Measuring College Students’ Perceptions and Attitudes toward ... - AECT

Measuring College Students' Perceptions and Attitudes toward AntiPlagiarism Detection Tools and Their Behaviors, Beliefs, and Moral

Judgment regarding Plagiarism

Aaron Kim and Jean-Marc Wise Florida State University

Abstract The focus of this study is on measuring the impact of using anti-plagiarism tools (APT's) on students' behavior and attitudes toward plagiarism. A survey instrument was developed and validated in order to collect self-reported data on perceptions/attitudes, behavior, beliefs, and moral judgment with respect to plagiarism and the use of APT's. The survey was deployed with an undergraduate section at a southeastern research I university, yielding a response rate of 72.3% (n=133). Results found significant correlations between APT perception and plagiarism behavior, moral judgment, and neutralization (i.e. justification of unethical behavior). Findings also confirmed previous research with respect to correlations between attitudes, moral judgment, and plagiarism behavior. Details on survey development and validity, as well as suggestions for future research are provided.

Literature Review It is fair to say that plagiarism may threaten the integrity of higher education in colleges and universities throughout the U.S. About one in two students have cheated on tests and 71% and 85% of college students who participated in a study reported plagiarism and cheating on assignments, respectively (Stephens et al, 2007). Although Stephens defines cheating and plagiarism rather widely, figures from other studies are also staggering. For example, 70% of 18,000 high school students admitted they had cheated; and 60% plagiarized (Villano, 2006). Even more alarmingly, most of the students reported in Villano's article do not consider cheating as wrong and they feel justified in their behavior. "They don't understand why they should be held to a higher standard" said McCabe, who has studied the field of academic integrity for decades, in an interview with Slobogin from CNN (Slobogin 2002). The report is consistent with the findings from Stephens' study in which the students who reported cheating and plagiarizing showed a lower sense of moral responsibility and a higher tendency of neutralization or justification of their behaviors than non-cheating students (Stephens 2001). Students' behaviors and moral sense have been further investigated by Lovett-Hooper et al. They looked at the relationship between academic dishonesty and norm/rule-violating behaviors (Lovett-Hooper et al., 2007). The study found a positive relationship between all three Academic Dishonesty subscales (self-dishonest, social falsifying, and plagiarism) and an imagined norm/ruleviolating future. In short, the study suggests academic dishonesty may lead to rule-violating behavior in the future. In 2002, the Association of American Colleges and Universities (AAC&U) published a report on higher education of the 21st century to suggest comprehensive principles and models that improve learning for all undergraduate students entitled "Greater Expectations: A New Vision for Learning as a Nation Goes to College" (AAC&U, 2002). It envisions what society expects from college education and identifies three characteristics of learners as being empowered, informed, and responsible. Among dishonest behaviors, plagiarism has become a focal point recently as information technology advances and the use of the Internet is commonplace. In surveys run from 2002 through 2005, 38% of 18,000 college-aged students said they had `cut-and-pasted' from the Internet, 25% of graduate students also reported engaging in plagiarism, and 80% faculty observed plagiarism (McCabe, 2005). In an effort to fight plagiarism, many institutions put in place anti- plagiarism tools (APT) such as an online plagiarism detection service. There exist several commercial services including EVE2 (Easy Verification Engine: . eve/), Glatt Plagiarism Services (. ), SafeAssignment ( ), and TurnItIn (htpp://). These tools are generally well received by students because of the convenience of online submission, and by instructors as an additional tool to address the plagiarism issue. But only few studies have evaluated such tools systematically. In a 2006 study, TurnItIn was evaluated in terms of how the system works, user experiences, and data generated by the system (Evans, 2006). The author suggestes that the service detects problematic practices but it may not be practical to check every script thoroughly. The verdict was `working' but `impractical.' The growing popularity of these tools is likely due to a variety of reasons. For one, there seems to be a general agreement among instructors that awareness of such a tool can be a deterring factor; students do not risk getting caught if such a tool is in place. As shown in a recent study by Dahl (Dahl, 2007) one of the main advantages of

256

APT's may be their psychological effect. Awareness of TurnItIn among the participants in the study increased awareness of academic integrity and plagiarism (e.g., what is considered plagiarism, honor code, etc). The awareness also clearly brought about some behavioral changes: More than half of the participants learned how to reference correctly after using TunrItIn, which tells us that such a tool can also have educational benefits for higher academic standard (Dahl, 2007).

