American National Election Studies, Time Series Study, 2016
DOI
10.17605/OSF.IO/ZCVTACitation
Hutchings, V., Brader, T., Iyengar, S., Segura, G., & Jackman, S. (2020, March 3). American National Election Studies, Time Series Study, 2016.Summary
"The ANES 2016 Time Series is a continuation of the series of election studies conducted by the ANES since 1948 to support analysis of public opinion and voting behavior in U.S. presidential elections. This year's study features a dual-mode design with both traditional face-to-face interviewing (n=1,181) and surveys conducted on the Internet (n=3,090), and a total sample size of 4,271."Study Content Highlights:
Data collection for the ANES 2016 Time Series Study began in early September and continued into January 2017. Pre election interviews were conducted with study respondents during the two months prior to the 2016 elections and were followed by post-election re-interviewing beginning November 9, 2016.
"As in 2012, face-to-face interviewing was complemented with data collection on the Internet. Data collection was conducted in the two modes independently, using separate samples but substantially identical questionnaires. Web-administered cases constituted a representative sample separate from the face-to-face.
(ANES. 2017. User's Guide and Codebook for the ANES 2016 Time Series Study. Ann Arbor, MI and Palo Alto, CA: the University of Michigan and Stanford University.)
The ARDA has added four additional variables to the original data set to enhance the users' experience on our site.
Data File
Cases: 4271Variables: 1196
Weight Variable: V160101, V160101F, V160101W, V160102, V160102F, V160102W
To analyze the complete dataset (both web and face-to-face modes) using variables from the post-election survey only or from both the pre-election and post-election questionnaires, use the weight V160102. For analysis of the complete dataset using only the pre-election questionnaire data, use the weight V160101.
Weight Use for analysis of the...
V160102 full sample using post-election survey only or both pre and post
V160101 full sample using pre-election survey data only
V160102F face-to-face mode alone, using the post-election survey or both pre and post
V160101F face-to-face mode alone, using pre-election survey data only
V160102W Internet mode alone, using data from both pre- and post-election or post alone
V160101W Internet mode alone, using data from only the pre-election survey
Data Collection
Pre-election survey: September 7 - November 7, 2016 Post-election survey: November 9 - January 8, 2017Original Survey (Instrument)
American National Elections Survey (ANES) 2016 Time Series StudyFunded By
This material is based upon work supported by the National Science Foundation Under Grants SES-0937727 and SES-0937715. Additional funding from NSF was made available in conjunction with the Department of Homeland Security. ANES is also supported by the University of Michigan: Center for Political Studies, and Office of the Provost and Stanford University: Department of Political Science, Institute for Research in the Social Sciences, and Office of the Vice Provost and Dean of Research.Collection Procedures
The study interviewed respondents in a pre-election survey between September 7 and November 7, 2016. Election day was November 8. The study re-interviewed as many as possible of the same respondents in a post-election survey between November 9 and January 8, 2017.In-person interviews were normally conducted in the respondent's homes, but were allowed to be conducted at any location that was convenient for the respondent. Internet questionnaires could be completed anywhere the respondent had access to the Internet, on a computer or on a mobile device. In-person interviews were conducted by trained interviewers using computer assisted personal interviewing (CAPI) software on laptop computers. During a portion of the in-person interview, the respondent answered certain sensitive questions on the laptop computer directly, without the interviewer's participation (known as computer assisted self-interviewing). Respondents were only eligible to compete the survey in the mode for which they were sampled; no one sampled for the Internet study was interviewed by a live interviewer, and no one sampled for the face-to-face interview completed the study online.
The pre-election and post-election questionnaires were designed for a median administration time of approximately 80 minutes each, not counting screening procedures. Interviews in both modes were conducted in English or Spanish.
(ANES. 2017. User's Guide and Codebook for the ANES 2016 Time Series Study. Ann Arbor, MI and Palo Alto, CA: the University of Michigan and Stanford University.)
Sampling Procedures
"The study has two independently drawn probability samples that describe approximately the same population. The target population for the face-to-face mode was 222.6 million U.S. citizens age 18 or older living in the 48 contiguous states of the USA or the District of Columbia, and the target population for the Internet mode was 224.1 million U.S. citizens age 18 or older living in the 50 US states or the District of Columbia. In both modes, the sampling frame was lists of residential addresses where mail is delivered, and to be eligible to participate, a respondent had to reside at the sampled address and be a U.S. citizen age 18 or older at the time of recruitment."Face-to-face face mode: The sample for in-person interviews was a multi-stage stratified cluster sample. This is similar in general design but different in details from past designs of ANES Time Series studies. Sixty primary sampling units (PSUs) were selected from across the 48 contiguous states and Washington DC. The PSUs were counties, or combinations of counties to form a minimum population of 50,000, or, in the case of Los Angeles County, half-counties (i.e. L.A. County was divided into two PSUs due to its large size). Counties were stratified by Census region, prevalence of poverty and members of minority groups, and population size, and then selected at random with probability proportional to the number of adult citizens, except that Cook County, IL, Harris County, TX, Maricopa County, AZ, and both halves of Los Angeles, CA, were selected with certainty.
"Within each PSU, four smaller areas were drawn (secondary sampling units, which were Census Block Groups), and households were selected at random from within these areas from the US Postal Service's computerized delivery sequence file (DSF). An advance letter was sent to each selected address that included a study brochure and $5 in cash, after which an interviewer visited the address to complete a screening interview to randomly select one person to be interviewed from among the adult U.S. citizens living there. Those interviewed were given $25 or $50 during the initial phase of the field period (with $50 offered in PSUs where predicted response propensity was low and $25 offered elsewhere), escalating to $100 after initial non-response.
"During the last two weeks of data collection, subsampling was performed in which half of the remaining eligible cases from the face-to-face were dropped from the study to focus field efforts on the other half. The weights account for this sub-sampling by up-weighting the cases retained at this stage.
"Internet mode: The sample for the Internet mode was a random draw from the DSF (excluding 'drop point' addresses), with all included residential addresses across the 50 states and Washington DC having equal probability of selection. Selected addresses were sent a series of letters to recruit one household member to go online to complete a survey. The invitation letter included $20 in cash and promised $40 for completing a survey online (later escalated to $80 for initial non-respondents). This online survey included a screening instrument to randomly select one person from among the adult U.S. citizens living at the address to complete the ANES questionnaire online and receive $40 or $80.
"Mode sample differences: The face-to-face component of the study does not include respondents from Alaska or Hawaii; the Internet component does. This is the main substantive design difference in eligibility and coverage for the two modes and for most analytic purposes it is probably ignorable because the combined population of Alaska and Hawaii is less than 1 percent of the study population."
(ANES. 2017. User's Guide and Codebook for the ANES 2016 Time Series Study. Ann Arbor, MI and Palo Alto, CA: the University of Michigan and Stanford University.)
