Jbm.johogo.com
Fall 1998 Vol. 6, No. 1
| Diane H. Roberts |Evaluation of the Validity of Management Representations as Audit Evidence |
Jacob M. Chacko Internet-Based -Marketing
Randy Larson
| Robert W. Stone |Computer Self-Efficacy and Outcome Expectations and Their Impacts on|
|John W. Henry |Behavioral Intentions to use Computers in Non-Volitional Settings |
| Owen P. Hall, Jr. |Utilization of Artificail Intelligence Technologies in Adult |
|Farzin Makjidi |Distance Learning |
| | |
| Lara Preiser-Houy |Providing Value to the Users of Information Systems: A Theory of |
|Carole E. Agres |the IS-User Relationship Development |
Published jointly by the Western Decision Sciences Institute and the School of Management, California State University, Dominguez Hills
JOURNAL OF BUSINESS AND MANAGEMENT
THE OFFICIAL PUBLICATION OF THE WESTERN DECISION SCIENCES INSTITUTE (WDSI)
The Decision Sciences Institute is a professional society dedicated to the development and application of quantitative and behavioral methods to administrative problems. Most functional areas of business are represented among the membership. Through its journals, national and regional meetings, and other activities, the Decision Sciences Institute serves as a vehicle to advance and disseminate the theory, application, pedagogy, and curriculum development of the decision sciences.
Western Regional Officers 1997-98
President, Richard L. Jenson, Utah State University
President-Elect, Karen L. Fowler, University of Northern Colorado
Vice President for Programs Marc Massoud, Claremont McKenna College
Associate Program Chair, James Taylor, Claremont McKenna College
Vice President for Programs-Elect, Paul Mallete , Colorado State University
Vice President for Member Services, Eldon Y. Li, California Polytechnic State University, San Luis Obispo
Secretary/Treasurer, Christine A. McClatchey, University of Northern Colorado
JOURNAL OF BUSINESS
AND MANAGEMENT
Vol. 6, No. 1 Fall 1998
|EDITORS |Franklin Strier |
| |Burhan F. Yavas |
|EDITORIAL ASSISTANT |Eileen D. Hall |
Editorial Offices:
JOURNAL OF BUSINESS AND MANAGEMENT
School of Management
California State University, Dominguez Hills
1000 East Victoria Street
Carson, California 90747
Phone: (310) 243-3472, (310) 243-3501
Fax: (310) 516-3664, (310) 217-6964
Published jointly by Western Decision Sciences Institute (WDSI) and the School of Management, California State University, Dominguez Hills. The purpose of the JOURNAL OF BUSINESS AND MANAGEMENT is to provide a forum for the dissemination of contributions in all fields of business, management and related public policy of relevance to academics and practitioners. Original research, reports and opinion pieces are welcome. The style should emphasize exposition and clarity, and avoid technical detail and jargon.
The views expressed in articles published are those of the authors and not necessarily those of the Editors, Executive Board, Editorial Board, WDSI or California State University, Dominguez Hills. All submissions will be reviewed initially by the editors and, if judged appropriate, will be sent to knowledgeable referees for review. The authors assume responsibility for the accuracy of facts published in the articles.
Copyright ©1998 WDSI and by the School of Management, California State University, Dominguez Hills. Subscriptions are $16/year. Manuscripts should be double-spaced and submitted in triplicate. Manuscripts and comments should be directed to the editors.
JOURNAL OF BUSINESS AND MANAGEMENT
Executive Board
Richard L. Jensen, President, WDSI
Karen L. Fowler, President -Elect, WDSI
Donald L. Bates, Dean, School of Management, CSUDH
Franklin Strier, Editor
Burhan F. Yavas, Editor
Editorial Board
Dr. Joseph R. Biggs
California Polytechnic State University, San Luis Obispo
Dr. Henry Brehm
University of Maryland
Dr. Terry E. Dielman
Texas Christian University
Dr. Moshe Hagigi
Boston University
Dr. Ronald H. Heck
University of Hawaii at Manoa
Dr. Richard C. Hoffman
Salisbury State University, Maryland
Dr. Marc T. Jones
University of Otago, Dunedin, New Zealand
Dr. Erdener Kaynak
Pennsylvania State University
Dr. Thomas Kelly
State University of New York, Binghamton
Dr. George R. LaNoue
University of Maryland
Dr. George A. Marcoulides
California State University, Fullerton
Dr. John Preble
University of Delaware
Dr. Arie Reichel
Ben-Gurion University of the Negev, Israel
Dr. Elizabeth L. Rose
University of Auckland, New Zealand
Dr. Anne S. Tsui
The Hong Kong University of Science and Technology, Hong Kong
Dr. Michael Useem
University of Pennsylvania
Reviewer Acknowledgments
The editors of the Journal of Business and Management wish to express their appreciation to the following individuals who have reviewed manuscripts submitted for consideration in this issue of the Journal of Business and Management.
Dr. Zeynep Bilgin
Dr. Chee W. Chow
Dr. Cheryl Cruz
Dr. Stephen Ferraro
Dr. Charles Fojtik
Dr. James Heath
Dr. Paul Herbig
Dr. Swinder Janda
Dr. Albert King
Dr. Eric Kirby
Dr. Miriam Lacey
Dr. George Marcoulides
Dr. Mark Simkin
Dr. Elizabeth Trybus
JOURNAL OF BUSINESS
AND MANAGEMENT
TABLE OF CONTENTS
|From the Editor’s Desk |7 |
| | |
|Evaluation Of The Validity Of Management Representations As Audit Evidence |8 |
|Diane H.Roberts | |
|Internet-Based-Marketing |22 |
|Jacob M. Chacko | |
|Randy Larson | |
|Computer Self-Efficacy And Outcome Expectations And Their Impacts On Behavioral |45 |
|Intentions To Use Computers In Non-Volitional Settings | |
|Robert W. Stone | |
|John W. Henry | |
|Utilization of Artificial Intelligence Technologies In Adult Distance Learning |59 |
|Owen P. Hall, Jr. | |
|Farzin Madjidi | |
|Providing Value To The Users Of Information Systems: A Theory Of The IS-User |77 |
|Relationship Development | |
|Lara Preiser-Houy | |
|Carole E. Agres | |
| | |
FROM THE EDITOR’S DESK
Articles for this issue were selected from the best papers of the Western Decision Sciences Institutes’ 26th Annual Conference held in Reno, Nevada on April 7-11, 1998.
Auditors have greater responsibilities for detection of fraudulent and other deceptive management representations than ever before. Diane Roberts’ article presents her study of deception detection where management bonus is a significant risk assessment factor.
Marketing over the Internet is one of the fastest growing and most distinctive activities of modern business. Jacob Chacko and Randy Larson explore marketing applications over the Internet for corporations and individuals.
Do individuals who intend to use technology tend to be more innovative and creative than those who use technology because it is mandatory? Robert W. Stone and John W. Henry develop a structural equation model to investigate outcome expectancy and computer self-efficiency and their effects on behavioral intentions to use computers. Their results indicate outcome expectancies have significant influences on behavioral intentions to use computers.
Owen P. Hall, Jr. and Farzin Madjidi propose a preliminary model of an expert system based learning programs for working adult students. Such programs offer promise for meeting the business education challenges in the 21st. century.
The computerization of organizations has accelerated in the 1990’s. Information technology (IT) and information system specialists (IS) are two critical elements in this transformation. Lara Preiser-Houy and Carole E. Acres present a theory of how IS specialists provide value to their users through the development of IS-user relationships.
Frank Strier
Burhan F. Yavas
EVALUATION OF THE VALIDITY OF MANAGEMENT
REPRESENTATIONS AS AUDIT EVIDENCE
Diane H. Roberts*
Auditors' fraud risk assessment includes detection of deceptive client representations. Prior psychology deception detection research found low accuracy but high confidence that may lead to inappropriate judgments. In an audit context, subjects were more accurate than chance but had substantial errors. Accuracy was greater for overstatements than for understatements. The warning of low evaluated management integrity aided detection and significantly increased assessed fraud likelihood. Confidence was greater with audio tape cues than with transcript-only; however, accuracy was not significantly greater. The proliferation of Internet fraud makes knowledge about the ability to detect deception with only transcript cues important.
INTRODUCTION
S
AS No. 82 (AICPA, 1997) formalized auditors' responsibilities for discovering fraud during an audit to include detection of deception in client representations as a component of fraud risk assessment. Inquiry of client personnel has been found to be a cost effective means of discovering potential errors in the audit planning phase (Hylas and Ashton, 1982; Wright and Ashton, 1989); however, deception detection accuracy rates have been found to approximate chance in both everyday and work-related situations. Low deception detection accuracy is a particular concern in communication channels with limited deception cues such as transcripts-only conditions like the Internet, email, and audit workpapers. Poor calibration due to low accuracy and relatively high levels of confidence in deception detection abilities (DePaulo and Pfeifer, 1986) may lead to inappropriate judgements.
The types of lies generally told by clients tend to be those that make the client company appear to be more financially sound than it actually is. If the auditor perceives the client as more honest than the client actually is, then audit procedures may be reduced or misdirected and audit effectiveness compromised. If the auditor perceives the client to be more deceptive than the client really is, then audit testing may be needlessly extended and audit efficiency reduced. Inaccurate deception detection can also create risks in other business contexts.
This study uses an experimental approach to examine the accuracy of deception detection and judgment confidence in an audit context where one of the SAS No. 82 risk factors, a significant portion of management's compensation is from bonuses, is present. The impact of high or low management integrity and type of communication channel are examined. The communication channels are audio tape and transcripts-only as cues as the possibility of Internet fraud makes knowledge about the ability to detect deception with only transcript cues important. An overall assessment of the likelihood of the presence of fraud is elicited and related to accuracy of deception detection.
The remainder of the paper is organized as follows. The next section provides a review of the relevant literature from the auditing and psychology domains. The third section details the experimental design and the fourth section presents the results. The final section provides conclusions and recommendations for future research.
LITERATURE REVIEW AND HYPOTHESIS DEVELOPMENT
Use of Unaudited Data in Audit Planning
Evidence the auditor obtains during management inquiry has limitations: it has not been observed directly, and the evidence was derived from the entity itself, thus is not independent. Wild and Biggs (1990) derived a Bayesian decision theoretic model to determine the audit impact of utilizing unaudited book values in analytical procedures. The model indicated that reliance on unaudited book values with inaccurate assessments of no intolerable error unnecessarily exposes auditors to increased risk of incorrect acceptance and audit costs. Auditors' reliance upon unaudited management responses to auditor inquiry without accurate error assessment has a similar risk.
Unaudited financial statements can be considered management's representation of the company's financial status. Professional skepticism is used to evaluate any management representation and the possibility of errors or irregularities considered (AICPA, 1997). Determination of an account's expected value should be based upon sources independent of the audited entity, thus theoretically, unaudited book values should not be used to perform preliminary analytical procedures (AICPA, 1980).
Tversky and Kahneman (1974) found decision makers would use a provided value as an anchor for their judgments and then adjust in the correct direction, but insufficiently compared to a value reached without the anchor's presence. Kinney and Uecker (1982) anticipated that auditors would use this heuristic to determine their analytical review investigation boundaries and the anchor would be the current year's unaudited amount. Auditors who received two prior years' audited data and unaudited current year data in the lower (higher) value range established lower (higher) analytical review investigation boundaries. If the unaudited value contains a material error or irregularity, then the non-investigation range based on that value will incorrectly increase the probability of not investigating the account.
A richer data set of five years of prior audited results did not remove the impact of the unaudited book value anchor, although auditors were slightly less biased in the direction of the anchor (Biggs and Wild, 1985). Unaudited data containing any type of trend influenced auditors (Heintz and White, 1989). A decreasing trend had a greater impact than an increasing trend, perhaps due to conservatism. A trend reversal had a greater impact than did a consistent trend.
Auditors with no or incomplete explanations from management had comparable judgments of the most likely cause of a fluctuation observed in analytical procedures (Bedard and Biggs, 1991). Most auditors either disconfirmed or did not adopt management's incomplete explanation and experience measures by rank did not have a significant effect. This suggests that when management fails to supply the correct explanation, intentionally or not, audit efficiency and effectiveness could be reduced.
Deception Detection
Deception detection is a general domain task practiced throughout life. When outcome feedback is unsystematic and the task's structure is poorly understood, learning from experience is unlikely (Einhorn, 1982). The outcome feedback from deception detection tasks can be accurate or can be confounded by either a disbelieving-the-truth error or a believing-a-lie error. Suspected liars cannot be trusted to provide completely honest outcome feedback, thus reliable information for improving deception detection abilities is unavailable. This may partially explain low levels of accuracy and difficulties in development of appropriate confidence.
Since people have generally been found to be inaccurate at detecting deception, psychology researchers have attempted to determine what, if any, cues lead to success in catching liars. Ekman's (1985) theory of communication channels stated that less controllable channels have greater leakage of cues to deception detection and properly attending to these channels may enable greater detection of deception. Facial expressions are readily controllable but other parts of the body and vocal intonations are more difficult to control. The highly controllable verbal content cues of plausibility, consistency, and concreteness (detailed instead of vague) were found to have Pearson correlations between actual truth and subjects' judgments of .42, .24, and .20, respectively (Kraut, 1978). Planned lies were found to be no more detectable than unplanned lies; however, planned responses, whether deceptive or true, were perceived as being more deceptive (DePaulo et al., 1983).
Visual cues, especially facial visual cues, decrease deception detection accuracy (DePaulo et al., 1983; Riggio and Friedman, 1983; Ekman, 1985). Vocal cues were found to be helpful to detect deceit (Bond et al., 1985), while relying only on transcripts was detrimental to detection accuracy (DePaulo et al., 1983). Body viewing conditions, both with and without sound, had significantly better than chance detection accuracy (Riggio and Friedman, 1983). Training that included valid feedback information was more effective than practice without feedback in improving detection accuracy, but improvement did not generalize to evaluating other people's response (Zuckerman et al., 1984).
Studies whose subjects' profession involved deception detection found that job experience did not significantly improve accuracy (Kraut and Poe, 1980; DePaulo and Pfeifer, 1986; Vrij, 1993). U. S. Secret Service agents were the exception with a 64 percent detection rate (Ekman and O'Sullivan, 1991).
Professional customs inspectors may develop detection strategies on the job (Kraut and Poe, 1980); however, the absence of verifiable outcome feedback did not allow any conclusions about the strategies' normativity. Although some experimental settings approximated the work environment (Kraut and Poe, 1980; Vrij, 1993), the accuracy in these studies was not better than in those studies where the experimental task did not attempt to capture job dynamics (DePaulo and Pfeifer, 1986; Ekman and O'Sullivan, 1991).
Law enforcement officers with an average of seven years of experience were not more accurate than students but were significantly more confident in their judgments (DePaulo and Pfeifer, 1986). Confidence and accuracy have not been found to have a consistent relationship over a broad spectrum of contexts and tasks (Koehler, 1991). Peterson and Pitz (1988) hypothesized that this may be due to the absence of a universal relationship between accuracy and confidence. Other studies (Einhorn, 1980; Hogarth, et al., 1991) focused on the difficulties of learning from feedback as a possible reason for inappropriate confidence.
This prior research and discussion suggest the following hypotheses:
H1: Accounting students will be more accurate than chance (50%) in detecting deception in an audit context.
H2: Accounting students who have access to verbal cues will be more accurate in detecting deception in an audit context than those who access to transcripts-only.
H3: Accounting students' confidence in their judgment will be greater than the accuracy level achieved.
EXPERIMENTAL DESIGN
Subjects
The experiment was completed by 153 advanced accounting students in their final semester of undergraduate study. Deception detection is a general domain task and the students' degree course work provides the background to understand the audit/business context of the task.
Task and Procedures
Background information was provided about a hypothetical, continuing audit client with unqualified opinions in prior audits. The moderate internal control and the materiality threshold were unchanged from the prior year and the base rate (50/50) of honest and deceptive client responses was disclosed. Client responses to auditor inquiries were provided for each of two asset and one liability accounts. For each client response, the students judged its truthfulness or deceptiveness and rated their judgment confidence. Then, based upon their belief in the veracity of the information received from all the client responses for an account, the students indicated the percentage likelihood of fraud at the client. Lastly, an overall fraud likelihood judgment was made.
Independent Variables
The experiment is a 2 X 2 factorial design. Management integrity and communication channel are between-subjects factors. The red flags approach was used as a theoretical basis for operationalization of the high and low management integrity levels. Red flags are significant items that should alert the auditor to a higher than normal possibility of management fraud, thus they also serve as indicators of a low level of management integrity. The five most significant indicators from Albrecht and Romney's (1986) study were combined to form the low level, and their opposite formed the high level. These indicators included overlooking controls, and domination of the company by individuals who lived beyond their means and who were characterized as 'wheeler-dealers.' A manipulation check was included in the debriefing questions.
The two communication channels were (1) a prerecorded audio tape condition and (2) a transcript-only condition. The difference was limited to the additional verbal cues on the audio tape. All auditor questions of the client and other information were presented in a written format for all conditions.
Dependent Measures
Students indicated honesty or deception judgments on a seven-point Likert scale with end points of “very honest” and “very deceptive.” To obtain the accuracy score the student's response was compared to the known truth or deception status of the client's representation and the percentage of correct judgments was computed. The truth or deception status was determined by comparing the client's response to the facts of the experimental case.
Confidence judgments were indicated on a seven-point Likert scale with the end points of “very confident” and “not at all confident.” Each scale point was assigned a numerical value and the responses were summed to yield an overall confidence score for each student. Fraud likelihood judgments were the subjective probability judgments of the students stated as percentages.