Therefore, it is important to examine how using APTs can affect students' behaviors, attitudes, and perceptions with respect to ethical conduct. What psychological effects and educational benefits of using APT's can bring is an important question. Thus, in this current study we investigate the following questions:

1. What are college students' awareness, perceptions and attitudes, and beliefs with respect to anti-plagiarism tools?

2. What is the relationship between perceptions/attitudes toward plagiarism, plagiarism behavior, moral judgment and beliefs, and perceptions/attitudes toward anti-plagiarism tools?

3. What are the effects of using anti-plagiarism tools? 4. What is the relationship between effects of using APT's and other variables of interest?

To answer the research questions, a self-report questionnaire was developed and used to collect data. Correlational analysis was the main method used to examine the relationships among variables based on the students ratings on the corresponding statements. The questionnaire included statements related to specific effects and benefits of APT's suggested in the literature as an attempt to address the questions. The statements include `After I learned my instructor was using an APT, I still plagiarized' and `I decided not to plagiarize' (deterrence effect); `I became more aware of plagiarism' (awareness effect); `I tried to find out how to avoid committing plagiarism,' `I tried to learn how I should reference something correctly because I experienced an APT' and `I feel confident about quoting other people's work correctly' (educational benefit). Other variables' operational definitions are as follows.

Variables of interest Plagiarism awareness is the level of students' awareness or knowledge about what constitutes plagiarism.

Plagiarism behavior represents how many times a student engaged in plagiarism. Perception/attitude on plagiarism includes three subsets: moral judgment, belief (neutralization), and plagiarism perception. In a 1991 study, moral judgment was shown to have a strong relationship with college students' dishonest behavior (i.e., cheating on exam) and suggested a strong predictor of such behaviors (Beck & Ajzen, 1991). The study suggested that students with high moral judgment level are less likely to engage in dishonest acts. In another study, neutralization was also shown as a strong predictor of academic dishonesty (Diekoff et al., 1996). Neutralization means students' tendency to justify dishonest behaviors attributing external and internal factors. Thus, the study suggested that students with high neutralization level are more likely to engage in dishonest acts. Perception/attitude of APT's is about how students feel about APT's, whether they support the use of APT's, feel intimidated, see APT's as a reliable tool, and so forth.

Method Participants

Participants were 184 students enrolled in a non-major course in southeastern public university. The researchers decided to use one course that is representative of the university population and work with the instructor of the course to ensure reasonable response rate. The instructor of the course gave extra credit to the students who completed the survey. The response rate was 72.3% (133 students participated). Since the survey should be anonymous, the researcher assured the students that the fact of their participation would be stored separately from the survey data and that their responses would not be associated with their identity.

The respondents consisted of 6 freshmen (4.5%), 55 sophomores (41.4%), 56 juniors (42.1%), and 15 seniors (11.3%). 64% of the respondents were female. The majority of the students major in Communications Studies (42%) and other majors included English, Marketing, Media Production, Business, Public Relations, Political Science, and International Affairs.

Procedures The first announcement regarding the study and the procedure was made during the class in September 2009

and was followed by an email advertisement and announcement on the course website. All students enrolled in the course were emailed with a link to the online survey along with a brief summary about the study. Anonymity of

257

participation was emphasized in all announcements and individual inquiries regarding anonymity and procedures were answered during the class visit and via emails by the researchers and the instructor. Once the students opened the questionnaire, they needed to provide their informed consent in order to proceed.

The online questionnaire was available for one week. In addition to the announcements in class and on the course site, and the initial email invitation, a reminder email was sent out on the third day and another one on fifth day of the data collection.

Measures A survey instrument was developed for this study, which included newly developed items and other items

that were adapted from the existing instruments. It is a self-report questionnaire that consists of forty five statements, most of which are rated using 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree), and eleven statements with slightly different scales. Please see Appendix A for a copy of the instrument. New statements were written in order to measure the variables of interest: the students' awareness, perceptions and attitudes, beliefs, and morality with respect to plagiarism and APT's. Subscales of moral judgment and beliefs were adapted from a moral responsibility scale (Beck & Ajzen, 1991) and a neutralization scale (Diekoff et al., 1996). Students' academic dishonesty was measured by items adapted from the Academic Dishonesty Student Survey (McCabe, 1992).

