ANES Pilot Studies
| Study Title and Date |
ICPSR Study Number |
Key Design Features |
Number of Cases |
Number of Variables |
Special Study Content |
| 1979 Pilot Study: March and April 1979 |
7709 |
The 1979 pilot study was carried out in two waves. Wave I (March 5 to March 16) encompassed
280 interviews, and Wave II (March 26 to April 13) obtained 236 reinterviews. Each wave used two questionnaires,
Form A and Form B, administered to different half-samples. The 1979 pilot study differed from other pilots in
two respects. Its sample was independent of the previous year's (1978), and it employed personal interviews
instead of telephone interviews. |
280 |
930 |
Candidate evaluations (traits and affects); dimensions of partisanship; inflation vs. unemployment
tradeoff; social context (friends and neighborhood); most important problem follow-ups. EXPERIMENTS: Minimally
labeled feeling thermometer (labels at 0, 50 and 100 degrees) compared to a format in which the feeling thermometer
is labeled at 9 points. Ratings of "Republicans" and "Democrats" compared to ratings of "Republican Party" and
"Democratic Party" feeling thermometers. For issues, branching format was compared with 7-point scales, and
alternate wording and differing salience measures were compared. |
| Method Comparison Project: November 1982 |
8233 |
Because of parallel work in the development of computer-assisted telephone interviewing at Berkeley
and Michigan, the telephone sample was randomly allocated to two different interviewing sites, one operated by the
University of Michigan's Survey Research Center, and one by University of California-Berkeley's Program in
Computer-Assisted Survey Methods. This study took place in conjunction with the 1982 National Election Study.
Telephone respondents were administered a somewhat abbreviated version of the 1982 NES in-person interview
schedules, adapted for use over the telephone where appropriate, all interviewing conducted in the three months
following the 1982 elections. |
2416 |
964 |
The broad purposes of the data collection were to permit the NES staff and user community to assess
overall differences between the two data collection techniques in sample and measurement quality, and to carry out
a series of additional methodological experiments concerned with question format, computer-assisted procedures, and
survey organization. Focus is on comparing data obtained by personal interview and national data probability sample
with telephone computer-assisted interviews, sampling by RDD. EXPERIMENTS: Experimenting with use of seven point
scales versus branching format (in which the respondent is asked to place him/herself in stages, with the first
stage being a choice and the second stage being a strength or intensity followup of some kind) |
| 1983 Pilot Study: July and August 1983 |
8178 |
A sample of respondents from the 1982 NES Pre-Post Study was selected (probability sample) for two
waves of telephone interviews. Wave I (July 1 to July 25) encompassed 314 interviews, and Wave II (August 1 to
August 22) obtained 274 reinterviews. Each wave used two questionnaires, Form A and Form B, administered to
different half-samples. |
314 |
1183* |
Personal and national economic well-being, group identification, political participation, values,
and predispositions; candidate evaluations; attitudes toward gender/racial equality. EXPERIMENTS: Comparison of
branching (original question with follow-up for intensity/strength) format of liberal-conservative self- placement,
versus the seven-point liberal conservative scale self-placement. One year retrospective time frame versus
six-month time frame, in economic evaluations. |
| 1985 Pilot Study: November 1985 - January 1986 |
8476 |
From a sample of respondents from the 1984 NES Pre-Post Study selected for the 1985 pilot study,
15 replicates were drawn. Replicates 1-10 represent the cross-section, and replicates 11-15 constitute an
oversample of elderly (over 60). The 1985 study was carried out, via telephone interviews, in two waves. Wave I
(November 11 to December 13) encompassed 429 interviews, and Wave II (December 6, 1985, to January 16, 1986)
obtained 345 reinterviews. Each wave used two questionnaires, Form A and Form B, administered to different
half-samples (randomly assigned). Sample also limtited to those 1984 respondents who completed a pre- and a
post- election interview and had a telephone. |
429 |
1652* |
Measures of political knowledge; group identification of elderly, blacks, and women; attitudes
toward racial groups, and opinions on traditional moral values; important positions of U.S. Senators and
Representatives; political mobilization; civic duty; experiments on question wording and format. EXPERIMENTS:
Civic duty items: half sample received standard NES questions (these are agree/disagree) and other half received
these questions in a quantitative format, measuring either "how often" or "how strongly." Proximities
(placements) of presidential candidates on seven- point issue scales asked immediately after asking for
