THE FIELD (CALIFORNIA) POLL



THE FIELD (CALIFORNIA) POLL

Late November-early December 2013

Codebook 13-03

Data file citation:

Field (California) Poll Late November/early December 2013 [machine-readable data file] San Francisco, CA: Field Research Corporation, 2013, Field (California) Poll 13-03.

UNIVERSE: CALIFORNIA REGISTERED VOTERS

INTERVIEWING PERIOD: NOVEMBER 14-DECEMBER 5, 2013

METHOD OF INTERVIEW: TELEPHONE USING LIVE INTERVIEWERS

SAMPLING METHOD: REGISTRATION-BASED LISTS

NUMBER OF CASES: 1,002 REGISTERED VOTERS STATEWIDE

CODING OF NOT APPLICABLE: Form A or Form B or AS INDICATED

About the Field Poll

The Field Poll is an independent, media-sponsored and non-partisan survey of California public opinion founded in 1947 by Mervin Field. Throughout its long history, The Field Poll has conducted regularly scheduled surveys in California examining voter preferences in major statewide candidate and proposition election contests, assessing public opinion of their elected officials and major issues facing the state, measuring public reactions to political and social developments, and covering other topical news stories of general public interest.

The Field Poll is owned and operated by Field Research Corporation, with headquarters in San Francisco, California. The Field Poll receives continuing financial support from leading California newspapers and television stations, which purchase the rights of first release to Field Poll reports in their primary viewer or readership markets. The Poll also receives funding from the University of California and California State University systems, which comprise the Poll’s academic consortium, as well as from foundations, non-profit organizations, and others as part of its policy research sponsor program. Since 1958 each Field Poll data file has been released to through the University of California’s UC Data Program data archive, where it is accessible to the public.

Description of Sampling Procedures

Field Poll surveys conducted prior to 1979 were typically administered through in-person interviews conducted door-to-door across the state. Since 1979 virtually all Field Polls have been conducted by telephone. During the period 1979-2005 these telephone surveys employed a random digit dial (RDD) sampling methodology. Since 2006, each Field Poll survey has employed a registration-based sampling (RBS) methodology sampling voters at random from lists of the state’s registered voter population. The following is a description of the RBS procedures employed by the Poll.

Lists of citizens who are registered to vote in California are compiled and regularly updated by local County Registrars. Commercial sampling vendors aggregate and maintain data files of these voter lists, and them available to public polling firms like The Field Poll. The lists contain a wealth of information about each voter, including their name, address, county, city, zip code, census tract and other geographic identifiers, gender, date of birth (age), party registration, voting participation and other political descriptors. Telephone numbers are appended to the list from a variety of sources, including phone numbers included when voters initially register to vote, or when they update their voter registration information with their counties, as well as by cross referencing their name and address information against directory and other telephone matching services. Since an increasing number of voters list their cell phone as their main contact telephone number when registering to vote or when providing phone contact information to companies or other agencies, the lists now include a large and increasing number of cell phone listings, as well as listings of voters’ residential landline phones.

Once a voter’s name and telephone number have been randomly selected by Field as part of an RBS sample, interviewers hand dial each listing in an attempt to reach and interview that voter. Because the RBS sample frame is drawn from a list of individuals and not households, this eliminates the need to implement respondent selection procedures once a contact is made. The specified voter is simply asked for by name and the interview only proceeds once that person is reached.

An advantage of sampling voters from RBS listings is that all persons interviewed are by definition known to be registered voters. This is superior to alternative approaches that must rely on respondent testimony, which is often unreliably reported. Since each voting record also contains the voter’s age, gender, party registration, and city and county of residence, this information is used to establish post-stratification weighting parameters to align each sample to parameters of the registered voter population.

This also enables Field to overlay a stratified random sample design onto its surveys. For example, target sample quotas can be assigned by age group, gender or regional geography to ensure that the composition of voters interviewed in the survey are consistent with the characteristics of the larger survey population.

The address information contained in an RBS listing also lends itself well to efficiently augmenting samples of voters within specific sub-geographies of a defined population. This can include supplemental samples of voters living within a city, set of zip codes, or even a collection of census tracts. Similarly, specific subpopulations such as men or women or voters of a certain age category can be targeted.

In recent years Field has also successfully employed RBS listings to augment the number of interviews conducted among specific ethnic voter populations, such as Hispanic Americans, Chinese Americans, Vietnamese Americans and Korean Americans. This is done by cross-referencing the voter surnames found on the voter lists against common ethnic surnames within that ethnic population.

Programming the Questionnaire Onto CATI

Once a survey questionnaire has been finalized in English, it is translated into Spanish (and other languages as needed) by professional translators. Each questionnaire is then programmed onto our computer-assisted telephone interviewing (CATI) system by staff programmers.