RESULTS
To verify the manipulation of the management integrity variable, students were asked to indicate their evaluation of the management's integrity level. The manipulation check levels were assigned values as follows: low, 1; neutral, 2; and high, 3. Students in the low management integrity condition assigned a mean of 1.5775 to the case's management and those in the high management integrity condition assigned a mean of 1.9459. These were significantly different integrity levels per the results of an univariate ANOVA (F = 7.2449, p = .008) with the intended higher integrity level perceived as higher by the students. Students primarily identified the low integrity level as low; however, in the high condition 21 students rated it as high, 28 as neutral, and 25 as low. It may have been difficult to assign a high integrity rating when the provided deception base rate was 50 percent. Due to this evaluation, analysis was performed for both the experimental management integrity levels (high and low) and the evaluated management integrity levels (high, low, and neutral). Experimental management integrity results are presented unless the evaluated management integrity level analysis was different.
The first hypothesis suggested accounting students would be more accurate than chance accuracy of 50 percent in detecting deception in an audit context. This hypothesis was supported as overall students were more accurate than chance (t = 6.51, p = 0.000). Mean accuracy levels by account and experimental group are shown in Table 1. On an individual account basis, students were more accurate than chance for each of the asset accounts; however, the accuracy for the liability of accounts payable was not significantly different than chance (t = 0.89, p = 0.374). As lower liabilities indicate financial soundness, the deceptive responses for the accounts payable attempted to prevent recording an additional amount and represented understatement errors. The deceptive responses for the asset accounts attempted to increase their values and focused on overstatement. Students had difficulty detecting understatement lies but were more successful with overstatement lies.
The highest accuracy level was attained on the first account in the experiment, inventory, and was statistically greater than chance (t = 7.45, p = 0.000). As this was the initial account evaluated the increased accuracy level can not be attributed to a learning effect. Further evidence of the lack of a learning effect can be seen from the mean accuracy rates achieved for different client personnel shown in Table 2. The second client responding was always the same individual to simulate the controller or primary financial officer who is the main contact for the audit. The first client responding was specific to that account to simulate the individual accountant responsible for that portion of the financial statements. The highest accuracy levels for the second client, repeated respondent are for the first account, inventory, and decline as more responses are provided. Subject fatigue is possible but unlikely as the overall accuracy for the last account is higher than that of the second account. Although accuracy on the accounts payable responses did not exceed chance, accuracy for the notes receivable was significantly greater than chance (t = 2.43, p = 0.016).
Table 1
Mean Accuracy of Deception Detection by Account
|Inventory |Transcript |Audio Tape |Total Integrity Level |
| High Management Integrity |.5344* |.6244* |.5772* |
| Low Management Integrity |.6453* |.6179* |.6314* |
| Total Communication Channel |.5891* |.6201* |.6045* |
|Accounts Payable | | | |
| High Management Integrity |.4954 |.4930 |.4942 |
| Low Management Integrity |.5463* |.5023 |.5240 |
| Total Communication Channel |.5205 |.4978 |.5092 |
|Notes Receivable | | | |
| High Management Integrity |.4764 |.5659* |.5200 |
| Low Management Integrity |.5829* |.4895 |.5356* |
| Total Communication Channel |.5290* |.5267 |.5278* |
|All Accounts | | | |
| High Management Integrity |.5044 |.5697* |.5362* |
| Low Management Integrity |.5937* |.5403* |.5666* |
| Total Communication Channel |.5484* |.5546* |.5515* |
* Significantly different than chance accuracy of 50 percent.
Greater accuracy for the highest materiality account, inventory, is a positive indicator of audit effectiveness; however, this enhanced accuracy may not be attributable to increased effort on the part of the subjects in response to the significance of the account. The individuals who provided the responses to be evaluated may have been differentially detectable. All client replies to auditor questions were from male respondents for inventory but the first client respondents for the other accounts were female. There is no clear gender effect although the highest level of account accuracy was for the all male inventory account. However, for the accounts payable and notes receivable accounts, there was greater accuracy for the female first client than for the male second client.
Table 2
Mean Accuracy of Deception Detection for Account Specific Client
|First Client |Inventory |Accounts Payable |Notes Receivable |
| |(Male) |(Female) |(Female) |
|Transcript, High Management Integrity |.5813 |.5390 |.5385 |
|Transcript, Low Management Integrity |.6661* |.5795 |.6400* |
|Audio Tape, High Management Integrity |.6668* |.5403 |.7619* |
|Audio Tape, Low Management Integrity |.6833* |.5426 |.5600 |
|Total Accuracy |.6490 |.5416 |.6232 |
Mean Accuracy of Deception Detection for Repeated Client
|Second Client |Inventory |Accounts Payable |Notes Receivable |
| |(Male) |(Female) |(Female) |
|Transcript, High Management Integrity |.5090 |.4531 |.4141 |
|Transcript, Low Management Integrity |.6237* |.5129 |.5353 |
|Audio Tape, High Management Integrity |.5816 |.4822 |.4900 |
|Audio Tape, Low Management Integrity |.5513 |.4609 |.4205 |
|Total Accuracy |.5658 |.4770 |.46542 |
* Significantly different than chance accuracy of 50 percent.
The second hypothesis proposed that accounting students with access to verbal cues will be more accurate in detecting deception in an audit context than those with access to transcripts only. Hypothesis two was not supported as students with access to the audio tape's additional cues were more accurate, but not significantly so per ANOVA (F = 0.158, p = 0.693). Experimental management integrity level did significantly effect accuracy (F = 3.972, p = 0.048) as the low management integrity condition had greater total accuracy in all accounts. The interaction of channel and experimental integrity level was also significant (F = 15.648, p = 0.000). The low management integrity, transcript group had the greatest accuracy for each account and overall. The high management integrity, transcript group was the least accurate on an overall basis and for inventory and for notes receivable.
An additional ANOVA was performed using the evaluated integrity level. The evaluated integrity level was significant (F = 3.413, p = 0.036); however, the interaction of channel and evaluated integrity level was not significant. The low evaluated integrity level had the greatest accuracy on an overall basis and for inventory and accounts payable. The high evaluated integrity level had the greatest accuracy for the notes receivable account.
There was a greater range of achieved accuracy rates in the transcript condition than in the audio tape condition. The warning provided by the experimental low management integrity condition information apparently aided detection as students in the low management integrity, transcript condition were the most accurate for all individual accounts and overall with total accuracy of .5937. One of the factors that contributed to their performance is that this group had the best detection rate for the second client, repeated respondent. The lowest accuracy of any group was .5044 for the high management integrity, transcript condition. This group had the lowest accuracy rate for both the second client, repeated respondent and the inventory account where all other groups were their most accurate. For the audio tape condition, the high management integrity group was more accurate at .5697, while the low management integrity group had accuracy of .5403.
Table 3
Evaluations of Client Responses Compared to Base Rate of 50/50 in Numbers and Percentages of Students
| |Inventory |Accounts Payable |Notes Receivable |All Accounts |
| |Number Percent |Number Percent |Number Percent |Number Percent |
|Deceptive |55 36% |35 23% |81 53% |54 35% |
|Base Rate |25 16% |17 11% |12 8% |3 2% |
|Honest |73 48% |101 66% |60 39% |96 63% |
|Total |153 100% |153 100% |153 100% |153 100% |
As accuracy rates are near or slightly above chance, the types of judgments subjects are making of the client replies is of interest and are shown in Table 3. The base rate of deception was provided as 50 percent so students who judged half of the responses to be deceptive had base rate judgments. Those students who judged more than half of the responses as deceptive (honest) considered the client to be more deceptive (honest) than the base rate. For the inventory account more students judged the client's responses to be honest (48%) than to be deceptive (36%). Accuracy was greatest for the inventory account, and it had the highest percentage of base rate judgments (16%). The account with the lowest accuracy was accounts payable which maybe due in part to 66 percent of the students judging the client's replies to be honest. Notes receivable was the only account where more people (53%) judged the responses to be more deceptive than honest (39%). On an overall basis, the client personnel's replies were judged to be more honest (63%) than deceptive (35%) with only two percent achieving the base rate.
Table 4
Mean Confidence by Evaluated Integrity and by Account*
|Inventory |Transcript |Audio Tape |Total Integrity Level |
| High Evaluated Integrity Level |4.2662 |4.4508 |4.3245 |
| Low Evaluated Integrity Level |3.9174 |4.5497 |4.2511 |
| Neutral Evaluated Integrity Level |3.1846 |4.2659 |3.8643 |
| Total Communication Channel |3.9222 |4.4416 |4.1770 |
|Accounts Payable | | | |
| High Evaluated Integrity Level |4.3885 |4.4792 |4.4171 |
| Low Evaluated Integrity Level |3.8294 |4.6471 |4.2609 |
| Neutral Evaluated Integrity Level |3.2062 |4.1859 |3.8220 |
| Total Communication Channel |3.9414 |4.4684 |4.1959 |
|Notes Receivable | | | |
| High Evaluated Integrity Level |4.2704 |4.7967 |4.4366 |
| Low Evaluated Integrity Level |3.7703 |4.6600 |4.2399 |
| Neutral Evaluated Integrity Level |3.1800 |4.3091 |3.8897 |
| Total Communication Channel |3.8579 |4.5658 |4.2069 |
|All Accounts | | | |
| High Evaluated Integrity Level |4.3065 |4.5742 |4.3911 |
| Low Evaluated Integrity Level |3.8382 |4.6171 |4.2493 |
| Neutral Evaluated Integrity Level |3.1685 |4.2568 |3.8526 |
| Total Communication Channel |4.0800 |4.4569 |4.1907 |
*Values Assigned to Likert Scale Points: Very Confident, 6; Moderately Confident, 5;
Weakly Confident, 4; Neutral, 3; Weakly Not Confident, 1; and Not At All Confident, 0.
The third hypothesis proposed that accounting students' confidence in their judgement will be greater than the accuracy level achieved. Hypothesis three was supported as the ANOVA results reveal that communication channel had a significant effect on confidence levels (F = 14.177, p = 0.000); however, communication channel had no significant effect on accuracy. Students with access to the audio tape's additional vocal cues were significantly more confident than those with transcripts. The enhanced confidence from the extra cues in the audio tape communication channel was not accompanied by an enhanced accuracy effect so it may be inappropriate or excessive confidence. Correlation of total accuracy and total confidence was .2227 and was significantly different from zero (p = .006). This positive correlation indicates that as confidence increases, accuracy increases weakly.
The experimental management integrity level did not significantly impact confidence (F = 0.794, p = 0.374) but the students' evaluated management integrity level did significantly effect confidence levels (F = 5.307, p = 0.006). Confidence levels ranged from above neutral (3) to below moderately confident (5) as shown in Table 4. The lowest confidence level was for the neutral evaluated integrity group for all accounts. If the neutral evaluation represents an inability to evaluate the management integrity as either high or low instead of an actual assessment, the comparative lack of confidence is appropriate. For the transcript condition, the greatest confidence was for the high evaluated integrity group. Dealing with a more reputable company gave more confidence in assessing responses provided in an impoverished deception cue set. In the audio tape condition, the most confidence was for the low evaluated integrity group on an overall basis and for each account except notes receivable.
A fraud likelihood judgment was elicited after all client representations. ANOVA analysis did not show any significant effect of experimental management integrity level or communication channel. The evaluated integrity level did have a significant effect on the likelihood of fraud per ANOVA results (F = 3.589, p = 0.032). Students who judged the client's management integrity level to be lower had higher fraud likelihood judgments as shown in Table 5. There was a significant interaction between evaluated management integrity and communication channel (F = 3.125, p = 0.047). Audio tape cues led to more extreme fraud assessments with the highest (lowest) fraud rating on an overall basis in the low (neutral) integrity, audio tape condition. Correlation of the probability of fraud and total accuracy level was a very low .0971 and was not significant (p = 0.257).
Table 5
Mean Fraud Likelihood Judgments by Evaluated Management Integrity Level
| |Transcript |Audio Tape |Total Integrity Level |
|High Evaluated Management Integrity |.5364 |.5000 |.5250 |
|Low Evaluated Management Integrity |.6103 |.6789 |.6493 |
|Neutral Evaluated Management Integrity |.6417 |.4550 |.5250 |
|Total Communication Channel |.5881 |.5845 |.5885 |
On an overall basis, higher probability of fraud, .6493, was judged in the evaluated low integrity condition compared to the .5250 assessed by those who evaluated the integrity as high or neutral. These fairly high judgments of fraud may be partially attributed to the task. Students were required to focus directly on deceptiveness and were informed of the 50 percent base rate of deceptiveness in the experiment. Also, students and auditors with lesser amounts of experience have been shown to have more conservative judgments in the evaluation of management integrity compared to more experienced auditors (Roberts, 1997).
CONCLUSIONS
Accuracy levels achieved were greater than chance accuracy of 50 percent; however, a substantial error percentage remains (between .4063 and .4956 depending on experimental condition). This is consistent with accuracy levels obtained in other deception detection studies using non-business task contexts. The absence of outcome feedback in the experiment is similar to real life deception detection. Any real world outcome feedback has significant lack of reliability concerns, thus people can only act upon their subjective belief in the veracity of the information they receive. Auditors should combine deception detection in client representations with other techniques when implementing SAS No. 82's (AICPA, 1997) fraud risk assessment. Considerable caution should be exercised before relying on or acting upon unsupported management representations.
When the client's responses were judged as more deceptive than the base rate, there was excessive disbelief of client representations. In an audit context this may lead to an unnecessary increase in audit work, thus reducing audit efficiency. When client responses were judged more honest than the base rate, the client was able to successfully misdirect the subjects. If successful misdirection from a sensitive client issue is achieved, then audit work maybe reduced. Audit effectiveness is impaired and the risk of an audit failure may be increased.
The proliferation of Internet commerce and the possibility for fraud makes knowledge about the ability to detect deception with only transcript cues important. Although there was no significant communication channel effect on accuracy, both the experimental and the evaluated management integrity levels did have an effect. As the experimental low management integrity warning did aid detection of deception in the transcript condition, it highlights the importance of knowing the background or business reputation of the entity one is dealing with. Under current electronic commerce conditions it is problematic how a user/consumer obtains information about management integrity level. This indicates a significant practice development opportunity for accountants, one which the AICPA has started to tap with its Web Assure product.
Students were more successful in the detection of overstatement lies for the asset accounts than with understatement lies for accounts payable. DeFond and Jiambalvo (1991) found more overstatement of net income errors reported as prior period adjustments. Understatement of liabilities is often accompanied by understatement of expenses, resulting in overstatement of net income. Prevention of liability understatement is thus an important audit issue.
The degree of confidence a person has in their judgment is a significant component of how willing they are to act upon their belief. There was greater confidence than accuracy especially for the audio tape conditions where the additional cues provided as added level of comfort. The transcript condition had lower confidence than the audio tape condition. This was a more realistic assessment and perhaps may indicate a more cautious approach to judgment. The transcript condition's highest confidence was for the high evaluated integrity group which again indicates the importance of knowing about the company one is dealing with.
The greatest (least) accuracy was in the most (least) material account which raises the issue of whether materiality affects accuracy. Future research could directly address this issue by having the same individual(s) provide the responses to be evaluated for all materiality levels. This would remove the effect of the differential delectability of the people making the responses. Deception detection could also be studied in other business or accounting contexts, with income tax being a particularly interesting area.
REFERENCES
Albrecht, W.S. & Romney, M.B. (1986). “Red-Flagging Management Fraud:
A Validation.” Advances in Accounting, 3, 323-333.
American Institute of Certified Public Accountants. (1997). Statement on Auditing
Standards No. 82, Consideration of Fraud in a Financial Statement Audit. New
York, NY: AICPA.
_____, (1980). Statement on Auditing Standards No. 31, Evidential Matter. New York,
NY: AICPA.
Bedard, J.C. & Biggs, S. F. (1991). “The Effect of Domain-Specific Experience on
Evaluation of Management Representation in Analytical Procedures.” Auditing: A
Journal of Practice & Theory, 10 (Fall), 77-90.
Biggs, S. F. & Wild, J. J. (1985)., “An Investigation of Auditor Judgment in Analytical
Review.” The Accounting Review, LX (October), 607-633.
Bond, C. F., Kahler, K. N. .& Paolicelli, L. M. (1985). “The Miscommunication of
Deception: An Adaptive Perspective.” Journal of Experimental Social Psychology, 21, 331-345.
DePaulo, B., Lanier, M. K. & Davis, T. (1983). “Detecting the Deceit of the Motivated
Liar.” Journal of Personality and Social Psychology, 45, 1096-1103.
_____ & Pfeifer, R. L. (1986). “On-the-Job Experience and Skill at Detecting
Deception.” Journal of Applied Social Psychology, 16, 249-267.
DeFond, M. L. & Jiambalvo, J. (1991). “Incidence and Circumstances of Accounting
Errors.” The Accounting Review, 66 (July), 643-655.
Ekman, P. (1985). Telling Lies. New York: W. W. Norton & Company.
_____, & O'Sullivan, M. (1991). “Who Can Catch a Liar?” American Psychologist,
46(September), 913-920.
Einhorn, H. J. (1980). “Overconfidence in Judgment.” New Directions for Methodology
of Social and Behavioral Science, Vol. 4, Pp. 1-16. San Francisco: Jossey-Bass, Inc.
_____. (1982). “Learning from experience and suboptimal rules in decision making.” In
D. Kahneman, P. Slovic, and A. Tversky (Eds.) Judgment under uncertainty:
Heuristics and biases. Pp. 268-283. Cambridge: Cambridge University Press.
Heintz, J. A. & White, G. B. (1989). “Auditor Judgment in Analytical Review--Some
Further Evidence.” Auditing: A Journal of Practice & Theory, 8 (Spring), 22-39.