respondent's (self) placement -- versus, placing the presidential candidates several minutes after asking for
respondent's placement of him/herself. Framing experiment on two racial policy questions:"giving blacks
advantages they haven't earned" versus "discrimination against whites". Asking half the sample about the problem
of "inflation" in the last year versus asking about the cost of living relative to R's income over the last
year. |
| 1987 Pilot Study: May - June 1987 |
8713 |
A sample of respondents from the 1986 NES Post Study was selected for the 1987 pilot study,
which was carried out, via telephone interviews,in two waves. Wave I (May 5 to May 30) encompassed 457 interviews,
and Wave II (June 2 to July 2) obtained 360 reinterviews. Each wave used two questionnaires, Form A and Form B,
administered to different half-samples (randomly assigned). The sample was drawn form 1986 form A respondents who
had provided a telephone number for recontact. The sample was stratified on political information. The sampling
fraction for the stratum was proportional to predicted response rate. |
457 |
2147* |
New measures of system support, foreign policy predispositions, and morality; roles of U.S. Senators
and Representatives. EXPERIMENTS: Location of feeling thermometer on the questionnaire: beginning versus the end
of the questionnaire; this experiment has a test/retest component. Trust in government items: asked with
agree/disagree format and with standard NES format, i.e., "how much/how many." Experiment in priming: preceding a
Reagan approval question with anti-Reagan primes, (e.g., questions about budget deficit) versus preceding the
approval item with pro-Reagan primes (e.g. items reminding Rs about tax cut and lower inflation). Two series of
questions about what R thought were proper responsibilities of Senators and Representatives were asked widely
separated in Wave 1 and adjacent to each other in Wave 2. The order of a series of items about how U.S. should
respond to "terrorist acts" was reversed for half the sample. Experiments with making survey response more stable
and reliable: Asking R to "stop and think" before responding to a question about an issue; versus asking "what did
you have in mind when you answered that question" after the response is given (with test/retest). Experiment with
ways to reduce misreport of turnout: preceding traditional NES vote question with a request for the respondent to
report on turnout in the three elections preceding the last one versus simply asking the traditional NES vote
question for the most recent election. "Forced choice" partisan identification versus standard NES party ID: the
difference amounts to leaving out the response option "an independent" when the question is read. Asking R about
"welfare recipients" or "welfare programs" as opposed to "the poor" or "programs to assist the poor". Reading the
liberal-conservative and party ID questions with a filter: "or don't you think of yourself that way" versus reading
the questions without this filter. Asking R's opinion on several standard issue questions with and without the
filter "or haven't you thought much about this?" |
| 1989 Pilot Study: July - September 1989 |
9295 |
A subsample of respondents from the 1988 NES Pre-Post Study was selected for telephone interviews in
two waves. The sample (1988 Pre- and Post-election respondents with telephones) was stratified on level of political
information. The sampling fraction for each stratum was proportional to predicted response rate. Wave I (July 6 to
August 1) encompassed 614 interviews, and Wave II (September 6 to October 6) obtained 494 reinterviews. Each wave
used four questionnaires, Forms A through D. |
614 |
1934* |
New measures of religious affiliation, religiosity, and media exposure; questions about U.S. Senators
and Representatives. EXPERIMENTS: Using a full 100-point feeling thermometer compared to a 10 point scale.
Experimenting with use of seven point scales versus branching format (in which the respondent is asked to place
him/herself in stages, with the first stage being a choice and the second stage being a strength or intensity
followup of some kind). Unipolar versus bipolar question formats: that is, asking respondents to endorse or not
endorse, in successive questions, opposite poles of an issue dimension, versus asking respondents to chose between
bipolar opposites on an issue dimension. Interchanged order of recall of names of candidates for the House of
Representatives with recall of Senate candidate names. Interchanged order of asking about contacts with Senators
and Representatives. Half the sample received "Since January 1st" time frame for Congressional contact batteries;
the other half received the standard NES question (where "ever" is the implicit referent). Experiment with time
frame of reported media usage: half sample were asked about usage "in the past week;" the other half about usage
"in a typical week." Experiment with ways to reduce misreport of turnout: Half of the Rs were asked the usual NES
vote turnout question, while the other half were asked whether they "missed out" on voting.