Structure of the measurement instrument The measurement instrument consists of 5 sections measuring the variables of interest in addition to APT

experience (section 4) and the demographic information (section 7). The following Table 1 summarizes the structure of the instrument with the different sections and associated variables.

Section Variables 1. Plagiarism Awareness

2. Plagiarism Behavior

3. Perception/Attitude toward Plagiarism

4. Anti-Plagiarism Tool (APT) Experience

5. Perception/attitude of AntiPlagiarism Tool (APT)

6. Effects of APT

7. Demographic Information

Table 1. Structure of the Measurement Instrument

Subscale (if any)

Scale

Disagree, Not sure, and

Agree.

Never, 0, to More than three

times, 5

- Moral judgment

5-point Likert scale ranging

- Neutralization

from 1 (strongly disagree) to

- Plagiarism perception 5 (strongly agree).

- APT Awareness

- Yes, No

- Experience

- Number of courses and

instructors

5-point Likert scale ranging

from 1 (strongly disagree) to

- Educational Benefits 5 (strongly agree) with N/A

option

Variable type Interval (Pretest total score) Interval

Interval

- Categorical - Interval

Interval

Interval

Categorical

Section 2 uses a modified version of McCabe's Academic Dishonesty Student Survey (1992). Most statements in the section 3 were developed in accordance with items in sections 5 and 6. Section 3, perception/attitude on plagiarism, contains subscales of moral judgment and beliefs that were adapted from a moral responsibility scale (Beck and Ajzen, 1991) and neutralization scale (Diekoff et al. 1996).

Development of the measurement instrument Newly developed items were drawn from the literature and tested and revised through a panel review, an

expert review, and formative evaluation. The instrument with the new items was reviewed by twelve doctoral students who are experienced in survey research and instrument development. The revised instrument then was reviewed by an expert who has been conducting and teaching survey research methods at a Research I University for more than 20 years.

Formative evaluation involved four representative students with different majors and years in school. During the formative evaluation session students engaged in open discussion and critique of the survey instrument (e.g., clarity of statements, overall structure, effective communication, etc). The instrument was once again revised based on the result of the formative evaluation.

258

Validity and reliability test Formative evaluation provided opportunities for assessing not only the quality of items but also content

validity. Since there is no external criterion that can be compared to in terms of plagiarism behavior, perception, and morality, criterion validation is beyond the scope of this study. Establishing construct validity, however, may be possible for some items that were modified from the existing items. The authors made a judgment based on how the results of this study relate to the previous studies that used the same instrument. For example, the instrument measuring moral judgment and beliefs revealed that plagiarism is highly correlated with moral judgment and beliefs of students (Beck & Ajzen, 1991). Thus, construct validity of the modified items of moral judgment and belief subscales can be assessed in addition to the content validity.

The reliability estimates for each section and subscales are summarized in Table 2.

Section Variables

Table 2. Reliability Estimates N of Items

Cronbach's Alpha

1. Plagiarism Awareness

6

N/A ? a test score

2. Plagiarism Behavior

9

.763

3. Perception/attitude on Plagiarism

11

.693

a. - Moral judgment

(4)*

(.660)

b. - Neutralization

(2)

(.440)

c. - Plagiarism perception

(5)

(.737)

4. Anti-Plagiarism Tool (APT) Experiences

5

N/A**

5. Perception/attitude of Anti-Plagiarism

13

.701

Tool (APT)

6. Effects of APT

7

.727

7. Demographic Information

6

N/A

*subscales in parenthesis

**2 questions pertains to the number of courses and instructors the students counted as experiences with APT and

plagiarism discussion.

Results College students' awareness, perceptions and attitudes, and beliefs with respect to anti-plagiarism tools

The awareness level was measured by six questions each asking whether a specific act was a type of plagiarism. The majority of the students scored 3 or less, which may indicate the lack of knowledge about plagiarism among the students. Moral judgment, plagiarism perception, neutralization, and APT perception are all quite high. APT effects showed high agreement that their experiences with APT affected their behaviors and perception. The following descriptive statistics table summarizes the results.