Hogarth, R. E., McKenzie, C. R. M., Gibbs, B. J., & Marquis, M. A. (1991). “Learning
From Feedback: Exactingness and Incentives.” Journal of Experimental
Psychology: Learning, Memory, and Cognition, 17 (4), 734-752.
Hylas, R. E. & Ashton, R. H. (1982). “Audit Detection of Financial Statement Errors.”
The Accounting Review, LVII (October), 751-765.
Kinney, Jr., W. R. & Uecker, W. C. (1982). “Mitigating the Consequences of Anchoring
in Auditor Judgments.” The Accounting Review, LVII (January), 55-69.
Koehler, D. J. (1991). “Explanation, Imagination, and Confidence in Judgment.”
Psychological Bulletin, 110 (3), 499-519.
Kraut, R. E. (1978). “Verbal and Nonverbal Cues in the Perception of Lying.” Journal
of Personality and Social Psychology, 36(4), 380-391.
_____, & Poe, D. (1980). “Behavioral Roots of Person Perception: The Deception
Judgments of Customs Inspectors and Laymen.” Journal of Personality and Social
Psychology, 39 (5), 784-798.
Peterson, D. K. & Pitz, G. F. (1988). “Confidence, Uncertainty, and the Use of
Information.” Journal of Experimental Psychology: Learning, Memory, and
Cognition, 14 (1), 85-92.
Riggio, R. E. & Friedman, H. S. (1983). “Individual Differences and Cues to Deception.”
Journal of Personality and Social Psychology, 45 (4), 899-915.
Roberts, D. H. (1997). “Audit Experience: Aid or Detriment to Evaluation of
Management Integrity?” Proceedings of the Institute of Business Administration and
Technology 1997 Conference Pp. 20-28, London, UK (July 7-14).
Tversky, A. & Kahneman, D. (1974). “Judgment Under Uncertainty: Heuristics and
Biases.” Science, (September 27), 1124-1131.
Vrij, A. (1993). “Credibility Judgments of Detectives: The Impact of Nonverbal
Behavior, Social Skills, and Physical Characteristics on Impression Formation.” The
Journal of Social Psychology, 133 (5), 601-610.
Wild, J. J. & Biggs, S. F. (1990). “Strategic Considerations for Unaudited Account
Values in Analytical Review.” The Accounting Review, 65 (January), 227-241.
Wright, A. & Ashton, R. H. (1989). “Identifying Audit Adjustments with Attention-
Directing Procedures.” The Accounting Review, LXIV (October), 710-728.
Zuckerman, M., Koestner, R. & Alton A. O. (1984). “Learning to Detect Deception.”
Journal of Personality and Social Psychology, 46 (3), 519-528.
INTERNET-BASED MARKETING
Jacob M. Chacko*
Randy Larson**
The popularity of the Internet makes it a powerful marketing tool for companies as well as individuals. Marketing applications on the Internet challenges the effectiveness of traditional modes used for such purposes. Businesses are vigorously exploring the possibilities of life online and have already experienced immense cost savings and successes. Individuals have found that the Internet can be an affordable and highly effective vehicle for personal marketing. This paper looks at marketing functions and explores their application over the Internet for corporate and personal marketing.
INTRODUCTION
T
he concept of managing a marketing program using the Internet can be referred to as “Virtual Marketing” (VM). This paper explores use of the Internet for marketing activities effectively. Both business and personal marketing issues will be discussed. In examining the potential use of VM in a marketing program, an Internet site is referred to for each stage to illustrate at least one aspect of its usefulness in business and personal marketing. The personal marketing illustrations will focus on job hunting, where job seekers are considered “products,” and employers are considered “consumers.” Marketing can be defined as a social process involving the activities necessary to enable individuals and organizations to obtain what they need and want through exchanges with others and to develop ongoing exchange relationships (Boyd, Walker, and Larreche 1995). In the past, marketing had been sales-driven. Organizations focused their energies on changing customers' minds to fit the product. As technology developed and competition increased, some companies shifted their approach and became customer driven. These companies expressed a new willingness to change their product to fit customers' requests. In the 1990s, successful companies are becoming market driven, adapting their products to fit their customers' preferences. It is marketing that is oriented toward creating rather than controlling a market (McKenna 1991).
The old approach to marketing and product development involved getting an idea, conducting traditional market research, developing a product, testing the market, and finally going to market. This typically is a slow, unresponsive, and turf-ridden method. Given the fast-changing marketplace, there is less and less reason to believe that this traditional approach can keep up with real customer wishes and demands or with the rigors of competition (McKenna 1991). The new approach is called Relationship Marketing.
Relationship marketing focuses on developing a continuous relationship with consumers across a family of related products and services. It involves the use of computer database technology to assist in filtering through masses of electronic consumer information in order to divide the entire potential market into smaller pieces, or segments. This helps an organization concentrate its marketing efforts on specific groups of consumers who may be the most interested in the product. These targeted groups of consumers will likely be the most satisfied after the sale is made since an effort was made before the sale to identify their needs and wants. The Internet represents a powerful method of reaching target customers.
The recent worldwide explosion in popularity of the Internet gives it the potential to be a perfect marketing tool. In fact, for companies of all sizes, the Net might soon replace or greatly compliment traditional business elements such as department store buildings, warehouses, paper catalogs and flyers, and TV and radio advertisements. It is entirely possible that in the not-so-distant future companies may no longer have a need for corporate offices or many of the traditional promotion and distribution vehicles. Potentially, all of these elements can be replaced by Internet applications.
Many businesses are vigorously exploring the possibilities of life online and have already experienced immense cost savings and successes. These savvy Internet optimists, or Netrepreneurs as they are becoming known, are moving quickly to build new forms of business that take advantage of the benefits offered by the Net.
Building an Internet program is an investment in any company’s future. Skeptics wallow in short-term thinking, asking themselves “how can the Internet enhance my existing operation today?” They are ignoring the technological revolution that is happening all around them and they do not realize that the traditional methods of conducting business have various inefficiencies. As we will see shortly, Internet tools can be useful in all aspects of marketing.
Any good business plan includes a marketing section that addresses the specific marketing needs of the company. According to Dollinger (1995), the basic components are marketing research, marketing strategy, marketing mix, and sales forecasting. The following sections will explore how the Internet can aid in performing the various marketing activities.
VIRTUAL MARKETING
Marketing Research
Marketing research is systematic and objective process of gathering, coding, and analyzing data for aid in making marketing decisions. This includes identifying the customers, their demographic characteristics, and their locations. It also helps in identifying existing and potential competitors (Dollinger 1995).
The Internet is a technological innovation that can ideally be thought of as a gigantic database of almost infinite proportions. Various search tools, or “browsers,” are available which can assist marketers in finding almost any kind of information they desire. Perhaps most of the Internet's potential for marketing researchers is to gather data for coding or analysis.
A source of secondary statistical data found on the Internet is the U.S. Statistical Abstract website, which contains census data published by the U.S. Government. Following is a sample of the data that can be retrieved from this site:
[pic]
One way this information can be utilized is by plugging it into forecasting software tools to help project future consumer populations in target markets. See the Sales Forecasting section for more details.
Another aspect of marketing research involves identifying customer demographics. To serve this need, the company InfoSeek has developed a browser with which it gathers detailed demographic data and builds lists of consumers with specific traits (Rebello, Armstrong, and Cortese 1996). Tools such as this can prove quite useful to marketers as a method of collecting data for conducting research on potential customers, or prospects, and for segmenting these markets into smaller, more attainable pieces. Also, marketing research involves identifying customers and their locations. See the Advertising section to see how prospect information can be located on the Internet.
An aspect in personal marketing that involves marketing research is finding relevant information about companies, industries, careers and the like. There are hundreds of websites that specialize on career initiation. Few examples of such site are HOTJOBS () which has company profiles of over 2000 firms and CareerMosaic () with information on both domestic and international firms. In addition, there are web sites of public agencies that provide both industry and company information – The Better Business Bureau (members/search.html), FedWorld Information Network () , and Thomas J. Long Business Library (sunsite.berkeley.edu) to mention a few. There are also sites that specialize on providing comprehensive information on the topic of researching companies which take readers through several stages of searches and provides links to hundreds of sites like the ones mentioned above. One such web site is Dr. Randall Hansen’s Guide to Researching Companies (stetson.edu/~rhansen/researching_companies.html).
With increasing numbers of students opting to be entrepreneurs, there are several Internet sites that provide valuable information on setting up and running small businesses, including resources available from various public and private sources. Some of examples of such web sites are Entrepreneur’s Cyber Shop (), Entrepreneur America (entrepreneur-), Small Business Administration (), National Entrepreneur Alliance (nea-), and Virtual Entrepreneur ().
Marketing Mix
Marketing mix decisions must be made once the target market(s) are selected. The mix is commonly referred to as the Four P's-Price, Product, Place, and Promotion.
1. Price
Price decisions involve calculating the exchange value of the company's goods and services. To achieve a desired strategic competitive position for a product or service in its target market, the manager must take competitors’ costs and prices into account (Boyd, Walker, and Larreche 1995). One way to quickly research competitor prices is to access their home pages and browse through their online electronic catalogs.
As an illustration, Honda positions its Accord automobile to compete with the Ford Taurus, Mazda 626, and Toyota Camry. Suppose Honda is considering following a low-priced strategy to beat the competitors’ prices. By browsing the web sites of these companies, Honda can quickly and easily find out the current base price ranges of each competing model, and can use this information to set its own prices lower. The following base price listings were taken from the Ford, Mazda, and Toyota web-sites:
Ford Taurus ()
[pic]
Mazda 626 ()
[pic]
Toyota Camry ()
CE:
5-Speed Manual Overdrive 4-Cylinder: $16,868
4-Speed Electronically Controlled Automatic Overdrive (ECT) 4-Cylinder:
$17,668
5-Speed Manual Overdrive V6: $19,728
LE:
4-Speed Electronically Controlled Automatic Overdrive (ECT) 4-Cylinder:
$20,348
4-Speed Electronically Controlled Automatic Overdrive With Intelligence
(ECT-i) V6: $22,658
XLE:
4-Speed Electronically Controlled Automatic Overdrive (ECT) 4-Cylinder:
$22,298
4-Speed Electronically Controlled Automatic Overdrive With Intelligence
(ECT-i) V6: $24,508
1997-based MSRP as of 4/97 includes $420 destination charge. Destination
charge is $40 higher in AL, FL, GA, NC and SC, and $27 higher in AR, LA,
MS, OK and TX. Excludes taxes, license, title, optional or regionally required
equipment. Actual dealer price may vary.
Based on a quick review of the feature comparisons and price ranges listed above, it is possible for Honda to accurately set the Accord’s price range to competitively position it in the marketplace. The Taurus has a base price range of $18,545-22,160, the 626 a range of $15,695-$22,995, and the Camry a range of $16,868-$24,508. Honda can use this current and accurate information to help decide if it is feasible to follow a low-priced strategy.
Obviously, this online pricing method is not limited only to the automotive industry. Any business of any size or type can research prices on the Internet as long as the competitors have web sites.
From a personal marketing perspective, job seekers can use the Internet to prepare for salary negotiation - Infoseek Careers Channel (getting_it-done/find_a_job). To help estimate the salaries they will require (or the “prices” employers will need to pay to acquire their services) when relocating to a different city based on a cost of living index:
crime3/salcalc3/citypick.html[pic]
[pic]
As you can see above, someone moving from Grand Forks, North Dakota to Minneapolis, Minnesota should expect a small increase in the cost of living. Therefore, if the job seeker makes $50,000 in Grand Forks, then he or she should ask for a “price” of at least $3,000 more when offering his or her job skills to employers in that area.
2. Product
Product decisions determine the bundle of physical and psychological attributes that form or are associated with the core benefit to be delivered. This includes packaging, colors, sizes, options, warranties, and after-sale service. These factors are directly related to quality.
Internet is an ideal medium for marketing products that are targeted toward consumers who surf the web. This is especially true for product in highly dynamic and ever-changing markets such are personal computers and other consumer electronics items. Almost every company marketing such products use the Internet to provide up to date information to their customers. In addition, a product guarantee promises exceptional, uncompromising quality and customer satisfaction (Richardson 1995). Service is a related concept. Marketers who appreciate the importance of [service] focus on building customer relationships (McKenna 1991). Companies that ignore or minimize their guarantees and services are not relationship-oriented. Following Great Plains Software's introduction of guaranteed response times with its product support services, the company has experienced many payoffs, including average renewals of 85%, and revenue from product support at nearly 20% of total revenues (Great Plains Software 1992).
By providing after-sale information to the consumer through its web-site, Federal Express believes it actually enhances the value of its products to track shipments online (Judson 1996). For instance, one of the online services Federal Express offers is package tracking, which allows customers to determine the status of their shipments after sending them:
[pic]
From a personal marketing perspective, the job seeker is the product and also its marketing manager. The Internet offers several services to aid in the development and positioning of the product [job seeker] based on the target market preferences [career choice]. An example of this is the array of career planning services such as skills assessment, self evaluation, goal setting etc. available on web sites such as Career Builder (), and Infoseek Careers Channel (getting_it-done/find_a_job). These sites provide valuable information and strategies that help individuals to be competitive in the job search process.
3. Place-Distribution
Place or distribution decisions and activities relate to the location of the business and the choice and availability of distribution channels. The major objective of a distribution or location decision is to get the venture's goods or services to the target market. The Internet as a marketing medium represents a viable method of offering and moving goods and services from sellers to consumers. Using the Net, manufacturers may find it possible to eliminate many parts of the traditional distribution elements, including retailers and wholesalers. For instance, instead of driving to a retail store to pick up their products, consumers now have the luxury of having the products delivered to them. All they need to do is cybershop through an electronic catalog located on the Internet, select the products they like, fill out an electronic order form, and specify the payment and shipping details (e.g., line of credit, overnight delivery, etc.).
JCPenney, L.L. Bean, and Lands' End are examples of companies who offer merchandise electronically. Below is an illustration of the Lands' End home shopping page. When entering the home page, you are presented with a list of alternative categories, or “departments” to choose from:
[pic]
[pic]
If we enter the Luggage department (by clicking on “Luggage”), we are offered a sublisting by luggage type. Say we select “Travel,” and are interested in a sturdy three-piece set of luggage. We see this listed as an option, select it, and are presented with all relevant details including a picture of the set, its price, dimensions (not shown), and a short description:
[pic]
To order, we use the online order form. From here we can select options such as color, quantity, and monogram:
[pic]
All that's left to enter is personal information such as name and address, and choose payment type (e.g., C.O.D., or credit) and delivery options. Companies selling online may send their customers copies of the order forms, by electronic mail if possible.
From a personal perspective, the Internet can be helpful in moving (or distributing) job seekers (the products) to the employers (the consumers). One way is by offering job search services online. The Minneapolis Star Tribune uses this technique to improve distribution by helping job seekers and employers locate each other more efficiently:
[pic]
[pic]
Another service available on Internet that assists with distribution in personal marketing is help in deciding on a place to live or relocate to. One example is the Home Buyer's website, which has useful information for people planning to sell or buy houses and move to a different city. For instance, a family interested in moving from Grand Forks, North Dakota to Minneapolis, Minnesota might first want to compare the crime rates between the two cities:
early97/crime/crimelab.html
[pic]
[pic]
4. Promotional Mix
A promotional mix includes such activities as personal selling, advertising, sales promotions, and publicity. Promotional activities encompass the methods and techniques that businesses use to communicate with their customers and with other stakeholders. The purposes of promotion are to inform and persuade (Dollinger, 1995).
a. Personal Selling
Personal selling is oral presentation supplemented by other media (e.g., overhead slides and graphics) either in a formal setting or in informal conversation, for the purpose of making a sale to prospective buyers (Dollinger, 1995).
A relationship-oriented approach is particularly important in sales, because successful salespeople develop customer relationships, not just orders (McKenna, 1991). According to McKenna (1991), the future for selling will include voice input sales where customers use television, telephone, and computerized “salespeople” to do their in-home shopping.
A Midwest company recently used the Big Yellow online Yellow Pages research tool as an inexpensive segmentation technique to quickly and efficiently identify and contact the desired target market in its serviceable region. The electronic raw data (saved as a text file from the Internet to a PC hard drive) from Big Yellow was manipulated slightly and then transferred directly into a local office PC database system (ACT! by Symantec Corporation) for use as a contact management system.
The firm prioritizes prospective customers based on their responses to the mailings, and uses periodic reports to keep the staff up-to-date on changes in customer relationships. This information is entered into the system whenever contact (in any form - telephone, mailings, meetings, etc.) was made with prospects. In effect, this gives the company the memory of an elephant by storing a history of past contacts to be used as a basis for managing future interactions (i.e., by listing “next actions”). This helps the company sales force effectively and efficiently keep track of its customers and prospects. Most importantly, it helps them “remember” to keep their promises...a basic requirement for keeping customers satisfied and loyal!
From a personal perspective, hundreds of web sites such as The Monster board (), Infoseek Career Channel (getting_it-done/find_a_job), and Career Path () provide extensive information on developing interviewing skills, polishing one’s business etiquette and on all aspect of personal selling. In addition, job seekers are able to expand their personal network using the Internet. Many personal networking services exist on the Net today. The one below is offered by Career Magazine. This website can deliver updates on personal networking techniques, including online networking strategies, directly to a subscriber's email address:
newsarts/networking.html
[pic]
[pic]
(For further information, see newsarts/current/networking.html)
b. Advertising
Advertising can be defined as any paid form of nonpersonal presentation and promotion of ideas, goods, or services by an identified sponsor, usually made using television, radio, and newspaper (Dollinger, 1995). The Internet represents a new advertisement media.
