Work and Leisure Today 2 Survey, United States, 2015 Archival Version


The Work and Leisure Today 2 survey is a dual frame random digit dial survey of landline and cell phone households, collected in Summer 2015. Adults in landline households were selected within the household following the Rizzo, Brick and Park selection method, with the next birthday method used to select adults in households with 3+ adults. Cell phone respondents were selected as the telephone answerer. The survey took about 15 minutes to complete. The survey contains questions about the sampled adult's employment, attitudes toward their job, leisure activities, and technology use. See the questionnaire for the question wording. Two versions of the questionnaire were fielded with experimental variation on many items for question wording and other contextual information. Each interview was audio-recorded, transcribed, and behavior coded at the conversational turn level. As such, in addition to the traditional survey data, the datasets include these behavior coded data, call record data, demographic characteristics about the interviewers, paradata files, and acoustic measurements of the interviewer's question reading for a subset of questions. Demographic variables include gender, race, income, and educational level.

This is a methodological study on computerized data collection and interviewer-respondent interaction, conducted as part of the Nebraska node of the NSF-Census Research Network.

The first component of data collection in the study is a computer-assisted telephone interview (CATI) survey. The survey contains substantive questions about sampled adults' employment, attitudes toward their job, leisure activities, computer activities, and demographics (including age, sex, marital status, family makeup, and level of education). No incentive was offered. The survey was administered only in English. Each version of the survey took 15 minutes on average to complete. All questions were administered by trained live interviewers. Each interview was audio-recorded and transcribed. Investigators then collected data about the interviewers (gender, race, and tenure), call records (date, respondent disposition), and survey paradata (questions, order of questions, and duration). Each survey interview was behavior-coded at the conversational turn level, in fields including actor, action, outcomes related to questions with parentheses, laughter, speech disfluencies, and interruptions. Additional coding was conducted to evaluate the accuracy of how interviewers recorded open-ended textual and numeric questions. Characteristics of screening and survey questions were coded using Flesch-Kincaid question reading levels and the Question Understanding Aid (QUAID) web tool.

Variables about individuals include work and leisure activities, demographics (including gender, race, income, and educational level), survey participation attitude, and conversational behaviors. Variables about survey questions include category, length, and readability.

Response Rates: Version 1: American Association for Public Opinion Reserach (AAPOR) RR3=7.54%; Version 2: AAPOR RR3=8.96%

Datasets:DS0: Study-Level FilesDS1: Survey Response Version 1 DataDS2: Survey Response Version 2 DataDS3: Interviewers DataDS4: Calls DataDS5: Paradata DataDS6: Behavior Codes DataDS7: Screencodes Version 1 DataDS8: Screencodes Version 2 DataDS9: Waveform Data

Adults living in landline households and adults with a cell phone Smallest Geographic Unit: Census region

The target population for this study was US adults in telephone (landline and cellular) households in the 48 contiguous US states. The survey has two independent probability-based random digit dial samples of telephone numbers in the contiguous United States. Sample was selected separately from the landline and cellular frames. A total of 32,599 phone numbers were dialed for version 1 (22,608 landline; 9,991 cell) and 30,996 phone numbers were dialed for version 2 (21,802 landline; 9,194 cell). Adults were randomly selected within households using the Rizzo, Brick, and Park method; adults in households with 3 or more adults were selected using the next birthday method.

Funding institution(s): National Science Foundation. Directorate for Social, Behavioral and Economic Sciences (SES-1132015).

computer-assisted telephone interview (CATI)

Metadata Access
Creator Olson, Kristen; Smyth, Jolene D.
Publisher Inter-University Consortium for Political and Social Research
Contributor National Science Foundation. Directorate for Social, Behavioral and Economic Sciences
Publication Year 2020
Rights Download; This version of the study is no longer available on the web. If you need to acquire this version of the data, you have to contact ICPSR User Support (
OpenAccess true
Contact Inter-University Consortium for Political and Social Research
Language English
Resource Type Dataset; survey data
Discipline Social Sciences