Framing experiment on two racial policy questions:"giving blacks advantages they haven't earned" versus
"discrimination against whites". Experimenting with modifier in response options to liberal/conservative
questions: half the sample received "very" liberal/conservative while the other half "extremely"
liberal/conservative. |
| 1991 Pilot Study: June - July 1991 |
9673 |
One wave of telephone interviews. Attempt to re-interview all respondents to 1990 NES. This study was
also part of a larger panel study (1990-1991-1992). All respondents were asked the questions in the Panel section
(on the political consequences of war); the pilot aspects of the 1991 study appeared in three separate forms, which
were randomly assigned. 1385 interviews were conducted June 4 to June 31. |
1980 |
1124* |
New content includes ethnic politics, gender consciousness, attitudes toward social security and
medical care for the elderly, social altruism, and political knowledge. EXPERIMENTS: Government spending and
services seven-point scale (Form 1) versus reading seven labeled response options Form 2) versus branching format
(Form 3). Feeling thermometer for President Bush preceded by three different Democratic candidates on the on three
forms of 1991 Pilot. Feeling thermometer for "environmentalists" on half the sample and on "people seeking to
protect the environment" on the other half. |
| 1993 Pilot Study: September - November 1993 |
6264 | An attempt was made to reinterview all 1,005 'fresh' cross-section
respondents from the 1992 Study (respondents not previously interviewed in 1990). Telephone interviews. Of the 1005
cases from 1992, 750 were interviewed over a period from September 23 to November 24, 1993. Respondents were
randomly assigned to one of four forms. |
1005 |
1474* |
New content on attitudes toward homosexuality; group interest. Several experiments in response formats.
EXPERIMENTS: Policy choice items: half sample received strength (strongly/not strongly) follow-ups; other half
sample received quantity (a lot/a little) follow-ups. Placement of President Clinton on liberal-conservative seven
point scale, before and after R is asked to place him/herself. Experiment in wording of the vote choice for U.S.
Representative question: reading candidate name as well as party versus reading only party labels for candidates.
Description of the Clinton package as "budget package" versus "deficit reduction package". Certainty probe on
Liberal-Conservative scale. Question order experiment in 'interest basis of politics' section, where R evaluates
the effects of a government policy upon list of objects (order of objects). |
| 1995 Pilot Study: August and September 1995 |
6636 | Conducted between August 3 and September 10th, 1995. Respondents
were seleted from 1994 'fresh' Cross-section cases (respondents not previously interviewed in 1992). In order to
include all of the content, and to test between competing instrumentation, there were two forms of the
questionnaire. Rosters of items, such as the thermometer, were randomized in administration to minimize order
effects. |
486 |
1124* |
The content of the study reflects the NES commitment to improve measures of candidate evaluation, the
impact of the campaign, values and predispositions, the comparative study of elections,and other responses to a
stimulus letter calling for ideas for content sent to the user community on November 4, 1994. EXPERIMENTS: (1) an
experiment using different measures of affective reactions to political figures, (2) a module of items that is part
of a comparative study of politics (CSES), (3) a set of 12 items asking respondents to make tradeoffs between
programs, taxes, and the budget deficit, (4) a set of items designed to measure attitudes toward the environment
and environmental policy, (5) a new measure of "humanitarianism," and (6) an extensive set of items regarding
attention to the media, intended to capture exposure to the political campaigns. |
| 1997 Pilot Study: September 5 - October 1, 1997 |
2282 |
This study is a one-wave reinterview of a randomly-selected subset of 1996 respondents with
telephones; most of the selection from 1996 were fresh' Cross-section cases (respondents not previously interviewed
in 1994). |
551 |
1850** |
The content of the study reflects the NES commitment to improve measures of group mobilization,
interest articulation and representation, group-based political reasoning, race and racial attitudes and policy,
issue attitudes, human and social capital, social choice, and theories of the survey response.
EXPERIMENTS: (1) a battery designed to improve NES instrumentation on nonelectoral political participation and
mobilization, (2) testing of NES instrumentation on group closeness, group difference, and group conflict as a
basis of current mass politics, and group threat as a basis of group-based politics, (3) evaluations of the
President, Congress, and the Supreme Court using a new battery of items, and (4) the role of religion in citizens'
political thinking. The use of CATI enabled a number of experimental treatments within the survey instrumentation,
including random assignment, early-late placement, and presentation order. In addition, rosters of items, such as
the thermometer, were randomized in administration to minimize order effects. |