Computers and the Internet not only can help gather and analyze market data and identify target market segments, they can also help Marketers reach these markets. The Internet, for instance, can be useful to marketers as a method of collecting business addresses for direct-mail campaigns. One tool that can be used for such a purpose is the online Yellow Pages developed by Bell Labs (called “Big Yellow”). Big Yellow is a potentially less expensive data-gathering technique as opposed to purchasing addresses from an independent researcher. Below is an example of the Big Yellow search screen:
[pic]
As mentioned earlier, a Midwest company recently used Big Yellow to pinpoint specific agricultural businesses in a tri-state area for its promotional mailings. The company was able to locate and categorize prospects by using state and business type as search criteria. Big Yellow retrieved business names, addresses, telephone numbers, and in some cases, contact persons. The data retrieved from Big Yellow was imported into the company's marketing database and used for managing contacts and for printing mailing labels. Following is a subset of this data:
[pic]
Marketing on the Internet does not necessarily have to follow the traditional one-way communication methods such as direct mail. For instance, the search tools mentioned earlier can help customers quickly and efficiently locate home page catalogs of information about the products and services that would best suit them, resulting in a target marketing “bullseye.” Also, customers increasingly have the option to order products and services using Internet order forms.
More traditional marketing methods on the Internet may involve such technology as “webcasting” (a way to push information across the net rather than have customers find it). Webcasting involves dispatching collections of ordinary web pages, news updates, and live sound and video customized for a particular audience or person (Hof, 1996).
The TV manufacturers, retailers, and broadcasters have recognized that the Internet has huge potential to affect their industry, and are racing to prepare for the changes. For instance, combination PC/TVs are soon coming to the U.S., European, Japanese, and Asian markets (Gross, 1996). Broadcasting giant NBC is combining forces with software giant Microsoft to create MSNBC, a form of cable TV for Internet users. Together, the two companies created the MSNBC home page, where it airs its Internet broadcasts. As illustrated below, in addition to its broadcasts, this page has a variety of other information to offer on a 24 hour basis:
[pic]
Both NBC and Microsoft hope that MSNBC has a significant number of net viewers to prove the viability of an advertising-driven Internet site (Rebello and Lesley 1996). Other marketing methods on the Internet are likely to include direct e-mail (electronic mail), and electronic telemarketing (the Internet is currently heavily based on telephone technology).
Web advertising revenues currently do not compare to the tens of billions of dollars spent on TV spots every year. However, Jupiter Communications predicts that web ad revenue will jump to $5 billion by the year 2000. Then, it figures, 50 million people will be connected to the Net, giving advertisers the chance to reach TV-size audiences. In fact, Forrester Research says that Net merchants will sell $6.6 million in goods by the turn of the millennium. Presently, advertisers can expect to pay $30-$100 per 1,000 “impressions,” counted as each time an ad is viewed (Rebello, Armstrong, and Cortese, 1996). It appears that Internet advertising represents an excellent source of potential revenue, at least in the near future.
From a personal perspective, advertising comes into play whenever job seekers attempt to promote themselves (the products) to employers (the consumers). One way to effectively package oneself is by writing a great resume. There are a plethora of career counselors on the Internet who offer resume writing services. One counselor that offers free online resume writing tips is the Boston College Career Center. Besides using online resume, another way job seekers can promote themselves on the Internet is by posting their resumes and other personal information on their own homepages.
bc.edu/bc_org/svp/carct/resume.html
[pic]
[pic]
c. Publicity
Publicity can be defined as nonpersonal stimulation of demand for product, service, or business unit by planting commercially significant news or by obtaining favorable presentation of it on a published or broadcast medium that is not paid for by the sponsor (Dollinger, 1995).
Perhaps one of the most popular web sites today is published by NASA to cover the Mars Pathfinder mission. Although NASA pays for its own website, many other organizations post Mars mission updates on their homepages, free of charge to NASA. The space agency uses this ingenious marketing tool to publicize up-to-date findings to a worldwide audience. This serves the multiple purposes of informing and educating the public, generating enthusiasm for space exploration, and creating lots of new tax-paying fans (which NASA hopes will result in more funding for its programs in the long-run). Following are excerpts from the Mars Pathfinder home page:
[pic]
[pic]
[pic]
The benefits of such a publicity approach are obvious--the Internet gives the space agency a means of dispersing controlled, varied, and easy-to-update information, especially when compared to the more traditional, one-time press-release method. Below is an illustration of online snapshots from the mission, including a panoramic view of the Martian landscape. Notice how NASA has capitalized on the tremendous attention its mission has received over the Internet, by giving the rocks in the picture nicknames of popular characters and interesting objects found in everyday life:
[pic]
From a personal perspective, publicity can be an important way for job seekers to become known to employers. One way that job seekers can publicize themselves and their abilities to prospective employers is by posting their resumes at free online resume database website such as Work Avenue (located at ). Another avenue for job seekers to publicize themselves is through informational interviews, where the job seekers initiates contact with prospective employers to set up meetings for sole purpose of gathering information on the employers and establishing contacts with key people in the firm. This is especially appropriate for seniors who are about a year away from graduation. Several of the web sites that have tips on interviewing have specific sections on informational interviewing.
d. Sales Promotions
Sales promotions are attention-getters designed to stimulate customer purchasing. Examples include price reductions through coupons and volume discounts, and other nonroutine selling efforts such as trade shows and exhibitions (Dollinger, 1995).
Price-oriented sales promotions (as practiced frequently in traditional marketing programs) often attract the “wrong” customers, those who are interested only in price, reducing both loyalty and profitability in the long run (Richardson, 1995). On the contrary, relationship marketing does not seek a temporarily increase in sales...the ultimate goal is increased usage over time (Richardson, 1995).
Online merchants are finding new ways to gain repeat customers. They're finding that contests, give-aways, and “sweb-stakes” keeps the clientele coming back. For example, , an online bookseller, has a quarterly drawing in which the winner gets a free book a week for a year. Smart Games, Inc. offers over $50,000 in cash prizes to customers who score well on its Smart Games Challenge CD-ROM game. Its web site, , attracts some 500 people a day who download a demo version of the game--helping to generate $1 million in CD sales (Rebello, Armstrong and Cortese, 1996).
Rather than using sales promotions, Virtual Marketers can use new database and communication technologies to deal with their customers as individuals in order to tailor messages that make the consumers more receptive to their claims. For example, instead of offering a temporary price reduction, a relationship-oriented software company might offer free copies of its software online for trial periods to new computer buyers, so that the potential customers can see for themselves the benefits the product has to offer. Microsoft is one of many companies offering free beta versions of its software products:
[pic]
From a personal perspective, less-experienced job seekers such as college students or recent graduates can promote themselves by interning for companies where they might not otherwise have the opportunity to work as full-salaried employees. Internships typically pay very little (which means employees are working at a discounted “price”) but the potential long-term benefits are astronomical in terms of gaining work-experience and business contacts. Some organizations can be very hard to get into without related industry experience. One example is the National Aeronautics and Space Agency (NASA). Following is online information about NASA's internship program:
[pic]
D. Sales Forecasting
Sales forecasting combines marketing research and marketing efforts using data-based and judgmental methods to help estimate future sales levels. Forecasts provide a basis for negotiating credit and bank financing, and give the business a set of performance standards that can be used to evaluate progress (Dollinger, 1995).
Time-series forecasting is a quantitative method of predicting the future of a certain variable. In business, this method is widely used to predict future sales based on historical sales. Time-series considers only historical behavior of the variable in its attempt to predict future values (Eppen, Gould, and Schmidt, 1993). While time-series forecast methods can be fairly accurate predictors based on historical performance, they do not take into account all possible factors that might influence sales over a long-period of time (e.g., consumer population growth, new legislation, and fluctuations in the economy). Forecasters attempting to make long-run (and even short-run) predictions should be aware of these factors, and should possibly utilize a causal forecasting method such as linear regression. This model assumes there is a predictive relationship between a dependent variable (e.g., sales) and independent variables (e.g., changes in disposable income).
For a thorough estimate of potential sales, marketers should plug research data into software forecasting tools to obtain objective sales estimates. Some companies offer such tools on the Internet. Alt-C Systems, Inc., one such company, offers the TimeTrends forecasting software:
[pic]
[pic]
Business marketers can use the data retrieved from the marketing research phase (e.g., from the U.S. Census website) with the software to objectively estimate future sales.
From a personal perspective, job seekers might search the Net for information about the short-term economic outlook:
[pic]
Job Seekers can quickly learn from the information above that the economy is currently doing well but is expected to slow down next year. This may mean that jobs will be harder to come by a year from now, so job seekers might want to begin looking immediately for new employment while businesses are still hiring (or “buying”).
CONCLUSION
With the emergence of the information age, world is relying extensively on the Internet for a variety of purposes. Organizations and individuals have the opportunity to present themselves to a captive audience. Internet by the basis of being interactive, current, and low cost, attracts customers of all economic, cultural, and educational backgrounds. Today there is no excuse for marketers to not to take the initiative in utilizing the web to their advantage. Educational institutions have the responsibility of educating both marketers and customers on the new uses of this powerful medium, and thus add value to the marketing process.
INTERNET RESOURCES
Big Yellow ()
National Space Science Data Center ()
U.S. Statistical Abstract Census Data ()
HOTJOBS ()
CareerMosaic ()
The Better Business Bureau (members/search.html)
FedWorld Information Network ()
Thomas J. Long Business Library (sunsite.berkeley.edu)
Dr. Randall Hansen’s Guide to Researching Companies
(stetson.edu/~rhansen/researching_companies.html).
Entrepreneur’s Cyber Shop ()
Entrepreneur America (entrepreneur-)
Small Business Administration ()
National Entrepreneur Alliance (nea-)
Virtual Entrepreneur ().
Ford ()
Mazda ()
Salary Calculator ()
Toyota ()
Infoseek Careers Channel (getting_it-done/find_a_job)
Federal Express ()
Resume Writing Services ()
Career Builder ()
Infoseek Careers Channel (getting_it-done/find_a_job)
Job Search ()
Lands End ()
The Monster board ()
Career Path ()
Personal Networks-()
Microsoft NBC ()
Mars Pathfinder Project ()
Amazon Online Booksellers ()
Microsoft Corporation ()
NASA Internship Program ()
Smart Games, Inc. ()
Macroeconomic Forecast ()
TimeTrends Forecasting Software ()
REFERENCES
Boyd Jr., H, Walker Jr., O and Larreche, J. C. (1995). Marketing Management - A Strategic Approach With AGlobal Orientation, 2nd ed. Madison: Irwin, 4, 305.
Dollinger, M (1995). Entrepreneurship-Strategies and Resources. Austen/Irwin, 227-250.
Eppen, G.D. Gould, F.J. and Schmidt, C.P. (1993). Introductory Management Science, 4th ed. New Jersey: Prentice-Hall, 785.
Futrell, C. M. (1996). Fundamentals of Selling - Customers For Life, 5th ed. Madison: Times Mirror Higher Education Group.
Great Plains Software (1992), “Taking Names.” Inc. Magazine, (September), 5.
Gross, N (1996). “Defending The Living Room: How TV makers intend to fend off
Cyberlopers.” Business Week, 24 (June) 96-98.
Hof, R. D. (1996). “Don't Surf To Us, We'll Surf To You.” Business Week, 9 (September), 108-109.
Hof, R. D. (1997). “Who Will Rule Digital TV?” Business Week, 21 (April), 34-36.
Judson, B (1996). Netmarketing - How Your Business Can Profit From the Online
Revolution. New York: Wolff New Media, 60.
Levinson, J, and Rubin, C (1995). Guerrilla Marketing Online - The Entrepreneur’s Guide To Earning Profits On the Internet. Boston/New York: Houghton Mifflin.
McKenna, R (1991). Relationship Marketing: Successful Strategies For The Age Of The Customer. Addison-Wesley, 2-3, 16, 49, 219.
Peppers, D, and Rogers, M (1993). The One To One Future - Building Relationships One Customer At A Time, 1st ed. New York: Bantam Doubleday Dell.
Rebello, K. Armstrong, L and Cortese, A (1996). “Can You Make Money On the Net?” Business Week, 23 (September), 104-118.
Rebello, K and Lesley, E (1996). “Network Meets Net: How big an audience is there for Microsoft and NBC's cable-Web news venture?” Business Week, 15 (July), 68-70.
Richardson, J. E. (1995). Annual Editions - Marketing 95/96, 17th ed. Guilford:Dushkin, 22, 61-62, 33.
Smedinghoff, T. J. (1996). Online Law - The SPA’s Legal Guide To Doing BusinessOn the Internet. Addison-Wesley Developers Press.
Wheelen, T and Hunger, J. D (1995). Strategic Management and Business Policy, 5th ed. Addison-Wesley.
COMPUTER SELF-EFFICACY AND OUTCOME
EXPECTATIONS AND THEIR IMPACTS ON
BEHAVIORAL INTENTIONS TO USE COMPUTERS IN
NON-VOLITIONAL SETTINGS
Robert W. Stone *
John W. Henry**
A structural equations model of outcome expectancy and computer self-efficacy and their effects on behavioral intentions to use computers was developed for a non-volitional setting. The model was tested using subjects who are managers and executives across organizations and organizational levels. Several of the paths in the model were statistically significant at a 1% level. The significant paths to the computer self-efficacy measure were ease of system use and previous computer experience. Computer self-efficacy had significant paths to both work-related and personal outcome expectancies. Further, both outcome expectancies had significant influences on behavioral intentions to use computers.
INTRODUCTION
I
n recent years, there have been a number of articles modeling the effects of computer self-efficacy and outcome expectancy (Henry, 1989; Henry & Stone, 1995; Henry & Stone, 1997; Hill, Smith, & Mann, 1987;(Venkatesh & Davis, 1996). Most of the models presented in these articles have been variants of the Theory of Reasoned Action (TRA Fishbein & Ajzen, 1975) and employed computer usage as the dependent variable (Hill, Smith, & Mann, 1987). However, as computer use has become, for the most part, mandatory in the workplace, inconsistency problems with attitude scales, which are a key component of models such as the TRA, have begun to appear in the literature (Gattiker & Hlavka, 1992; Gutek, Winter, & Chudoba, 1992; LaLomia & Sidowski, 1991; Pope-Davis & Twing, 1991; Safayeni, Purdy, & Higgins, 1989). For example, Gutek, Winter, and Chudoba, (1992) posit that these inconsistencies may be due to the increase in non-volitional information technology (IT) use. Doll and Torkzadeh (1996) also state that “the realization that, where use is mandatory, measures of system-use may indicate only compliance, not effectiveness” (pp. 1-2). Further, self-efficacy is related to “choice” behavior, not mandatory behavior (Bandura, 1986). These problems are likely to escalate as IT use becomes increasingly non-volitional. Thus, the inclusion of self-efficacy and attitudes in models that are variants of the TRA are not appropriate when computer use is mandatory. These studies suggest that intentions, not actual usage, may better capture the effects of computer self-efficacy, outcome expectations and their antecedents, and the “true” success of an information technology.
THE THEORETICAL FRAMEWORK
This research is founded in Bandura's self-efficacy theory (Bandura, 1982; Bandura, 1986) which provides a sound theoretical basis for examining the determinants of self-efficacy and outcome expectancy and the subsequent effect on an individual's behavioral intentions. Self-efficacy theory emphasizes the impact of the individual's cognitive state on outcomes such as loss of control, low self-confidence, lowered achievement motivation, and perceptions of future outcomes (Bandura, 1986; Meier, 1985; Seligman, 1990). It also provides a theoretical basis for describing behavioral and affective reactions to information technology (IT) (Martinko, Henry, and Zmud, 1995; Rafaeli and Sutton, 1986; Carey, 1992, Kahn and Robertson, 1992). Self-efficacy theory is part of a larger group of psychological theories described as expectancy-value theories (Maddux, Norton, and Stoltenberg, 1986). Self-efficacy theory proposes that an individual's expectations are the primary determinants of affective and behavioral reactions in numerous scenarios involving motivation, performance, and feelings of frustration associated with repeated failure. Specifically, self-efficacy theory states that environmental and personal factors such as verbal (i.e., social) persuasion, actual experience (enactive mastery), and emotional arousal influence expectations that subsequently affect individual outcomes (Bandura, 1986; Lent, Lopez, and Bieschke, 1991. Bandura, 1982 and 1986) separated expectations into self-efficacy and outcome expectancy and posits that these expectancies affect individual behavioral and affective outcomes. Self-efficacy refers to an individual's belief in their ability to accomplish a task (Bandura, 1986). Self-efficacy affects persistence and influences the individual's perception of future outcomes. Outcome expectancy refers to an individual's belief that task accomplishment (i.e., a satisfactory level of performance) leads to desired outcomes. It has been shown that this outcome expectancy is actually two distinct constructs, one related to personal outcomes and the other to work outcomes (Henry & Stone, 1993). Self-efficacy theory posits that these expectancies are directly or indirectly a result of inactive mastery, vicarious experience, verbal persuasion and emotional arousal. The value of expectancies lies in the notion that not only is there a direct relationship between expectancies and behavioral and affective outcomes, but that the relationship is causal (Sadri and Robertson, 1993). Some success has been achieved in the identification and operationalization of the antecedent constructs of computer self-efficacy and outcome expectancy in the IT literature. Perhaps the most salient factors derived from this body of research are management and peer support (Leonard-Barton & Deschamps, 1988; Zmud, 1984), ease of system use (Davis, 1989 Franz, 1991 Guimaraes, Igbaria, & Lu, 1992) and previous experience; (Glass & Knight, 1988). Such a notion implies that self-efficacy and outcome expectancy actually serve as mediating variables (Shell, Bruning and Murphy, 1989; Lent et al., 1991).
