(1) Overview

Collection Date(s)

2019–2020, United States of America.

Background

The dynamics of political attitudes are of interest to social scientists because they help us understand if attitude change is possible and under what conditions this occurs. Moreover, these dynamics can help scholars understand the underpinnings of political attitudes. Much of the work on political attitudes uses data that makes the study of attitude change dynamics difficult. Although scholars often go beyond cross-sectional data, longitudinal data often comes from different participants at each time point (such as in the World Values Survey, American National Election Studies, or the General Social Survey) or from the same participants with time points approximately 1 year apart (such as the Longitudinal Internet Studies for the Social Sciences Panel and the New Zealand Attitudes and Values Survey). Although this is valuable data, if attitudes change at a faster rate, or have dynamics that occur over shorter time periods, they cannot be detected in such studies. Moreover, many longitudinal studies in this domain only include 2 or 3 time points, making it difficult to observe some types of dynamic processes (e.g., [9]; although for exceptions see Longitudinal Internet Studies for the Social Sciences Panel and the New Zealand Attitudes and Values Survey). Here we aimed to collect data that would provide a more fine-grained look at the dynamic processes underlying the maintenance and change of political attitudes.

We included measures of constructs that would have broad interest to social scientists, including political attitudes and identities [2], perceived threats and stresses [6], political engagement [3], and social distance from political groups [5]. Each of these constructs have a long tradition in the political psychological literature. Our study aims to provide data about their variation over the course of one year. We also included items related to the COVID-19 pandemic, which overlapped with the end of our study. This makes it possible to assess COVID-19-related attitudes over a part of the study.

(2) Methods

The methods for this data collection were preregistered. There are three preregistrations associated with the project. The first preregisters the procedure and original measures used in the study. It was completed prior to the start of data collection, and is available here: https://osf.io/7h5ds. The second preregistration documents measures of behavioural intentions we added after data collection began, and was completed in December 2019. It is available here: https://osf.io/7r4bk. The third preregistration documents measures we added related to COVID-19. It was completed in March 2020 after data collection had started. It is available here: https://osf.io/rxcqf.

There were changes to the study that are not documented with preregistrations (noted below). These include a measure related to climate policy (added for Wave 2–26) and two attention check items in two different waves.

Sample

Participants were recruited on Prolific (see Procedure below for details) using quota sampling based on vote choice in the 2016 U.S. elections. In the first wave, we had 552 participants from the United States (M age = 34.7, SD = 12.4, Range [18, 73]; 278 women, 271 men, 2 indicated their gender was not listed). Our average response rate across the waves was 75%, SD = 8%, Range [63%, 93%]. The sample size for each wave and for all pairwise combination of waves is in Table 1. The number of waves completed by each participant is visualized in Figure 1. The sample size based on the 2016 U.S. election vote is in Table 2.

Table 1

Sample sizes per wave and pairwise sample sizes for each wave combination.


WAVE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

1 552 510 503 487 471 454 449 442 435 421 416 420 397 403 398 411 381 391 394 381 400 369 373 381 348 348

2 510 485 470 454 437 431 424 420 409 403 411 388 393 385 400 372 379 384 372 387 360 359 368 339 337

3 503 468 452 437 434 429 418 404 400 405 382 388 381 393 366 375 378 367 383 357 358 363 336 333

4 487 448 434 427 419 408 402 394 398 376 384 374 389 361 369 375 365 380 350 352 358 332 328

5 471 435 427 417 410 395 391 391 374 377 371 386 359 367 367 361 376 347 353 359 328 328

6 454 417 412 399 388 385 390 369 373 365 375 353 359 359 355 368 346 349 349 321 320

7 449 413 401 393 386 389 369 371 365 374 356 358 361 357 367 343 345 349 323 322

8 442 407 396 386 394 371 373 367 376 354 361 364 356 367 344 347 351 325 322

9 435 387 386 384 368 369 363 373 352 362 364 354 369 343 339 346 317 319

10 421 389 384 363 367 364 367 350 352 355 354 365 337 341 342 317 315

11 416 387 366 368 364 367 356 357 357 353 360 337 340 342 321 315

12 420 376 380 375 381 360 363 365 362 369 343 347 348 324 320

13 397 370 357 365 346 345 349 343 351 331 328 336 307 304

14 403 376 378 358 357 360 354 364 341 342 347 319 318

15 398 374 358 356 359 350 361 338 338 343 314 316

16 411 364 363 364 359 365 339 342 351 320 317

17 381 355 353 344 352 333 336 343 314 309

18 391 363 349 357 338 334 342 319 314

19 394 354 360 340 339 344 317 315

20 381 360 335 342 337 312 310

21 400 350 351 357 326 325

22 369 333 338 309 313

23 373 345 316 315

24 381 328 327

25 348 314

26 348

Figure 1 

Histogram of number of waves completed.

