American National Election Studies, Time Series Study, 2012

Data Archive > U.S. Surveys > General Population > National > National Election Studies > Summary



The ANES 2012 Time Series Study is the 29th study in a series of election studies conducted during years of Presidential elections since 1948 (the "ANES Time Series"). As with all Time Series studies conducted during years of presidential elections, respondents were interviewed during the two months preceding the November election (Pre-election interview), and then re-interviewed during the two months following the election (Post-election interview).

For the first time in Time Series history, face-to-face interviewing was supplemented with data collection on the Internet. Data collection was conducted in the two modes independently, using separate samples. For the face-to-face mode, all sampled persons were interviewed in person using Computer-Assisted Personal Interviewing (CAPI), which also incorporated an interview segment in each wave that was self-administered (CASI). For the Internet mode, all study participants were members of the KnowledgePanel, a panel of regular survey participants administered by GfK (formerly Knowledge Networks).

(ANES. 2014. User’s Guide and Codebook for the ANES 2012 Time Series Study. Ann Arbor, MI and Palo Alto, CA: the University of Michigan and Stanford University.)
Data File
Cases: 5,916
Variables: 2,249
Weight Variable: WEIGHT_FTF, WEIGHT_WEB, WEIGHT_FULL
Additional Weight Variable Information: There are three weight variables on the file, intended for different purposes. Each weight is preliminary and improved weights may be released with the next version of the data file if ANES personnel find that practical improvements can be made.

WEIGHT_FTF is intended for analysis of the face-to-face sample alone. It may be used for pre-election or post-election variables, or a combination of the two. The effects of attrition in the face-to-face sample are negligible for most analyses we have examined, therefore just one weight may be satisfactory for analysis of the face-to-face sample alone.

WEIGHT_WEB is intended for analysis of the Internet sample alone. As with the face-to-face sample, attrition effects are small, so this weight may be used for analysis of the pre-election variables alone, post-election variables alone, or the two in combination.

WEIGHT_FULL is intended for the analysis of the combined samples. If you want to include all available cases, use this weight.
Data Collection
Date Collected: Pre-election, Face-to-Face: September 8 – November 5, 2012 Pre-election, Internet: October 11 – November 6, 2012 Post-election, Face-to-face: November 7, 2012 – January 13, 2013 Post-election, Internet: November 29, 2012 – January 24, 2013
Original Survey (Instrument)
American National Elections Survey (ANES) 2012 Time Series Study
Funded 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
As with all Time Series studies conducted during years of presidential elections, respondents were interviewed during the two months preceding the November election (Pre-election interview) and then re-interviewed during the two months following the election (Post-election interview). Face-to-face interviews were conducted with one interview before the election and a second interview with the same respondents after the election. Internet interviews were conducted with two interviews before the election and two more interviews with the same respondents after the election.

For both the Internet and face-to-face data collection, the study was described using a name that sounded less political than “American National Election Studies.” This name is confidential so that future study participants will not be able to readily associate the study with the ANES. The name is a relatively general label for the study. Two names that were considered and rejected were the National Study of American Life and the Survey of the American Public. The actual name is similar to these examples. We use this generic name with respondents to reduce the potential for non-response bias related to interest in politics.

Internet Data Collection
Internet respondents were invited to the survey by email. They followed a link to a web address to complete the survey over the internet. The survey began with a consent screen that provided information about the study and included contact information for the study administrators. Prior to completing the ANES surveys, internet respondents had completed profile surveys on demographic and political topics. Data from these profile surveys are included in the release data files. In addition to completing the ANES questions, internet respondents also completed a Direct Democracy module, for which ANES will release data in future.

Face-to-Face Data Collection
Interviewers were provided with lists of sampled addresses in their work area and maps showing the locations of these addresses. Upon reaching a sampled address, interviewers completed the CHUM procedure described in part 3, if appropriate. Interviewers also made observations regarding the dwelling unit, such as noting its state of repair and the presence of political yard signs. Interviewers then attempted to complete a brief screening interview at the selected address. During the screening interview, interviewers sought to make a list of each eligible person residing at the household. For the main sample, this consisted of U.S. citizens age 18 or older by Election Day. For the black and Hispanic over-samples, this consisted of U.S. citizens age 18 or older by Election Day who were black, or Hispanic, respectively. After completing this household listing, the computer randomly selected one eligible person from the list. An interview could be conducted only with the selected person; no substitutions were allowed.

(ANES. 2014. User’s Guide and Codebook for the ANES 2012 Time Series Study. Ann Arbor, MI and Palo Alto, CA: the University of Michigan and Stanford University.)
Sampling Procedures
The ANES 2012 Time Series is a dual-mode survey (face-to-face and Internet) with two independent samples. Cases selected for the face-to-face sample could not be interviewed on the Internet, and cases selected for the Internet survey could not be interviewed in person. The target population for the two samples is U.S. citizens age 18 or older. Design criteria also included having sufficient numbers of black and Hispanic respondents to enable analysis of those subgroups.

Internet Sample:
Internet respondents were members of the GfK (formerly Knowledge Networks) KnowledgePanel. The KnowledgePanel is a large online panel of survey respondents who are invited to complete surveys several times each month on a variety of topics for a variety of investigators. Panelists are recruited using two probability sampling methods: address-based sampling (ABS) and random-digit dialing (RDD). Prospective panelists who do not have Internet access at the time of recruitment are furnished with free Internet service and free hardware to connect to the Internet. A sample of KnowledgePanelists selected from the KnowledgePanel to receive invitations to take the ANES Time Series survey. This sample was limited to U.S. citizens who would be at least 18 years old by Election Day, November 6, 2012, and was limited to one person per household.

Face-to-Face Sample:
The in-person (face-to-face) interviews were conducted using an address-based, stratified, multi-stage cluster sample in 125 census tracts. The sample includes a nationally-representative “main sample” and two “over-samples,” one of blacks and one of Hispanics. The target number of interviews was 1,400 responses from the main sample (including people of all racial and ethnic backgrounds) and 300 responses from the black over-sample and 300 responses from the Hispanic over-sample, for a total of 2,000 pre-election interviews. Note that in contrast to the ANES 2008 Time Series over-sample design, in which there was one sample with selection probabilities adjusted to produce over-samples of blacks and Hispanics, the 2012 design has distinct samples for the over-samples, making it possible to distinguish respondents as belonging to the main sample or an over-sample.

(ANES. 2014. User’s Guide and Codebook for the ANES 2012 Time Series Study. Ann Arbor, MI and Palo Alto, CA: the University of Michigan and Stanford University.)
Principal Investigators
Vincent Hutchings, the University of Michigan; Gary Segura, Stanford University; Simon Jackman, Stanford University.

Associate Principal Investigator: Ted Brader, the University of Michigan
Bookmark and Share