A review of the IT literature suggests that constructs similar to enactive mastery, verbal persuasion, and emotional arousal appear repeatedly as independent variables (e.g., DeSantis, 1983; Nelson, 1990; Davis, Bagozzi and Warshaw, 1989). These constructs are often implicitly and sometimes explicitly operationalized as antecedents to the reactions of IT end-users, but these distinctions are not often clearly made. However, there has been some IT research that has explicitly used measures of self-efficacy. A review of the studies making use of various measures of computer self-efficacy revealed moderately consistent findings. For example, Hill, Smith and Mann (1987) reported that self-efficacy influenced an individual's decision to use computers. Gist, Schwoerer and Rosen (1989) found that self-efficacy influenced an individual's decision to learn a computer language. In a study devoted to gender differences Miura (1987) found that men rated themselves higher than women on computer self-efficacy. In addition, numerous computer self-efficacy scales have been developed (Henry and Stone, 1997; Murphy, Coover, & Owen, 1989, Hill, Smith and Mann, 1987).
However, there is a lack of studies which directly measure outcome expectancy in the IT literature. These studies include Davis, Bagozzi and Warshaw (1989) who found that subjects developed behavioral intentions about using a wordprocessing program based on expectations that it would improve their performance in a MBA program. Hill, Smith, and Mann (1987) found that “outcome beliefs” influenced subjects’ attitudes and decisions about learning a computer language. Rafaeli and Sutton (1986) found that clerical personnel using word processors showed a decrease in their certainty about how using word processors would affect their job in the future.
The purpose of this research is to integrate the expectancies and their antecedents as described above in a model of IT use with behavioral intentions as the dependent variable. The examined model depicts the functional relationships among the antecedents, computer self-efficacy, outcome expectancies, and behavioral intentions. The model hypothesizes that outcome expectancies have direct impacts on the individual's behavioral intentions to use computers. The expectancies are influenced by computer self-efficacy. Finally, the antecedents of previous computer experience, ease of system use, boss' encouragement and support, and management support for the system are proposed to impact computer self-efficacy.
Enactive mastery skills are operationalized as computer experience (Gist et al., 1989; Hill et al., 1987). Emotional arousal is represented by ease of system use (Martocchio and Webster, 1992; Carey, 1992; Davis, Bagozzi and Warshaw, 1989). Ease of system use, in this study, refers to the degree to which the end-user likes the system and finds it easy to use, (i.e., the functionality of the system). Since support comes from many sources, the review of the IT literature showed that management and “boss” support were most relevant. Support represents the notion of verbal (i.e., social) persuasion. Management's message of support, although not explicitly stated in many studies, can be found in actions such as the development of training sessions, continuous updating of systems and re-training, and providing mechanisms for helping IT users solve problems. Thus, the theoretical model shown in Figure 1 examines how the impacts of support, computer experience, and system ease of use affect the end-user's behavioral intentions to use computers mediated by computer self-efficacy and outcome expectancy.
[INSERT FIGURE 1 ABOUT HERE]
RESEARCH METHODOLOGY
The Sample
The examination of these relationships began with the development of a questionnaire. The target population was business executives and managers who are computer users in their work. Questionnaires were mailed to 3000 of these executives selected randomly from a purchased, national mailing list. A total of 411 usable returns were received producing approximately a 14% response rate. Among the questions on the survey was an item allowing respondents to self-report whether or not their work-related computer use was voluntary or mandatory. Using the responses to this question, the sample was partitioned. The 105 individuals who reported volitional use were excluded from the sample leaving 306 respondents who were nonvolitional computer users. These nonvolitional respondents formed the sample that was used in the analysis.
Response bias was studied by comparing late respondents to early respondents. Late respondents were defined as the upper quartile of responses when ordered by response date. Early respondents were captured in the lower quartile of the ordered responses. The early respondents were used to simulate the respondents in the sample while the late respondents simulated nonrespondents. These two groups were then compared using t-tests for differences in the demographics (Armstrong and Overton, 1977). The results from the t-test found no meaningful differences. The specific t-values were: age -0.03; years worked for the current organization 0.74; gender 1.38; level in the organization 0.71; percentage of time using the system 0.25; and the number of employees in the organization -1.01. Thus, response bias should not present a problem in the study.
The Measures and Their Psychometric Properties
In order to evaluate the measurement part of the model (i.e., quality of the measures), the first step of a two-step method was used (Anderson and Gerbing, 1988). The first step evaluates the measurement model (i.e., the measures and their properties) and the second the structural model (i.e., the paths among the constructs). The measurement model was evaluated using a confirmatory factor analysis where all measures were exogenous (standard deviations set to one) and allowed to pairwise correlate. Each indicant had its path to its measure free to vary and a disturbance term that was also free to vary. The estimation used CALIS (i.e., Covariance Analysis of Linear Structural Equations) in PC SAS version 6.11 and maximum likelihood estimation. The standardized path coefficients between the indicants and the measures were used to evaluate the psychometric properties of the measures. The questionnaire items, the measures, and the psychometric properties are shown in Table 1.
[INSERT TABLE 1 ABOUT HERE]
The fit of the confirmatory factor analysis was good, as indicated by several statistics. The goodness of fit index was 0.87 and this index adjusted for degrees of freedom was 0.82. The root mean square residual was 0.13. Bentler's comparative fit index was 0.92 and Bentler and Bonett's non-normed index was 0.90 and the normed index was 0.87. Similarly, Bollen's normed and non-normed indexes were 0.84 and 0.92, respectively. The factor loadings (i.e., standardized path coefficients) ranged from 0.62 to 0.96. The lowest composite reliability coefficient was 0.76 and the highest was 0.87. Similarly, the percentages of shared variance for the measures ranged from 52% to 70%.
Discriminant validity was also examined. Discriminant validity focuses on whether or not the items composing a measure can differentiate between their own measure and all other measures in the study. A method evaluating discriminant validity is if the squared correlation between pairs of measures is less than the average percentage of shared variance for both measures. The shared variance represents the correlation among the items within the measure. The comparison of the shared variance to the squared correlation examines the between measure and within measure strengths of these correlations. Comparing the estimated squared correlations to the average percentage of shared variance for each measure indicated that all the measures satisfy discriminant validity (Fishbein & Ajzen, 1975). These squared correlations (computed from the confirmatory factor analysis) include 0.00 for the relationships between management support-previous computer experience; boss’ encouragement and support-work-related outcome expectancy; boss’encouragement and support-computer self-efficacy; management support-personal outcome expectancy; ease of system use-personal outcome expectancy; and previous computer experience-behavioral intentions to use computers. The relationships with squared correlations of 0.01 were: previous computer experience-ease of system use; previous computer experience-boss’ encouragement and support; previous computer experience-work-related outcome expectancy; management support-behavioral intentions to use computers; and boss’ encouragement and support- behavioral intentions to use computers.
The relationships between boss’encouragement and support-ease of system use; previous computer experience-computer self-efficacy; management support-computer self-efficacy; previous computer experience-personal outcome expectancy; and work-related outcome expectancy-personal outcome expectancy all had squared correlations of 0.02. All the following relationships had squared correlations of 0.05. These relationships are management support-ease of system use; boss’encouragement and support-personal outcome expectancy; and ease of system use-behavioral intentions to use computers. The remaining squared correlations were: 0.25 for management support-boss’encouragement and support; 0.07 for management support-work-related outcome expectancy; 0.27 for ease of system use-work-related outcome expectancy; 0.27 for ease of system use-computer self-efficacy; 0.11 for work-related outcome expectancy-behavioral intentions to use computers; and 0.30 for computer self-efficacy-behavioral intentions to use computer system. All of these squared correlations are less than the shared variances reported in Table 1. Thus, discriminant validity is satisfied.
In summary, the psychometric properties of these measures were satisfactory. This statement is based upon the following observations. First, item reliability was satisfied since all the factor loadings (i.e., standardized path coefficients) were larger than 0.61 (Fornell & Larcker, 1981). Second, because all the composite reliability coefficients were 0.76 or higher, reliability was satisfied (Nunnally, 1978). Third, since the average percentage of shared variance for each measure was 52% or larger, these measures display satisfactory values of shared variance (Igbaria & Greenhaus, 1992). From these results, it can be concluded that the measures display satisfactory convergent validity (Quick & Quick, 1984). With convergent validity and discriminant validity satisfied, it is implied that the measures possess construct validity (Quick & Quick, 1984).
Estimation of the Model
The proposed model was examined, in the second step of the two-step process, using the previously discussed questionnaire items and responses. The technique of structural equations with latent variables was used to estimate the model. The measures of previous computer experience, management support for the system, ease of system use, and boss' encouragement and support were exogenous in the model. These measures had standard deviations set equal to one. The remaining measures were endogenous and scaled by setting a path between an indicant and its measure to one. All other indicants had paths between their measure and themselves free to vary. Each indicant and measure were impacted by a stochastic disturbance term. The estimation procedure used was CALIS in PC SAS version 6.11 and maximum likelihood estimation.
THE RESULTS
The overall fit of the model to the data is illustrated by several measures. The goodness of fit measure was 0.87 and this same measure corrected for the degrees of freedom in the model was 0.83. The root mean square residual was 0.07. The normed Chi-square statistic was 1.88. Bentler's comparative fit index was 0.94 while the incremental fit indexes ranged from 0.86 to 0.94. For all the fit indexes, the closer its value is to one the better the fit between the model and the data. A rule of thumb for acceptable or good fit is for these indexes to be 0.90 or higher. However, for models with a large number of observations (i.e., greater than 200), values less than 0.90 can still indicate an acceptable fit. This is particularly true when the normed chi-square statistic is 2 or less (Hair, Anderson, Tatham, and Black, 1992). These results indicated that the fit of the model is acceptable (Henry & Stone, 1995). These statistics are displayed in Table 2.
Table 2
The Summary Statistics of the Model’s Fit to the Data
|Summary Statistic |Value |
|Goodness of Fit |0.87 |
|Adjusted Goodness of Fit |0.83 |
|Root Mean Square Residual |0.07 |
|Normed Chi-Square |1.88 |
|Bentler’s Comparative Fit Index |0.94 |
|Bentler & Bonett’s Non-normed Index |0.93 |
|Bentler & Bonett’s Normed Index |0.88 |
|Bollen’s Normed Index |0.86 |
|Bollen’ Non-normed index |0.94 |
Each indicant had a significant path from its latent variable, using a 1% significance level. Similarly, the disturbance terms for each indicant and latent variable were statistically significant at a 1% level. Additionally, the exogenous measures were allowed to pairwise correlate. Three of the six correlations were statistically significant at a 1% level. These significant correlations were between ease of system use and management support for the system; boss' encouragement and support and management support for the system; and boss' encouragement and support and ease of system use.
Several of the paths in the structural model were statistically significant at a 1% level. The significant paths to computer self-efficacy were from ease of system use and previous computer experience. The paths from computer self-efficacy to personal outcome expectancy and work-related outcome expectancy were also statistically significant. The significant paths to behavioral intentions to use computers were from both outcome expectancies. Thus, the antecedents of previous computer experience and ease of system use had significant, indirect impacts on behavioral intentions to use computers mediated by computer self-efficacy and the outcome expectancies. The details of these results are shown on Figure 2.
[INSERT FIGURE 2 ABOUT HERE]
DISCUSSION AND CONCLUSIONS
As mentioned above, previous experience and ease of system use both had positive impacts on computer self-efficacy. As predicted, computer self-efficacy had positive effects on both work-related and personal outcome expectancy. Moreover, work-related and personal outcome expectancies had positive effects on behavioral intentions to use computers. As stated by Boyd and Vozikis (1994, p. 65), “Self-perception, or the way in which a person perceives his or her abilities and tendencies, plays a role in the development of intentions. Similarly, self-efficacy affects a person's beliefs regarding whether or not certain goals may be obtained.” Intention is more closely related to a “state of mind” that reflects a perceived value in performing a certain behavior that cannot be captured in variables such as “computer usage”, especially in a nonvolitional context.
Furthermore, intentions are determined by rational/analytic thought (e.g., goal directed behavior) and vision (Boyd & Vozikis, 1994), suggesting that individuals who intend to use technology may be more innovative and creative than the individual who uses technology because such use is mandatory. This is extremely important in today's business climate with the emphasis on new product development and reengineering of business processes. Thus, individuals who possess intentions to use technology may demonstrate higher degrees of initiation, persistence, and performance.
The study also indicates that managers should examine intentions as well as actual computer usage when evaluating the success of any system. The notion of intentions is often overlooked since it is not a concrete measure of usage (e.g., time on system), but may be a more valid indicator of the extent of actual usefulness of the system. The identification of individuals who intend to use technology may be useful in determining success and performance, particularly in mandatory settings.
REFERENCES
Anderson, James C. and Gerbing, David W. (1988). “Structural Equation Modeling in Practice: A Review and Recommended Two-step Approach,” Psychological Bulletin, 103, 411-423.
Armstrong, J. Scott and Terry S. Overton. (1977). “Estimating Nonresponse Bias in Mail Surveys,” Journal of Marketing Research 14 (August), 396-402.
Bandura, A. (1982). “Self-efficacy Mechanism in Human Agency,” American
Psychologist, 37, 122-147.
Bandura, A. (1986). Social Foundations of Thought and Action: A Social
Cognitive Theory. New Jersey: Prentice-Hall, Inc.
Boyd, N. G., and Vozikis, G. S. (1994). “The Influence of Self-efficacy on the Development of Entrepreneurial Intentions and Actions,” Entrepreneurship Theory and Practice, Summer, 63-77.
Carey, J. M (1992). “Job Satisfaction and Visual Display Unit (VDU) Usage: an Explanatory Model,” Behaviour & Information Technology (11:6), , pp. 338-344.
Davis, F. D. (1989). “Perceived Usefulness, Perceived Ease of Use, and User Acceptance of Information Technology,” MIS Quarterly, 13: 319-340.
Davis, F. D., Bagozzi, R. P., & Warshaw, P. R. (1989). “User Acceptance of Computer Technology: A Comparison of Two Theoretical Models,” Management Science, 35(8), 982-1003.
DeSantis, G. (1983). Expectancy Theory as an Explanation of Voluntary Use of a Decision-support System. Psychological Reports, 52, 247-260.
Doll, W.J., & Torkzadeh, G. (1996). “Measuring How Information Technology Is Used in an Organizational Context,” Unpublished Manuscript.
Fishbein, M. & Ajzen, I. (1975). Belief, Attitude Intentions and Behavior: An Introduction to Theory and Research. Addison Wesley, Boston, Mass.
Fornell, C. & Larcker, D. F. (1981). “Evaluating Structural Equation Models with Unobservable Variables and Measurement Errors,” Journal of Marketing Research, (18), pp. 39-50.
Franz, C. R. (1991). “Descriptive Model for End-user Acceptance of Information Centers,” Information Resources Management Journal, 10: 14-27.
Gattiker, U. E & Hlavka, A. (1992). “Computer Attitudes and Learning Performance: Issues for Management Education and Training,” Journal of Organizational Behavior, 13, 89-101.
Gist, M. E., Schwoerer, C., & Rosen, B. (1989) “Effects of Alternative Training Methods on Self-Efficacy and Performance in Computer Software Training,” Journal of Applied Psychology (74), , pp. 884-891.
Glass, C. R., & Knight, L. A. (1988). “Cognitive Factors in Computer Anxiety,” Cognitive
Therapy and Research, 12: 351-366.
Guimaraes, T., Igbaria, M., & Lu, M. (1992). The Determinants of DSS Success: An Integrated Model. Decision Sciences, 12: 409-430.
Gutek, B. A., Winter, S. J, & Chudoba, K. M. (1992). “Attitudes Toward Computers: When Do They Predict Computer Use?” Proceedings of the Fifty-second Annual Meeting of the Academy of Management. Las Vegas, NV.
Hair, Joseph F., Anderson, Rolph E., Tatham, Ronald L., and Black, William C. (1992). Multivariate Data Analysis with Readings, third edition, MacMillian: New York.
Henry, J. W. (1989). “Learned Helplessness: A Factor Moderating the Impact of Technology in the Workplace,” Unpublished Doctoral Dissertation.
Henry, J. W., & Stone, R. W. (1993). “The Development and Validation of Computer Self-efficacy and Outcome Expectancy Scales”. Proceedings of The Twenty-ninth Annual Meeting of The Institute of Management Sciences Southeastern Chapter, 139-141.
Henry, J. W. & Stone, R. W. (1995). “A Structural Equation Model of Job Performance Using a Computer-based Order Entry System,” Behaviour & Information Technology, 14 (3), 163-173.
Henry, J. W. & Stone, R. W. (1997) “The Development and Validation of Computer Self-Efficacy and Outcome Expectancy Scales in a Non-Volitional Context,” Behavior, Research Methods, Instruments, & Computers, 29 (4), 45-53.
Hill, T., Smith, N. D., & Mann, M. F. (1987). “Role of Efficacy Expectations in Predicting the Decision to Use Advance Technologies: The Case of Computers”. Journal of Applied Psychology, 72, 307-313.
Igbaria, M. and Greenhaus, J. H. (1992). “Determinants of MIS Employee's Turnover Intentions: A Structural Equation Model,” Communications of the ACM, 35, 35-49.
Kahn, H., & Robertson, I. T. (1992) “Training and Experience as Predictors of Job Satisfaction and Work Motivation When Using Computers,” Behaviour and Information Technology (11:1), pp. 53-60.
Kanter, R. M. “Managing the Human Side of Change,” Management Review (74), 1985,
pp. 52-59.
LaLomia, J. J., & Sidowski, J. B. (1991). “Measurements of Computer Attitudes: A Review,” International Journal of Human-Computer Interaction, 3(2), 171-197.