Table 2

Sample size per 2016 vote choice.


TARGET N ACHIEVED N PROLIFIC OPTION

223 223 Did not vote/Prefer not to say or NA

161 161 Trump

166 168 Clinton

Materials

The materials for all of the waves are stored with the data. In addition, the materials are posted at the OSF page for the project (https://osf.io/x94rc).

Participants completed measures related to their positions on political issues, political identifications, political interest, presidential approval, feelings of internal threat (e.g., stress), feelings of external threat (e.g., from the economy), and social distance from political groups at each wave of the study. At the first wave, participants also completed demographic measures and reported their vote choice from the 2016 presidential election. This latter item is used to assess the correspondence between the prescreening measure provided by Prolific and our participants own self-report in the current study.

We preregistered that we may add items throughout the course of the study. In total we added seven new items to the survey. 1: We added an issue about climate change at Wave 2 that remained in the study until the end. 2: We added an item about the importance of the July 4th holiday in Waves 4 and 9 to help us test preregistered hypotheses about the effect of the holiday on political prejudice (see preregistrations for that study here https://osf.io/26bua/registrations). 3: We added an attention check item as the last item in both Waves 12 and 15. 4: We added behavioral intention items in Wave 17, 20, 23, and 26. 5–7: Finally, we added three items relevant to the COVID-19 pandemic in the last four waves. All items are listed in Table 3.

Table 3

Item summary.


NAME (TYPE) VARIABLE RESPONSE OPTIONS WAVE

def (issue) Should federal spending on defense be increased, decreased, or kept the same? 1. Greatly decrease defense spending
2.
3.
4. Keep defense spending about the same
5.
6.
7. Greatly increase defense spending.
8. Don’t Know
9. I haven’t thought much about it
All Waves

crime (issue) Should federal spending for dealing with crime be increased, decreased, or kept the same? 1. Greatly decrease spending for dealing with crime
2.
3.
4. Keep spending for dealing with crime about the same
5.
6.
7. Greatly increase spending for dealing with crime.
8. Don’t Know
9.I haven’t thought much about it
All Waves

terror (issue) Should federal spending on the war on terrorism be increased, decreased, or kept the same? 1. Greatly decrease war on terror spending
2.
3.
4. Keep war on terror spending about the same
5.
6.
7. Greatly increase war on terror spending.
8. Don’t Know
9. I haven’t thought much about it
All Waves

poor (issue) Should federal spending on aid to the poor be increased, decreased, or kept the same? 1. Greatly decrease aid to the poor
2.
3.
4. Keep aid to the poor about the same
5.
6.
7. Greatly increase aid to the poor
8. Don’t Know
9. I haven’t thought much about it
All Waves

health (issue) Should federal spending on healthcare be increased, decreased, or kept the same? 1. Greatly decrease healthcare spending
2.
3.
4. Keep healthcare spending about the same
5.
6.
7. Greatly increase healthcare spending
8. Don’t Know
9. I haven’t thought much about it
All Waves

econ (issue) Should federal spending to stimulate the economy be increased, decreased, or kept the same? 1. Greatly decrease economic stimulus
2.
3.
4. Keep economic stimulus about the same
5.
6.
7. Greatly increase economic stimulus
8. Don’t Know
9. I haven’t thought much about it
All Waves

abort (issue) Which one of the opinions below best agrees with your view of abortion? 1. By law, abortion should never be permitted.
2. The law should permit abortion only in case of rape, incest, or when the woman’s life is in danger.
3. The law should permit abortion for reasons other than rape, incest, or danger to the woman’s life, but only after the need for the abortion has been clearly established.
4. By law, a woman should always be able to obtain an abortion as a matter of personal choice.
5. Don’t Know
6. I haven’t thought much about it
All Waves