Table 3. Descriptive Statistics

Plagiarism Awareness Moral Judgment

N 124 123

Range

Min

Max

.00

6.00

6.00 15.00

Mean

Std. Deviation

2.137

.990

11.894

2.003

Skewness Statistic Std. Error

.383

.217

-.350

.218

Plagiarism Perception Neutralization

123

9.00

25.00

20.667

2.748

-.750

.218

122 2.00

8.00

3.943

1.501

.412

.219

APT Perception

124 24.00

59.00

40.742

7.204

-.042

.217

Data Preparation for statistical analysis Preliminary data examination showed that there were some missing cases and undifferentiated responses.

Further examination of such cases convinced the authors to drop the undifferentiated cases from the analysis. A total of 10 cases was removed from the data set because they either showed no variance in their answers to one or more sections of the questionnaire or provided answers that did not match when they were supposed to match such as reverse coded items.

In order to address the issue of missing data, we calculated the percent score for some variables instead of raw score using only answered items. Since there were only a few missing values and the missing values should not affect the overall score, a raw score divided by the maximum possible score would best represent the measures.

259

Perception/attitude on plagiarism (moral judgment, belief (neutralization), APT effects, and plagiarism perception used percentage scores while plagiarism awareness and behavior used raw scores.

Correlation Analysis This study examined the correlations among moral judgment, beliefs (in the neutralization subscale),

plagiarism behaviors, and attitudes/perceptions toward anti-plagiarism tools (APT's). The following correlational analysis table summarizes the results of this study. It shows relationships among variables and the effects of APT use. Educational benefit is a subscale under APT effects that included items such as `I became more aware of plagiarism'; `I tried to find out how to avoid committing plagiarism'; `I would prefer to get more guidance about what plagiarism is'; `I tried to learn how I should reference something correctly because I experienced an APT.'

Table 3. Correlations

Awrns

Plagiaris m

Moral Judgment

Plagiarism Perception

Neutralizatio n

APT Perception

Awareness

1

.036

.177

-.007

.011

.058

Plagiarism Behavior

1 -.339(**)

-.026

.305(**)

-.310(**)

Moral Judgment

1

.209(*) -.429(**)

.307(**)

Plagiarism Perception

1

-.206(*)

.158

Neutralization

1

-.382(**)

APT Perception

1

APT Effects

Ed. benefit

Experience

** Correlation is significant at the 0.01 level (2-tailed). * Correlation is significant at the 0.05 level (2-tailed).

Moral judgment and perception of APT are negatively correlated with plagiarism behavior while neutralization is positively correlated. Plagiarism behaviors represent the frequency of dishonest behaviors that the students reported. Thus, the study confirmed that the higher their moral judgment, the less likely it is for the students to involve in plagiarism activities. Likewise, the more the students have a positive perception of APT's, the less likely to plagiarize. On the other hand, the more students tend to justify their dishonest behavior, the more likely they are to plagiarize.

Perception of APT correlates positively with moral judgment and correlates negatively with neutralization, which could mean that those students who perceive plagiarism negatively tend to avoid plagiarism and perceive APT's positively. APT perception is also positively correlated with APT effects. In other words, those who see APT's positively tend to attribute their specific perception and behavioral changes to their experience with APT's, leading to a higher plagiarism awareness level or a decision not to plagiarize.

APT effects

Based on the students' responses, the majority of the students agreed that APT use affected their behavior

and awareness with respect to plagiarism. Table 4 summarizes the students' responses to the APT effects statements

and Table 5 summarizes the correlation between APT effects and other variables.

Table 4. APT Effects

Strongly

Strongly

N/A disagree Disagree Neutral Agree agree Total

1

25

64

28

6

123

20.3%

52.0%

22.8%

4.9%

2

6

5

11

19

68

12 123

4.9%

4.1%

8.9%

15.4% 55.3%

9.8%

3

30

5

5

23

43

16 123

24.4%

4.1%

4.1% 18.7% 35.0% 13.0%

4

15

11

10

29

50

8 123

260

 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download