Lent, R. W., Lopez, F. G., and Bieschke, K. J (1991). “Mathematics Self-Efficacy: Sources and Relation to Science-Based Career Choice,” Journal of Counseling Psychology (38:4), pp. 424-430.
Leonard-Barton, D., & Deschamps, I. (1988). “Managerial Influence in the Implementation of New Technology,” Management Science, 34: 1252-1265.
Maddux, J. E., Norton, L. W., & Stoltenberg, C. D (1986). “Self-Efficacy Expectancy, Outcome Expectancy, and Outcome Value: Relative Effects on Behavioral Intentions,” Journal of Personality and Social Psychology (51:4), pp. 783-789.
Martinko, M. J., Henry, J. W., and Zmud, R. W. (1996). “An Attributional Explanation of Individual Resistance to the Introduction of Information Technologies in the Workplace,” Behaviour & Information Technology, 15(5), 313-330.
Martocchio, J. J., and Webster, J. (1992). “Effects of Feedback and Cognitive Playfulness on Performance in Microcomputer Software Training,” Personnel Psychology, 45, 553-578.
Meier, S. T. (1985). “Computer Aversion,” Computers in Human Behavior, 1, 171-179.
Miura, I. T. (1987) “The Relationship of Computer Self-Efficacy Expectations to Computer Interest and Course Enrollment in College,” Sex Roles (165/6), pp. 303-311.
Murphy, C. A., Coover, D., & Owen, D. V (1989) “Development and Validation of the Computer Self-Efficacy Scale,” Educational and Psychological Measurement (49), 893-899.
Nelson, D. L (1990). “Individual Adjustment to Information-Driven Technologies: A
Critical Review,” MIS Quarterly (14), pp. 79-98.
Nunnally, J. (1978). Psychometric Methods (2nd ed.). New York: McGraw-Hill.
Pope-Davis, D. B., & Twing, J. S. (1991). “The Effects of Age, Gender, and Experience on Measures of Attitude Regarding Computers,” Computers in Human Behavior, 7, 333-339.
Quick, J. C., and Quick, J. D. (1984). Organizational Stress and Preventive Management. New York: McGraw-Hill.
Rivard, S., & Huff, S. L (1988). “Factors of Success for End-User Computing,” Communications of the ACM (31), pp. 552-561.
Sadri, G., & Robertson, I. T (1993). “Self-Efficacy and Work-Related Behaviour: A Review and Meta-Analysis,” Applied Psychology: An International Review (42), pp. 139-152.
Safayeni, F.R., Purdy, R.L., & Higgins, C.A. (1989). “Social Meaning of Personal Computers for Managers and Professionals: Methodology and Results,” Behaviour and Information Technology, 8(2), 99-107.
Seligman, M. E. P. (1990). Learned Optimism. New York: Alfred A. Knopf, Inc.
Shell, D. F., Bruning R. H., & Murphy, C. C (1989). “Self-Efficacy and Outcome Expectancy Mechanisms in Reading and Writing Achievement,” Journal of Educational Psychology (81:9), pp. 91-92.
Tannenbaum, S. I., Mathieu, J. E., Salas, E., & Cannon-Bowers, J. A (1991). “Meeting Trainees' Expectations: The Influence of Training Fulfillment on the Development of Commitment, Self-Efficacy, and Motivation,” Journal of Applied Psychology (76:6), pp. 759-769.
Venkatesh, V., & Davis, F. D. (1996). “A Model of the Antecedents of Perceived Ease of Use: Development and Test,” Decision Sciences, 27(3), 451-481.
Walton, R. E. Up and Running. Harvard Business School Press, Boston, 1989.
Zmud, R. W. (1984). “An Examination of 'Push-Pull' Theory Applied to Process Innovation in Knowledge Work,” Management Science, 30: 727-738.
UTILIZATION OF ARTIFICIAL
INTELLIGENCE TECHNOLOGIES IN ADULT
DISTANCE LEARNING
Owen P. Hall, Jr,*
Farzin Madjidi **
Working adults in evening business program often have difficulty with courses involving quantitative reasoning. The most common complaint among these students is the pace of the course and, due to the fact that they may be only on campus one night a week, it is often difficult for them to utilize the learning resources available to full-time students. Recent developments in AI technology have brought about the capability to significantly improve the self-learning process. This paper proposes and outlines the steps required to develop an expert system-based learning program for working adults who have returned to the classroom.
INTRODUCTION
S
tudents in both undergraduate and graduate business programs often have difficulty with courses involving quantitative reasoning. This problem is compounded in evening degree programs that are tailored toward working adults. A majority of working adults have been away from school for a number of years. Often, their skills in mathematics have deteriorated over time. Furthermore, many of these students report a less than satisfactory experience with these types of courses. As a result, a large of number of students in evening programs find themselves struggling in their courses involving quantitative reasoning, e.g., statistics, and a few even quit these programs. The most common complaint among these students is the pace of the course. They often claim, that if given more time to absorb the material, their overall understanding and performance would substantially improve. However, due to the fact that they are often on campus only one night a week, and are often involved in job related travel, it is difficult for them to utilize the learning support resources available to full time residential students (Kerka 1992).
As a result of these conditions, adult learners often seek other approaches for mastering the subject material. These alternatives include tutors, group processing and videos. Unfortunately, each of these approaches has a number of serious limitations. Computer-based information technology is a relatively new approach to providing students with access to self-paced study programs Although useful, a majority of the currently available tutorial software packages lack the sophistication and flexibility to accommodate the growing demands for adult education. System requirements for adult education and training are fundamentally different than those associated with most traditional business programs, (Newstrom 1991, pp.43-48). For example, “Adult learners have a deep and powerful drive to be self-directing to be in charge of their developmental destinies and to take control of their learning processes.” This observation is one of the basic assumptions of Malcolm Knowles’ andragogical paradigm which addresses the educational and training needs of the working adult (Lawson,1997, p.10).
In recent years, advances in technology have significantly enhanced the capabilities and affordability of PC’s for business education applications. For example, the availability of inexpensive CD-ROM (Compact Disks - Read Only Memory) drives has added the dimension of sound and motion to what used to be mundane programs. Presently, CD-ROM-based, interactive media, commercial software packages are available for teaching basic business subjects such as statistics and accounting. These packages incorporate sound, music, and animation and produce a comprehensive and entertaining teaching software package.
Over the past few years, technology has been used increasingly and successfully in teaching second languages. Under the umbrella of Computer Assisted Instruction (CAI), individual students, whether in a classroom or media center or over long-distance computer networks, are studying second languages. Learning components such as speaking, listening, reading, writing, and culture are identified first. Then, the appropriate technology for delivering each component is determined. For example, interactive audio programs allow students to practice with other students. Interactive video disk programs are used for listening comprehension activities. Techniques ranging from fill-in-the-blank, multiple-choice, word recognition, and word processing in the target language can be used to enhance a student’s writing skills. Cultural differences and similarities in their proper context can be demonstrated using video-based activities. And finally, computer-assisted testing can be used to measure the level of learning that has taken place (Willetts, 1992, p.4-7) These capabilities can be enhanced as technology becomes more sophisticated.
A common flaw with most CAI commercial applications is the “One-size-fits-all” nature of these packages. Most tutorial programs offer a “canned” package of learning experiences that disregard whether the learner is a novice or advanced, or even how a particular user learns best. This problem is even more critical when dealing with technical subjects such as statistics and operations management. Ideally, the software would include an “expert system” driven diagnostic module which would prescribe learning experiences for the user. That is, based on a series of questions and the responses, the software would diagnose the needs of the learner and determine the appropriate approach to a certain topic. These capabilities do not generally exist in CAI software. However, Expert System (ES) technology will allow for direct incorporation of these features.
This paper proposes the development of an Adaptive Learning System (ALS) software package that considers the specific needs of a user and provides the appropriate learning experience. The software features:
A series of topics divided into learning components that become the building blocks of this system. Each learning component is presented in a multi-step sequence of learning. The learning system consists of text material, video presentations and simulation analysis.
An adaptive routine that changes the sequence of learning and difficulty level within each learning component based on the user’s responses to questions posed.
A diagnostic feature using expert knowledge to determine the appropriate learning experience and mode of delivery.
An ability to reference additional support material for subsequent study.
A capability to save a partially completed learning session.
The capacity to provide this service via the Internet.
EXPERT SYSTEMS OVERVIEW
An Expert System (ES) is a decision support tool designed to computerize the sum of available knowledge and rules that emulate the human decision-making process. Perhaps one of the most comprehensive definitions of an ES is given by Parsaye and Chingnell (1988, p.1) as “A program that relies on a body of knowledge to perform a somewhat difficult task usually performed by a human expert. The principal power of an expert system is derived from the knowledge the system embodies rather than from search algorithms and specific reasoning methods. An expert system successfully deals with problems for which a clear algorithm solution exists.” The purpose of an ES is to assist or replace a manager or an operator. Expert systems augment conventional educational support programs such as databases, word processors, and spreadsheet analysis.
Some recent applications of ES have been in psychological testing, investment management research and development tourism management, total quality management, crime investigation, and qualitative and quantitative analyses, as well as numerous applications in marketing, strategic planning, chemical analysis, project/production management, and engineering. Most recent applications of expert systems are no longer stand-alone, but rather software applications “embedded” in a larger software system. Many commercial statistical analysis programs, data management systems, information management systems, and other data analysis systems now contain embedded heuristics that constitute expert system components of the package.
Expert systems contain an ES shell or an interpreter, a “knowledge base” or system of related logic or rules that enable the computer to approximate human knowledge, and a more sophisticated user interface. The ES shell simplifies the process of creating a knowledge base. It is the shell that actually processes the information entered by a user, relates it to the concepts and rules in the knowledge base, and provides an assessment or a solution to a problem. The knowledge base provides the connection between the concepts, ideas, and statistical probabilities that allow the reasoning part of the system to perform an accurate evaluation of a problem. They also rely on associative relationships among different concepts, statistical probabilities of certain solutions, or simply large databases of facts that can be compared to one another based on simple conventions intrinsic to the expert system. In arriving at decisions, expert systems use probabilities, fuzzy logic, or certainty theory. The newer Graphical User Interface (GUI) designs and pop-up menus are quite common in ES applications. Typically, expert systems can be classified into one of three basic types: production, frame-based and logic. This paper focuses on the use of a production expert system whose primary characteristics are as follows:
Knowledge Base - Contains the facts and data relevant to a particular problem.
Inference Engine - Provides the control and logic for applying the knowledge.
Explanation System - Furnishes a description of the proposed problem solution.
User Interface - Provides the linkage between the user and inference engine.
Knowledge Acquisition System - Updates the knowledge base with new facts.
A summary of selected commercially available production ES shells is given in Appendix A.
ADAPTIVE LEARNING SYSTEM STRUCTURE
Topics in quantitative reasoning are broken down into learning components. Each learning component is then segmented into a three-step sequence of learning: calculation, interpretation, and application. The steps in the sequence of learning are designed to be progressively more difficult. Every step builds on the previous step, thus providing continuity and repetition. The first step focuses on generation of a quantitative statistic or parameter. The focus of the second step is on analysis; that is, interpreting the measure within the specific context of the given problem. The third step requires synthesis or application of the measure. At this step, the user is given a small case study. The user must identify whether or not the quantitative principle being studied is appropriate given the facts of the case, has the measure been correctly interpreted, and are the conclusions drawn in the case accurate. Within each step in the sequence, a database of problems at varying levels of difficulty exists. The user begins at the least difficult level and progresses toward the most difficult level. Using an adaptive routine, the ALS will adjust the level of difficulty based on user responses. In this case, the ALS consists of three levels of difficulty. In practice, the system can consist of any number of difficulty levels, however, system complexity increases dramatically as additional levels are included. Figure 1 presents the logic structure for the prototype ALS. The learning process begins with the user selecting a particular study category from the topical selection menu, e.g., descriptive statistics. The user is then exposed to a brief review of the basic principles associated with the selected topic (mode 1 instruction).
The Adaptive Routine
The adaptive routine’s main function is to advance the user through the different layers of the ALS and to keep track of the learner’s progress. At each step, once the user completes a level of difficulty, the adaptive routine records that completion and moves to the next, more difficult step. Typically, each level would contain three or four questions. Once the user completes all levels in a section, the adaptive routine moves the user to the next section until the user completes the entire learning module. The adaptive routine notes areas where the user experiences difficulty, and alerts the knowledge base for a diagnosis and a suggested remedy.
The Diagnostic Feature
The diagnostic feature of the system uses expert knowledge to assess the user’s problems and to select the appropriate learning experience. Since the system is organized in three distinct sequential learning steps, namely calculation, interpretation and application, the diagnosis process also revolves around the same three steps. That is, the inference engine will first decide whether the user’s problem is with one of the three steps and then determine at what level of difficulty this problem exists. For example, a user may prove unable to calculate a statistic at the lowest level of difficulty. Obviously, the problem involves a calculation deficiency, therefore, no diagnosis will be required. So, the inference engine will select an appropriate learning experience as follows: Suppose the selected learning experience showed a video of the computational process. Further suppose that the user completes the learning experience and successfully completes another low difficulty calculation. The inference engine will keep track of the user’s success and the medium of delivery that was successfully used (a video of an example) to present the material. Now, suppose the user advances to the application level but cannot solve the problem presented.
FIGURE 1 - LOGIC STRUCTURE OF THE PROTOTYPE ALS
[pic]
The inference engine will determine if the inability to solve the problem is due to misunderstanding the application, interpreting it, or simply miscalculating it. If the problem is with the calculation, the inference engine will consider the user’s previous difficulty with calculating and recommend a calculation learning experience at the appropriate difficulty level. Furthermore, the inference engine may use another video for presentation. Once the presentation is completed, the user is returned to the last successful step and the process continues.
Selection of Instructional Delivery Mode
The following general methodology is used to determine the most appropriate mode of delivery:
Suppose a user answers a low level of difficulty question incorrectly. The rule requires that the knowledge base identify the source of the problem using one of the prescribed categories that users may have trouble with. Also, the rule requires the inference engine to outline the material using a video presentation (mode 2).
If the user answers a mid-level of difficulty question correctly, then the inference engine will record that the user responded well to a video presentation having this category of questions. A video presentation will then become the first choice of the knowledge base when another question of this type is answered incorrectly by the user.
FIGURE 2 - OVERVIEW SCHEMATIC OF THE ADAPTIVE LEARNING SYSTEM
[pic]
[pic]
[pic]
[pic]
Should the user answer the mid level question incorrectly, then the inference engine will determine that the video presentation was ineffective and present the user with a simulation (step-by-step solution - mode 3). If the user answers a high level of difficulty question correctly, then the knowledge base will record that the user responded well to a simulation when having difficulty with this category of questions.
Current technology allows the incorporation of a large database of questions with varying levels of difficulty, a number of presentations in each delivery mode, and the expert systems together on one CD-ROM. Another feature of the ALS is the ability to save a partially completed learning session. Figure 2 presents a schematic overview of the proposed adaptive learning system.
HOW THE ADAPTIVE LEARNING SYSTEM WORKS
The proposed ALS will be illustrated using statistical confidence intervals, a common but often misunderstood basic tenet of modern business management. The user may choose to begin at any of the three modes available. The ALS begins with a classical presentation of confidence intervals for the proportion. This presentation will be available in three forms: A textual description of confidence intervals similar to those found in textbooks; a specially prepared video presentation of a classroom lecture on confidence intervals; and a computer simulation that moves the student through all the pertinent steps. The user first completes the textual presentation. Upon completion of the textual presentation, the user is given a confidence interval calculation at the low difficulty level. If the user solves the easy problem, the adaptive routine will advance the user to the next difficulty level. If the user is successful in calculating the medium and high difficulty level problems, they will advance to the interpretation step.
At the interpretation step, as in the calculation step, the topic is first presented using the classical approach of textual presentation. Unless the user fails to solve a problem at one of the degrees of difficulty, they will go through the same process as in the calculation step and proceed to the application step where the same routine is repeated. Should the user fail to solve a problem correctly at any step, the adaptive routine alerts the knowledge base for a diagnosis. The knowledge base will first determine if the problem a user is having is mathematical, analytical or in application of the statistical measure. Once this determination is made, the inference engine, based on the level of difficulty the user was facing last, and the levels of difficulty the user had completed previously, will determine the level of difficulty appropriate for the user now. Finally, based on expert knowledge and the user’s previous response, the knowledge base will select the appropriate delivery mode of instruction for the user. Therefore, suppose a user has previously answered calculation problems at the low and medium levels of difficulty, but has had problems with calculations at the high level of difficulty. This user has been able to solve problems with which he or she has had trouble after viewing a computer simulation. The appropriate learning experience for this user would be a simulation of the calculation process.
Construction of the Knowledge Base
To construct the knowledge base, a series of rules representing expert knowledge is required. For example, in the case of constructing the confidence intervals for proportions, the ALS would required 12 presentations on the use of distributions (Four presentations in each of three modes of delivery), 12 presentations on how to physically calculate the statistic (Four presentations in each of three modes of delivery), 12 presentations on how to interpret the statistic (Four presentations in each of three modes of delivery), and, nine presentations on the application of the statistic (3 presentations in each of three modes of delivery), for a total of 45 presentations. To illustrate the knowledge base for this system and develop sample codes, a series of rules is developed for the learning component of confidence intervals for proportions. Presented below is a typical series of questions involving confidence intervals for the third step in the learning process (application) followed by the corresponding expert rules.
ALS Example: Confidence Intervals for Proportions (Step 3 - Application)
1. What information do you need to construct a confidence interval for the proportion of employees who are eligible for retirement?
a. Average number of employees, sample size, and the percentage of those who wish to retire.
b. Sample size, the number of those who are eligible for retirement, and the confidence.
c. Confidence level, maximum allowable error, and the proportion of those who retired last year.
d. Both a and b.