CATI controls the telephone scripts read to individual respondents by displaying the appropriate questionnaire items and their valid response code alternatives one at a time in their proper sequence on computer screens at each interviewer's booth. The interviewer simply reads each question aloud to the respondent from the screen and enters the pre-coded answer category through the keyboard directly to a computer disk. All answers are then automatically stored in computer memory.

Online interviewing using CATI allows for greater consistency in interviewing by controlling skip patterns, branches, randomization of items in a battery, "refer backs," and other control features during the call. CATI also affords greater opportunities for internal control, since the development and programming of the questionnaires remain under the direct control and supervision of the project director. The program will reject ineligible codes entered by an interviewer to all pre-coded questions, which means that the data contained in the survey’s raw data file are virtually clean and error free.

Once programming of the English version has been completed and is fully checked and proofed, the individual translations of each non-English language question are simply cut and pasted into the CATI script. Because the underlying CATI program does not vary across languages, this method ensures that all interviewing scripts are set up and implemented in a comparable fashion across all languages. Our CATI system is also designed to enable interviewers to seamlessly switch between the English and non-English language questionnaires during the call, enabling bilingual interviewers to initially choose which language to use when making their call attempts with potential respondents.

Field’s CATI system is directly linked to our internal Sample Management System (SMS), which incorporates each telephone listing into the survey samples. As interviewing proceeds the SMS stores and uses the call history information of each prior telephone attempt to guide all subsequent call attempts, arranging for callbacks at appointed times. When appointments cannot be scheduled, the SMS distributes callbacks so they are made at different days and different times of day to maximize the chances of reaching a respondent and assigns in-language callbacks only to interviewers fluent in that language.

The SMS also produces daily reports of all interviewing outcomes. Field's project director and staff supervisors closely monitor these reports to permit direct oversight and management of the telephone sample throughout data collection. In addition, the final disposition codes resulting from the SMS are used to calculate response rates, cooperation rates, and other indicators of survey performance at the conclusion of the calling effort.

Data Collection Procedures

All telephone interviewing is conducted in-house from Field’s central location telephone interviewing call center located in San Diego, where full-time staff professionals on hand to supervise, monitor and evaluate the performance of each interviewer. Field’s interviewing facilities consist of sound-protected booths where interviewers are stationed to do the calling using state-of-the-art computer-assisted telephone interviewing.

Field has staffing capabilities for administering surveys in many languages other than English. While Field has for decades routinely conducts most of telephone surveys of the California voting population in both English and Spanish, in recent years an increasing number have also been conducted in multiple Asian languages and dialects, such as Cantonese, Mandarin, Korean, and Vietnamese.

All interviewers working on this study are carefully trained by Field’s full-time staff supervisors in all the nuances of the questionnaire administration and their work are monitored throughout the interviewing period to assure uniform practices. The following measures are implemented to assure high quality and uniform telephone interviewing practices:

1. Interviewers are required to complete an interviewer training course, which provided both general and specific interviewing instructions, refresher reviews and online monitoring of telephone interviewing. The training course includes an introduction to survey research, a description of interviewer roles and responsibilities, general interviewing techniques and record keeping, and confidentiality procedures.

2. At the conclusion of their training, interviewers conduct mock interviews, and Field’s staff supervisors evaluate their performance.

3. Before the start of data collection, interviewers also attend a briefing session where the calling and interviewing procedures are discussed in detail by Field’s Study Director and staff supervisors. This session provides interviewers an overview of the study objectives and includes a question-by-question review of all survey items. The session also reviews recommended best-practice approaches for dealing with different interviewing situations, documenting the results of contact attempts, scheduling and confidentiality requirements.

4. De-briefings and re-training sessions are held as necessary by Field’s interviewing supervisors to be sure that all interviewers are administering the survey consistently. From time to time interviewers are asked to meet together as a group to discuss their interviewing experiences on the project. When questions arise and clarifications need to be made, written instructions are also prepared and distributed to all interviewing staff.

5. Throughout the interviewing period error correction forms are available to all interviewers to note respondent changes to answers after the initial recording of a response is given during the interview.

The survey questionnaire is typically pre-tested among a small sample of eligible voters within the target population. The pre-test is designed to assess the survey’s general ease of administration, and to provide an assessment of its average length of administration. These pretest calls are frequently monitored by Field’s staff supervisors and the Study Director to ensure that the survey is being implemented properly and consistently across all interviewers. Following the pre-test, changes are made to the questionnaire, as appropriate, before the launching of full-scale data collection.