2. An estimate based on a 90% confidence interval is?
a. More accurate than one based on a 95% confidence level.
b. Only allows 10% error in the estimate.
c. Only good for larger samples with at least 30 observation.
d. As reliable and accurate as an estimate at any other confidence level.
e. None of the above.
3. If you choose a sample of 29, the appropriate distribution to use is?
a. Z distribution
b. t distribution
c. Insufficient information to determine
d. Binominal distribution
4. Construct a 99% confidence interval for the proportion of those employees who are eligible for retirement, if a sample of 100 employees indicated that 29 were eligible. The correct answer is approximately?
a. 20% to 40%
b. 17.3% to 40.7%
c. 7.3% to 67.3%
d. 25.5% to 32.5%
Rule: The correct answers to parts 1 through 4 are “b,” “e,” “b” and “b.” If the user selects answers all parts correctly, the adaptive routine advances the user to the medium difficulty level.
If the user answers part 1 incorrectly, the knowledge base will determine that the user’s difficulty is with applying confidence intervals. Accordingly, the knowledge base, using an appropriate mode of delivery, will select a presentation that discusses and gives examples of how and when to apply confidence intervals and what data is needed to calculate them.
If the user answers part 1 correctly, but part 2 incorrectly, the knowledge base will determine that the user’s difficulty is with understanding how to interpret a confidence interval. Accordingly, the knowledge base, using an appropriate mode of delivery, will select a presentation that includes illustrated examples that describe the correct interpretation of a confidence interval.
If the user answers parts 1 and 2 correctly, but part 3 incorrectly, the knowledge base will diagnose the problem as difficulty using the appropriate distribution. The knowledge base will select an appropriate presentation on how and when to use which distribution and its accompanying table.
If the user answers parts 1, 2, and 3 correctly, but part 4 incorrectly, then the knowledge base determines that the user has difficulty with the mathematics of the problem and presents the user with a step-by-step solution of the formula using the appropriate mediu
An example rule base for this application is given in Appendix B. This summary contains the sequence of learning and the levels of difficulty. At each level of difficulty, the appropriate rule is assigned a rule number and the rules, corresponding commands, and the description of the rule are stated. This particular application was executed using the VP Expert shell.
SUMMARY AND CONCLUSIONS
Increasing pressures on working adults to remain technically competitive call for new methods of delivering high quality business education. This is particularly true for courses and subjects involving quantitative reasoning. An adaptive learning methodology holds considerable promise for addressing many of the problems associated with adult education programs, particularly those involving distance learning. Adaptive learning systems have inherent advantages over more conventional approaches to developing learning and training programs. Curry and Moutinho identify four such advantages:
More understandable and accessible expertise.
Models encapsulate qualitative reasoning as compared to numeric processing.
The models can conduct a consultation process with the user.
The models can provide an explanation of their conclusions and give advice.
The proposed ALS outlined in this paper takes advantage of these benefits. The diagnostic feature proposed in this model utilizes qualitative reasoning to guide the user and to give the user clear explanations. The ability to diagnose and pinpoint the reason the user fails to answer a problem allows for a specific remedy as opposed to just another attempt at the same problem. The diagnostic feature embedded in the adaptive routine allows the user to master all facets of a learning component while, at their pace, navigating through increasing levels of difficulty. The user not only learns how to perform calculations, a common feature in all tutorial software, but also learns how to interpret and apply the analysis; a must for any course in quantitative reasoning.
The construction of a knowledge base that contains enough rules for all learning components in a specific area of quantitative reasoning constitutes a major developmental effort. The challenge however is not one of complexity as illustrated in the sample knowledge base included earlier, but is one of volume. Developing a database of questions for each learning component will also be challenging. There are a number of test databases available; many of which accompany the more popular textbooks. These databases often come with three levels of difficulty. Many of them can be refined and incorporated in the proposed software design
The ALS described herein requires a marriage between decision support scientists and educators. In this marriage, neither role should be discounted. The quality of presentations and questions in each learning component is as critical as the development of rules. In prototyping, as much attention needs to be given to testing the learning and testing material as to the functionality of the software. During the prototyping process, particular attention needs to be given to the effectiveness of each presentation, as well as to the clarity of test questions. Unclear questions can lead to incorrect diagnosis which would defeat the purpose of the software entirely. If presentations are not clear and understandable, they would only add to the frustrations of the student and are counterproductive.
The proposed ALS is being implemented on several different delivery platforms including a stand alone PC version and an Internet version. The Internet in itself is another example of using information technology for improving education. This mode of delivery offers many benefits including ongoing database development and multi-user contributions. Additionally, it provides an effective medium for constructive feedback from both educators and student users.
The platform selected to provide the initial ALS Internet application is called Studynet. This system operates an Internet site at . Studynet represents the state-of-the Art in virtual teaching, training and education. It supports audio, video and a wide range of supplemental material, e.g., Harvard Case Study articles and reports. Instructors can create lesson and study plans that drives collaborative, asynchronous or synchronous real time learning. This, of course, is an important key to providing effective distance learning education. Used in this way, an Internet based ALS offers considerable promise for meeting the business education and management training challenges of the 21st Century.
APPENDIX A - SELECTED COMMERCIALLY AVAILABLE EXPERT SYSTEM SHELLS
|Vendor |Product |Price |Web Site/E-mail/Phone |
| | | | |
|Acquired Intelligence |ACQUIRE |$995 |http//ai/ |
|Arity Corp. |ARITY EXPERT | |73677.2614@ |
| |OS/2 |$495 |pgweiss@ |
| |DOS |$295 | |
|EXSYS Corp. |EXSYS RuleBook |$1,495 |WWW. |
| |EXSYS Professional |$2,900 | |
| |EXSYS Linkable Module |$5,000 | |
|Information Builders |LEVEL 5 OBJECT Professional |N/A |WWW. |
|Gold Hill Computers Inc. |GOLDWORKS III |N/A |WWW.goldhill- |
|Logic Programming Assoc. Ltd. |FLEX |$1,000 |WWW.lpa.co.uk |
|Template Software |KES | |703-318-1000 |
| |P.C. |$ 4000 | |
| |Workstation |$10,000 | |
| |Mini |$25,000 | |
| |Mainframe |$60,000 | |
| |SNAP (Mini) |$40,000 | |
|KDS Corp. |KDS | |708-251-2621 |
| |VOX |$15,000 | |
| |KDS 3+ |$1,795 | |
|ILOG Inc. |ILOG RULES |N/A |WWW.ilog.fr |
|Micro Data Base |GURU |N/A |info@ |
|Wordtech |VP Expert |$129 |WWW. |
| | | |510-689-1200 |
|The Haley Enterprise |The Easy Resoner | |WWW. |
| |16 bit Windows |$249 | |
| |Tool Kit | | |
| |32 bit Windows |$499 | |
| |Tool Kit | | |
| |OS/2 Tool Kit |$499 | |
| |Unix/Motif Tool Kit |$999 | |
|Procedural Reasoning |C-PRS |N/A |ingrand@ingenia.fr |
| | | |cprs@ingenia.fr |
|Comdale Technologies |Comdale /C |N/A | |
| |Comdale /X | | |
| |Process Vision | | |
APPENDIX B - EXAMPLE RULE BASED SPREAD SHEET
[pic]
REFERENCES
Alavi, M., Yoo, Y. & Vogel, D.R. (1997). “Using Information Technology to
Add Value to Management Education,” Academy of Management
Journal,40(8), 1310-1333.
Badiru, A.B. (1992). “Expert System Applications in Engineering and
Manufacturing,” Englewood Cliffs, NJ, Prentice Hall.
Badiru, A.B. (1990). “Systems Implementation for Total Quality Management,” Engineering Management Journal, 2(3), 23-28.
Badiru, A.B. et al., (1988). “AREST: Armed Robbery Eidetic Suspect Typing
Expert System,” Journal of Police Science and Administration, 16(3),
September, 210-216.
Bloor, R. (1993). “New Intelligence: Expert Systems are reemerging to satisfy
broader needs than their predecessors; Mission Critical View,” DBMS
6(13), 12.
Curry, B. & Moutinho, L. (1994). “Intelligent Computer Models for Marketing
Decisions.” Management Decision, 32(4), 32-35.
Ganascia, J.G. (1994). “Using an Expert System in Merging Qualitative and
Quantitative Data Analysis,” International Journal of Man-Machine
Studies, 27(3), 453-470.
Holsapple C.W.(1987). Business Expert Systems, Irwin.
Kaula, R. & Lander, L.C. (1995). “A module-based conceptual framework for
large-scale expert systems,” Industrial Management and Data Systems,
95(2), 15-23.
Kerka, S. (1992). “Higher Order Thinking in Vocational Education,”
Clearinghouse on Adult, Career, and Vocational Education, Eric
Digest No. 127, Columbus, OH.
Lawson, G. (1997). “New Paradigms in Adult Learning,” Adult Learning,
8(3), 10..
Luce, T. (1992). Using V.P. Expert., New York, Mitchell McGraw-Hill.
Newstrom, J.W. (1991). “One Size Does Not Fit All,” Journal of Training &
Development, 43-48.
Office of Educational Research, (1991). What is an Expert System?,
Washington, D.C.
Parsaye, K., & Chingnell, M. (1988). Expert Systems for Experts, John Wiley
& Sons.
Pickett, J.R. & Case, T.L (1991). “Implementing Expert Systems in R&D,”
Research Technology Management, July-August, 37-42.
Slater, J.R. (1993). “On Selecting Appropriate Technology for Knowledge
Systems,” Journal of Systems Management, 10-16.
Valentine, J. (1988). “Applying Expert Systems to Investment,” Financial Analyst Journal, November/December, 48-53..
Webster, J. & Hackley, P. (1997). “Teaching Effectiveness in Technology-
Mediated Distance Learning,” Academy of Management Journal, 40(8), 1282-1309.
Willetts, K. (1992). Technology and Second Language Learning, Office of Educational Research and Improvement, Washington, D.C.
PROVIDING VALUE TO THE USERS OF
INFORMATION SYSTEMS:
A THEORY OF THE IS-USER RELATIONSHIP
DEVELOPMENT
Lara Preiser-Houy*
Carole E. Agres**
The computerization of organizations has become increasingly important since the introduction of first computers in the late 1950s. The rapidly changing business environment and technological advances of the 1990s are triggering major transformation in the way companies organize their work and conduct their business. Information technologies (IT) and information systems specialists are critical enablers of this transformation. Successful provision of an IT-related service depends on the partnership between IS professionals and the user community. Providing value to the users is a means to an end in building these critical IS - user partnerships.
The purpose of this paper is to present a theory of how IS specialists provide value to their users through the development of IS-user relationships and the consequences of these value-adding activities to the organization. The theory involves three core processes which have been discovered to be relevant to the way IS professionals provide value. These processes are-inciting, intervening, and informating. Furthermore, the theory specifies the stages through which IS specialists provide value to their business clients. These stages are-building, maintaining or destroying the IS-user relationships. In this paper we describe the conditions under which value-adding activities take place, strategies by which IS specialists engage in value-adding activities, and the consequences of these activities to the organization.
INTRODUCTION
Background
C
hange seems to be the only constant in the IS field. The cost/performance improvements in the core information technologies (IT)--between 30 percent and 50 percent per year since the 1960s--have indeed been dramatic (Tapscott and Caston, 1994). As organizations progressed through periods characterized by computing hardware, software, and user relations’ constraints (Friedman, 1989), they employed the expertise and talent of many IS specialists to conceive, design, and diffuse technological solutions to business problems. However, as IT advanced and the number of IS specialists grew, so did the problems associated with IT implementations (Yourdon, 1992). Since the effective management of information flows inside and outside organizational boundaries is critical to the health and effectiveness of many organizations, IT implementation problems may negatively impact organizational performance. Recognizing the influence IS professionals can have on organizations, Orlikowski and Baroudi (1989) have called for increased research on IS as an occupation and the role of IS workers including their world views and behaviors.
As we approach the end of this century, literature suggests the Information Systems (IS) profession is plagued by a myriad concerns that IS specialists are not providing adequate value to the organization (Cusack, 1993; Markus and Keil, 1994; Markus and Robey, 1994; Walton, 1989). The value-added aspect of the IS service provision continues to be challenged by productivity paradox and outsourcing phenomena. For example, massive investments in information technologies have not resulted in sufficient productivity improvements (Brynjolfsson, 1993). Thus, many executives are faced with a formidable task of justifying the ever-increasing expenditures for the IT infrastructure and the IS human resources (Preiser-Houy, 1996). Outsourcing is yet another phenomenon foreshadowing IS.
Outsourcing is viewed by many business executives as a potential value-adding strategy. In fact, the results of a recent empirical study of IS outsourcing benefits indicate that both technical (i.e., technical expertise), business (i.e., lower costs of operation), and non-technical (i.e., better quality of IS service) benefits were significant motivators of outsourcing decisions (Loh and Venkatraman, 1995). The outsourcing business has grown rapidly in the past years and is expected to be a record high forty billion dollars by the end of 1990s (Burkett, 1993). This trend may eventually spell danger to the IS profession as users outsource and rightsize their IS organization, thus putting IS professionals out of work.
In the past thirty years, computer systems’ development practices have progressed through three distinct phases, each dominated by a certain constraint: hardware constraint-i.e., limitations of hardware cost and reliability, software constraint -- i.e., difficulties of producing quality systems on time and within budget), and finally, user relations constraint-i.e., inadequate servicing of user needs (Friedman, 1989). While hardware and software limitations are not as prevalent today as they were a decade ago, the dominant constraint of the 1990s is that of user relations. Providing value-added service through effective IS-user relationships is one of the ways IS specialists can successfully address the dominant user relations constraint, and thus, ensure the long-term survival of the IS profession.
Purpose of the paper
In this paper we present a theory of how IS specialists provide value to their users and the consequences of these value-inducing activities to the organization. The theory is comprised of three core value-adding processes: (1) inciting, i.e. instigating, being proactive, etc., (2) intervening, i.e. acting as a go between various stakeholders including users, management, and vendors, and (3) informating, i.e. educating or helping to 'digest' technology. This paper describes the conditions, strategies, and consequences relevant to the three value-adding processes of IS service provisions.
Methodology
This study was conducted using a grounded theory method. Grounded theory is an inductive, theory building methodology that enables researchers to develop theoretical accounts of the phenomenon under investigation by grounding these accounts in empirical data (Glazer and Strauss, 1967). The key aspect of the chosen method is the analytic activity of constant comparison. As data about the IS value-added processes was gathered, it was constantly fragmented, coded and compared to existing theoretical categories. Results of these comparisons were constantly fed back to both the analysis and the data gathering phases of the study.
The data collection strategy involved multiple sources of evidence including personal interviews with twenty IS specialists and their clients, follow up telephone interviews, and the review of company and IS project documentation. The focus of the semi-structured personal interviews with the IS specialists and their clients was on the tactics employed by IS specialists to provide the information systems-related services during IS implementation projects. The IS projects involved new system development efforts, post-production support and routine maintenance activities. IS specialists interviewed for this study had the following demographics: (1) two to eight years of IS experience, (2) 22 to 34 years of age, (3) 80% had college degree in IS or Business, and, (4) worked in banking, retail and communications industries.
THEORY
The theory of how IS specialists provide value to their business clients is summarized in Table 1. This theory is comprised of the key value-adding processes (i.e. inciting, intervening and informating), and the stages through which IS specialists progress in providing value to their clients (i.e., building relationships, maintaining relationships, and destroying relationships). Next, each of theses core value-adding processes will be described.
Table 1. Theory of IS value-adding activities: Stages and Processes
|Stages / Processes |BUILDING |MAINTAINING |DESTROYING |
| |RELATIONSHIPS |RELATIONSHIPS |RELATIONSHIPS |
| | | | |
|INCITING |Instigate users to “push” |Act as liaisons between |Identify ways to change |
| |for new technologies |users and IS management |users’ work without |
| | | |understanding political |
| | | |environment |
| | | | |
|INTERVENING |Bridge the “culture gaps” |Change roles depending on |Enhance authority and |
| | |what is needed at the time |status of IS function at |
| | | |the expense of the user |
| | | |group |
| | | | |
|INFORMATING |Educate users on how IT can|Continue education until |Provide no training for |
| |improve their jobs |users are self-sufficient |users |
Inciting Process
Webster' s Dictionary (1988) defines the word inciting as stimulating to action, either in a favorable or unfavorable sense. IS specialists stimulate users to action by identifying ways to change the workflow so that the company runs better and by suggesting information technologies to enable the new world of work. As one IS specialist commented, “I think it is part of my job to help people change the way they are thinking about manufacturing. I consider myself to be an educator. Part of what I do is explaining how important it is to be open to new ways to do things.” Other IS specialists used their social skills, combined with technical expertise, to instigate users to push for new technologies from their management when the users' management was unresponsive to the IS specialists' recommendations for the new system.
Many IS specialists recognize that the effectiveness of IT lies in enabling a more efficient and streamlined workflow. They view the activity of automating the currently inefficient business process as a 'Band-Aid' approach to solving a business problem, an approach that ultimately leads to a 'more expensive mess'. Furthermore, they understand that users may be resistant to change the ways they view and perform their work. According to one IS specialist, “It is a culture shock to reverse the current thinking about the process”. Yet, IS specialists also realize that in order to reap the benefits from IT and to use systems effectively, users may have to change the way they operate and perform their business activities.
While all IS specialists recognize the difficulty associated with proactively instigating users to rethink their processes, some of them are more successful than others at convincing users to change their old ways of doing business and to adopt various information technologies to enable the new world of work. Among the successful strategies are: (I) setting up training classes for users to educate them about IT and its potential to change the organization, (2) letting ideas about change emerge from the group discussions with users, IS, and the management, and, (3) viewing implementation more as a social rather than a predominantly technical process. Unsuccessful strategies to stimulate users to change include: (1) telling users what to do, (2) perceiving technology as an end, rather than the means to an end, (3) underestimating clients' resistance to change, (4) 'fighting fires' instead of finding ways to make IT improvements in the long run, (5) not framing the changes in the context of the political and cultural environment, and, (6) lacking solid functional knowledge.
The key consequence of successful instigative practice is that users are more willing to accept change. This change, in turn, may lead to substantial improvements in one or more contemporary measures of performance -- cycle time, cost, quality, and customer service. The consequences of the negative instigative practices range from deterioration in the IS-user working relationships to systems that fail to improve the organization. Once users start viewing the IS specialists as technicians, systems builders, and adversaries rather than agents of change, collaborators, and business partners, they become defensive towards IS professionals, and may fail to cooperate and contribute to the subsequent systems development efforts.
Intervening Process
Another way IS specialists engage in value-adding activities is by acting as the liaisons between various stakeholders in the systems development effort. The stakeholders are those individuals who make contributions to the development effort (e.g., allocate resources, ensure funding, provide requirements, offer specialized technical expertise, etc.) in return for certain inducements (e.g., money, product deliverables, etc.). This role could be adopted by users, management, external consultants, and internal IS specialists.
Many good high-tech ideas fail due to low-tech problems--inadequate perception of stakeholders' demands and/or inadequate servicing of their needs. Yet, technology that is never productionalized or successfully used in the aftermath of the implementation effort adds very little, if anything, to the financial bottom line of the company.
Some IS specialists view IT implementation as a complex technical as well as social process. Such views help them be more proactive in dealing not only with technical challenges of the IT project, but with potential conflicts and user resistance to new information systems. According to several IS professionals, . . .“when implementing technology you are not totally dealing with the artifacts, you are touching values all the time. . . you must portray an image that people trust, otherwise there is a good chance you will fail. . . ”
Any one of the potential stakeholders in IT implementation can make or break the implementation effort. Users may be unwilling to change, be generally afraid of technology, or aim to optimize their own workflow at the expense of other functional units. Managers may be concerned with the loss of power and control in the aftermath of technological diffusion since IT may change the organizational structure, the locus of control, the source of power, and the process of decision making in the organization. External consultants and other IS professionals may also have their own hidden agendas that effect their contribution to the implementation effort.
By viewing themselves as collaborators and facilitators of the systems implementation effort, IS specialists aim to bring the technical, business and social cultures together in order to accomplish a common goal. The ability to bridge the culture gap between different stakeholders and functional areas is an important relationship-building skill. While the stakeholders may share the same goal of implementing IT to make the company run better, they may have different means to reach that goal. Thus, one of the ways IS specialists provide value to their clients is by building consensus among functional areas with different stakes in the IT implementation effort.
IS specialists employ a number of different strategies to bridge diverse and at times conflicting interests during the process of implementing IT solutions to business problems. IS specialists set up the brainstorming sessions to build consensus among various users on what needs to be done and how. They talk the language that merges business and technology with a goal of user-friendly and productive interchange.
Another intervening tactic used by IS specialists is to develop/maintain personal relationship with their business clients and to open themselves to criticism as well as praise. IS specialists may admit to their users that they do not have all the answers, but they are willing to do their best to find the answers by enlisting the help of others. Furthermore, IS specialists try to anticipate potential conflicts and the sources of various social problems and work on addressing these problems before the implementation effort gets underway. Another ploy is to work hard at determining what makes different users 'tick' and present their ideas in such a way as to maximize the chances for their acceptance. System justifications are many times the responsibility of the IS function. Bias by the IS professionals in developing the cost-benefit analysis in order to 'sell' systems that are valuable to the organization but may not necessarily meet hard dollars and cents payback, was yet another strategy employed by IS specialists in this study.
The successful use of the above-stated strategies not only maximizes the chances for successful IT diffusion and subsequent IT use, but also improves the credibility of IS specialists in the eyes of their business clients. Users gravitate towards such individuals. They trust them and feel comfortable around them. Moreover, IS managers consistently rank these IS specialists as high performers.
Informating Process
The third way IS specialists engage in value-adding activities is by helping users develop competency and mastery in using information technology in their respective work settings. Users are only human. They tend to resist new information systems that are unfamiliar to them unless they get help overcoming their own fears of the unknown. Therefore, educating users about how the system can be used, as well as training them in the mechanics of that use, are critical to the success of the technology implementation process. This process of education and training starts and ends with making users feel comfortable with technology.
IS specialists use a number of different strategies and tactics to inform users about technology. Some IS specialists set up training sessions where they explain what IT is all about and how technology can potentially change the flow and content of users’ work. Other IS specialists go an extra mile to ensure that users actively participate in the systems development efforts and contribute to the emergent system design. Still other IS specialists work diligently on developing training guides and user manuals and organize one-on-one training sessions to make users feel more comfortable with IT.
The consequences of these strategies are many and varied. Informating users about the use of IT helps IS specialists build relationships with the user community. Users tend to view IS workers as helpful, trustworthy, and knowledgeable professionals. Users become more knowledgeable and, consequently, make increasingly valuable contributions to systems implementations. Users feel comfortable around IS specialists and make sure they are included in the subsequent IT implementation projects. Moreover, users are also likely to accept the new system much more readily if they are comfortable with it and have the knowledge of how it can be used to help them do their job better. But, probably the most significant consequence of informating tactics is a continuing business relationship and partnership between the IS and other functional groups.
On the other hand, prolonged 'hand-holding' may develop into increased reliance on IS specialists. Such reliance may hamper any subsequent efforts to make users self-sufficient. Consequently, from the organizational point of view, such intricate dependency may lead to sub-optimization of critical IS resources and dissatisfied users.
DISCUSSION
Some behaviors of IS specialists enable them to build and maintain effective and credible working relationships with users. Good working relationships between IS specialists and clients reduce chances of IS implementation failures (Preiser-Houy et al., 1997; Remenyi, 1996), enable IS specialists better manage client expectations and build systems clients actually use (Landeros et al., 1995; McConell, 1996), and work effectively with users to achieve common goals (Henderson, 1990). Such relationships, thus, add real value to the provision of IS service.
However, other behaviors of IS specialists may actually contribute to the deterioration of the IS-client working relationships. Such behaviors are counter-productive and add no value to the provision of IS service. Consequences of non-value adding behavior can be minor, such as a slip in schedule or increase in costs, or major, such as a failure of a complex system implementation effort. But, in either case, IS value-reducing consequences threaten the credibility of IS specialists and the IS profession. If the consequences of value-reducing behaviors are not good, why do some IS specialists behave the way they do?
Little research has addressed the behavioral aspects of the IS service provision and the way IS develop (or fail to develop) effective working relationships with their business clients. However, the findings from this exploratory research suggest that the cognitive style, personal characteristics, and social motivational needs of IS specialists are at the core of their relationship building behaviors.
For example, individuals with good interpersonal and social skills, collaborative attitude, and a sincere desire to work with people as well as technology are effective in developing good IS-client relationship, and thus, engage in value-adding activities. On the other hand, individuals with good technical, but poor interpersonal skills, low need for social interaction, and desire to work with technology rather than people, are not very effective in building good IS-client relationships. Consequently, since poor IS-client relationships may lead to failed IT projects, it is questionable whether IS specialists with relationship-destroying behaviors provide much value to their clients in the course of a service transaction. While more research is needed to better understand the interpersonal-level conditions under which IS specialists engage (or fail to engage) in value-adding activities in the context of relationship building, this exploratory research provides a lens through which the value-adding (and value-destroying) behaviors of IS specialists could be examined.
Inciting, intervening, and informating in the diametric form are value-reducing or alienating tactics; in the form present in this descriptive write-up, they become value-adding representations. The interviewees were generally adept at identifying their value-added behaviors and activities. In addition, their behaviors appear to fit a mental model that stresses the importance of good user relationships. Results of the interviews indicated a strong compulsion toward adopting the role of “collaborator” instead of a “pair-of-hands” approach for IS-user relationships. Collaborators apply their specialized skills to help clients diagnose their own problems (Schein, 1990). Furthermore, they negotiate issues of control and communicate well with clients (Block, 1981; Markus and Benjamin, 1995). On the other hand, IS specialists whose predominant role is that of a pair-of-hands adopt inactive role vis-à-vis the client and engage in a limited two-way communication with the clients. The collaborative approach to service provision seemingly fits the ideal model of the way things should be in order for IS specialists to add value to their organization.
In the course of this study, IS specialists consistently identified human resource dimensions (i.e., formal systems of expectations and rewards, organizational practices concerning the orientation, training and socialization of new employees, and career facilitation) as enablers (or barriers) to effective IS-user relationships. For example, when human resource policies of the organization promote service quality and encourage customer orientation through rewards, formal evaluations, career development, and socialization, IS specialists work well with their users and users report higher levels of satisfaction with the service they receive. But, human resources policies and procedures of the organization may provide barriers to effective IS-user relationships if they do not encourage, cultivate and promote collaborative, customer-focused behaviors on the part of their IS specialists.
Implications of these findings may establish a precedent for changing the way IS value is derived and determined. Value is in the eyes of the beholder, and as the productivity paradox and other theoretical frameworks portray problems in quantifying IT value, maybe it is the day-to-day behaviors of IS specialists and the contribution of these behaviors to the users that should be examined for value-adding measurements. There are some things that can not necessarily be measured in tangible dollars and cents or identified on the balance sheet bottom line. But if IS specialists take a proactive and perfunctionary role concentrating on value-adding behavior, the contributions of IS services to the organization will not be so vague and implicit. IS challenges, such as cost/schedule overruns and production problems, should be met with 'working smarter' not 'working harder' philosophy. Let the IT strategic planners continue trying to solve the seemingly impossible “big picture” problems of IS value. In the meantime, IS specialists can provide value-added services to their clients through behaviors that lead to good IS-client relationships. Such value-added provision of IS service will incrementally improve the concept of IS value and possibly change the way IS value is derived and measured!
REFERENCES
Block,P.(1981). Flawless Consulting: A Guide to Getting Your Expertise Used. San
Diego, CA: Pfeiffer & Co.
Brynjolfsson, E. (1993). “The Productivity Paradox of InformatioTechnology”. Communications of the ACM, 36(12), 41-55.
Burkett, G.T. (1993). “Outsourcing IT Services.” DATAPRO:Computing Systems Series - Overviews, Deltran, NJ: McGraw-Hill., 100-105.
Cusack, S. (1993). “The People Rollout: Key to Change.” Datamation, 39(7), 55-56.
Friedman, A. L. (1989). Computer Systems Development: History, Organization and Implementation, West Sussex, England: John Wiley & Sons.
Glaser, B. G and Strauss, A.L. (1967). The Discovery of Grounded Theory: Strategies
for Qualitative Research, New York, NY: Aldine De Gruyter.
Henderson, J. C. (1990). “Plugging into Strategic Partnerships: The Critical IS Connection.” Sloan Management Review, 31(3), 7 - 18.
Landeros, R., Reck, R., and Plank, R. E. (1995). “Maintaining Buyer-Supplier Partnerships.” International Journal of Purchasing & Materials, Management, 31(3), 3 - 11.
Loh, L. and Venkatraman, N. (1995). “An Empirical Study of Information Technology Outsourcing: Benefits, Risks, and Performance Implications.” In DeGross, J.I., Ariav, G., Beath, C., Hoyer, R. and Kemerer, C. (Eds.), Proceedings of the Sixteenth International Conference on Information Systems, Amsterdam, (December 11-15).
Markus, M. L. and Keil, M. (1994). “If We Build It They Will Come:Designing Information Systems That Users Want To Use.”Sloan Management Review, 35(3), 11-25.
Markus, M. L. and Benjamin, R. I. (1995). “Change Agentry -- The Next I.S. Frontier.” Working Paper WP 06-95. Claremont, CA: The Claremont Graduate School.
Markus, M. L. and Robey, D. (1995). “Business Process Reengineering and the Role of the Information Systems Professional”. In Grover, V. & Kettinger, W. (Eds.), Business Process Reengineering: A Strategic Approach Pp.569-589. Middletown, PA: Idea Group Publishing.
McConell, S. (1996). Rapid Development. Redmond, WA: Microsoft Press.
Orlikowski, W. J. and Baroudi J.J.(1991). “Studying Information Technology in Organizations: Research Approaches and Assumptions”, Information Systems Research, 2(1), 55-70.
Preiser-Houy, L. (1996). “Assessing the Payoff from an Information Technology Infrastructure: A Multi-Phased Approach.” Journal of Business and Management, 3(3), 53-79.
Preiser-Houy L.,Edberg D.T. and Agres, C. E. (1997). “MISImplementation: An Investigation of Success Factors.” In Proceedings of the Twenty-Sixth Annual Meeting of the Western Decision Sciences Institute Pp. 465-467,
March, Kamuela, Hawaii: The Decision Sciences Institute.Remenyi, D. (1996). “Ten Common Information Systems Mistakes.” Journal of Management, 21(4), 78-91.
Schein, E. H. (1990). “A General Philosophy of Helping: Process Consultation.” Sloan Management Review, 31(3), 57-64.
Tapscott, D. and Caston, A. (1993). Paradigm Shifts: The New Promise of Information Technology, New York, NY: McGraw-Hill, Inc.
Webster’ s New World Dictionary of American English. (1988). Neufeldt,V.and Guralnik, D. B. (Editors), 3rd College Edition, New York, NY: Simon & Schuster, Inc.
Walton, R. E. (1989). Up and Running: Integrating Information Technology and the Organization. Boston, MA: Harvard Business School Press.
Yourdon, E. (1992). Decline & Fall of the American Programmer, Englewood Cliffs: Prentice Hall.
CALL FOR PAPER
JOURNAL OF BUSINESS AND MANAGEMENT
The JOURNAL OF BUSINESS AND MANAGEMENT is soliciting papers for its upcoming issue. Its purpose is to provide a forum for the dissemination of fresh ideas and research in all areas of business, management and public policy which would be of interest to business persons, public officials and academics. The JOURNAL invites both empirical and theoretical articles. Reports on research and opinion pieces as well as original manuscripts are welcome. Book reviewers are requested to submit articles on recent publications in related topics. Only articles not previously published or currently under review elsewhere can be considered. All submissions are blind refereed. It should not be assumed that readers are completely familiar with the concepts and terminology of the specific subject under study. Directness and clarity of presentation are desired.
Manuscripts must be typed and double-spaced. Three copies should be submitted and become the property of the JOURNAL. The length of the paper should not exceed 20 pages, excluding notes, references, tables, figures and appendices. Notes and references should appear at the end of the article. Each chart, figure or table should be camera ready, that is, an original on a separate page with instructions indicating the placement in the article. A separate cover sheet indicating the title of the manuscript, as well as the name, affiliation, position and address of the author should accompany each submission. To facilitate blind review, the body of the text should include the title, but not the author's name. Include also an abstract of not more than 100 words summarizing the article.
The views expressed in published articles are those of the authors and not necessarily those of the Editors or the Editorial Review Board. Responsibility for the correctness of quotations and citations rests with the author.
The JOURNAL OF BUSINESS AND MANAGEMENT is a bi-annual publication of the Western Decision Sciences Institute and the School of Management, California State University, Dominguez Hills.
Submit Papers to:
Burhan F. Yavas, Editor
JOURNAL OF BUSINESS AND MANAGEMENT
School of Management
California State University, Dominguez Hills
Carson, CA 90747
Subscription to the
JOURNAL OF BUSINESS AND MANAGEMENT
The JOURNAL OF BUSINESS AND MANAGEMENT is published bi-annually by the Western Decision Sciences Institute and the School of Management, California State University Dominguez Hills.
The annual subscription fee is $20.00 for individuals, and $50.00 for institutions.
To receive the JOURNAL OF BUSINESS AND MANAGEMENT, please complete the form below, and make your check or money order payable to the CSUDH Foundation.
Burhan F. Yavas, Editor
School of Management
California State University Dominguez Hills
Carson, CA 90747
Name
Organization
Address
Please make your check or money order payable to the CSUDH Foundation.
* Diane H. Roberts is affiliated with the University of San Francisco
* Jacob M. Chacko is affiliated with the University of North Dakota
**Randy Larson is affiliated with the University of North Dakota
* Robert W. Stone is affiliated with the University of Idaho
**John W. Henry is affiliated with the Georgia Southern University
* Owen P. Hall, Jr. is affiliated with Pepperdine University.
**Farzin Madjidi is affiliated with Pepperdine University.
* Lara Preiser-Houy is affiliated with California State Polytechnic University, Pomona
** Carole E. Agres is affiliated with The Boeing Company
-----------------------
JOURNAL OF
BUSINESS AND MANAGEMENT
ACCOUNTING / FINANCE
MANAGEMENT/MARKETING
MANAGEMENT EDUCATION
QUANTITATIVE METHODS
MODE 2 - Video Presentation
MODE 1 - Review of Principles
MODE 3 - Simulation
User Interface
................
................
In order to avoid copyright disputes, this page is only a partial summary.
To fulfill the demand for quickly locating and searching documents.
It is intelligent file search solution for home and business.
Related searches
- getroman com reviews
- acurafinancialservices.com account management
- acurafinancialservices.com account ma
- getroman.com tv
- http cashier.95516.com bing
- http cashier.95516.com bingprivacy notice.pdf
- connected mcgraw hill com lausd
- education.com games play
- rushmorelm.com one time payment
- autotrader.com used cars
- b com 2nd year syllabus
- gmail.com sign in