Field typical makes up to six attempts (an initial call plus five callbacks) to reach each eligible voter selected for inclusion into the sample. Callbacks are made at different times and on different days to increase the probability of finding voters available for the interview. Voters contacted on a landline or cell phone are given the opportunity to offer another contact phone, such as a work phone, or alternate home, cell or VoIP phone, from which to conduct the survey. Callbacks are made at specified dates and times to maximize convenience and cooperation rates.

Rigorous procedures are employed to minimize respondent refusals and maximize survey response and cooperation. When refusals are encountered interviewers are instructed to categorize them into one of two broad categories – "hard refusals" for those in which the voter is adamant about not being called back again, and "soft refusals" for all others. No further telephone attempts will be made to refusals categorized as a "hard refusal." Refusals categorized as soft will often be re-opened and called back after a fixed period of time has passed by a specially trained team of refusal conversion interviewers. These calls are made at different times of day and on different days of the week than the initial refusal. Interviewers also consult previous details about the prior refusals to provide them with information that might be useful in helping them to convert refusals.

Field routinely employs a number of quality control measures during data collection. These include the following:

1. Only qualified and professionally trained interviewers will be conducting the telephone interviews under the supervision of Field’s full-time staff supervisors from our central location call center in San Diego.

2. Under the direction of the Field’s Study Director, a core team of Field staff professionals will supervise all aspects of interviewing, including the real-time monitoring of interviews in progress.

3. A variety of task management goals will be used to monitor and review the quality of the data collection on an interviewer-by-interviewer basis. Throughout data collection, debriefing and retraining sessions will be held as necessary to ensure that all interviewers are meeting our own internal high standards of professionalism and quality.

4. From time to time, interviewers will be asked to meet together as a group to discuss their interviewing experiences on the project. When questions arise or clarifications need to be made about issues relating to the administration of specific survey questions, written instructions will be prepared and distributed to all interviewing staff to ensure uniform, standardized interviewing practices.

Data Processing Procedures

All data processing is carried out from Field’s in-house data processing facility in San Francisco. This allows for close supervision and control of all processing functions by the Study Director and other project team members. The following is a description of the procedures employed by Field to complete the data processing tasks.

1. Data File Preparation: All survey information is systematically formatted in preparation for data cleaning and processing. As interviews are completed, each respondent is assigned a unique and anonymous identifier number.  In addition, some information included on the RBS file, such as political party affiliation, voting history, age and gender are appended to each voter’s survey record.

2. Data Cleaning and File Checking: Because CATI itself provides for the direct data entry of responses by the interviewer and does not permit ineligible or invalid data entries, the data file resulting from all CATI interviewing is itself virtually error-free. Nevertheless, as data collection proceeds, the survey data are edited and coded using both manual and machine cleaning procedures to verify that the data are complete, accurately recorded and internally consistent. Manual data editing involves various types of tasks. This includes managing and resolving problem cases which could involve reviewing specific interviewer comments about an answer, the key entry of information recorded onto error correction sheets by interviewers during or immediately following the completion of an interview, or reviewing permissible "other specify" replies to questions that fall outside established pre-coded response alternatives. A specially designed cleaning program will further scrutinize each record for internally inconsistent information.

3. At the conclusion of processing, respondent telephone numbers and any other identifying information obtained either from the RBS record or from the voter during the course of the interview are purged from the data file. During data collection and processing, access to all project files, and especially the RBS files containing identifiable information about the voter, are carefully controlled, with access to data files restricted to only key personnel at Field through a system of passwords.

4. Once the data file had been determined to be clean and error-free, post-stratification weights are applied to align the survey results to known population estimates of the overall registered voter population. To calculate individual weights Field uses its SPSS/Quantum v5e Rim Weighting software. The weighting software is quite sophisticated and internally calculates individual weights from specified "target values" established by the Study Director. For example, in developing weights for voters by age and gender, the researcher simply specifies the proportion of all voters in the overall target population who are men age 18-34, the proportion who women age 18-34, and so on as the target weight values desired in the final weighted sample. This greatly simplifies both the construction and proofing of the weights, since the correct target values can be displayed and checked in each weighted table at the end of the weighting process.

Reporting

Reports of The Field Poll cover a wide range of political, social and economic topics. Continuing measures are made of voter support for leading political figures vying for major state and federal elected offices, job ratings of important political figures, and reactions to significant political events. Voter awareness, understanding and predispositions for major campaign issues and controversial ballot propositions are also tracked over time. Each Field Poll release consists of two to twelve pages of text and statistical data presented in press release format, plus a background fact sheet reporting the details of the survey, its sample size, exact question wording and other technical details from that particular poll. All new and previous reports since 1995 are available to the public through our online archives at fieldpollonline.

-----------------------

THE

FIELD

POLL

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

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

Google Online Preview   Download