unemploy (issue) Should federal spending on benefits for the unemployed be increased, decreased, or kept the same? 1. Greatly decrease benefits for the unemployed
2.
3.
4. Keep benefits for the unemployed about the same
5.
6.
7.Greatly increase benefits for the unemployed
8. Don’t Know
9. I haven’t thought much about it
All Waves

blkaid (issue) Should federal spending to improve the social and economic position of blacks be increased, decreased, or kept the same? 1. Greatly decrease aid to blacks
2.
3.
4. Keep aid to blacks about the same
5.
6.
7. Greatly increase aid to blacks
8. Don’t Know
9. I haven’t thought much about it
All Waves

adopt (issue) Do you favor or oppose laws that prevent gay or lesbian couples from adopting children, or haven’t you thought much about it? 1. Favor strongly
2.
3.
4. Neither favor nor oppose
5.
6.
7. Oppose strongly
8. Don’t Know
9. I haven’t thought much about it
All Waves

imm (issue) Should federal spending to control immigration be increased, decreased, or kept the same? 1. Greatly decrease spending on immigration control
2.
3.
4. Keep spending on immigration control about the same
5.
6.
7. Greatly increase spending on immigration control
8. Don’t Know
9. I haven’t thought much about it
All Waves

vaccines (issue) Do you favor or oppose laws that require parents to vaccinate their children using common vaccines (e.g., polio, tetanus, measles, flu)? 1. Favor strongly
2.
3.
4. Neither favor nor oppose
5.
6.
7. Oppose strongly
8. Don’t Know
9. I haven’t thought much about it
All Waves

guns (issue) Should the federal government make it more difficult for people to buy a gun than it is now, make it easier for people to buy a gun, or keep these rules about the same as they are now? 1. Make it much more difficult
2.
3.
4. Keep the rules the same
5.
6.
7. Make it much easier
8. Don’t Know
9. I haven’t thought much about it
All Waves

djt (approval) Do you approve or disapprove of the job Donald Trump is doing as president? 1. Strongly approve
2. Approve
3. Disapprove
4. Strongly disapprove
5. Don’t know
6. I haven’t thought much about it
All Waves

interest (interest) How interested are you in politics? 1. Very uninterested
2.
3.
4.
5.
6.
7. Very interested
All Waves

friends_1 (prejudice) How willing would you be to be friends with people from the following groups? Liberals 1. I absolutely would not
2.
3.
4.
5.
6.
7. I absolutely would
8. Don’t know
9. I haven’t thought much about it
All Waves

friends_2 (prejudice) How willing would you be to be friends with people from the following groups? Conservatives 1. I absolutely would not
2.
3.
4.
5.
6.
7. I absolutely would
8. Don’t know
9. I haven’t thought much about it
All Waves

friends_3 (prejudice) How willing would you be to be friends with people from the following groups? Moderates 1. I absolutely would not
2.
3.
4.
5.
6.
7. I absolutely would
8. Don’t know
9. I haven’t thought much about it
All Waves

friends_4 (prejudice) How willing would you be to be friends with people from the following groups? Republicans 1. I absolutely would not
2.
3.
4.
5.
6.
7. I absolutely would
8. Don’t know
9. I haven’t thought much about it
All Waves

friends_5 (prejudice) How willing would you be to be friends with people from the following groups? Democrats 1. I absolutely would not
2.
3.
4.
5.
6.
7. I absolutely would
8. Don’t know
9. I haven’t thought much about it
All Waves

tense (internal threat) At this moment, I feel tense. 1. Fully disagree
2.
3.
4. Neither agree nor disagree
5.
6.
7. Fully agree
All Waves

death (internal threat) I have an intense fear of death 1. Fully disagree
2.
3.
4. Neither agree nor disagree
5.
6.
7. Fully agree
All Waves

ewry (external threat) I worry that I myself or someone from my family will be worse off financially in the near future. 1. Fully disagree
2.
3.
4. Neither agree nor disagree
5.
6.
7. Fully agree
All Waves

values (external threat) The values in our country have gone seriously off track. 1. Fully disagree
2.
3.
4. Neither agree nor disagree
5.
6.
7. Fully agree
All Waves

climate (issue) Do you think the federal government should be doing more about climate change, should be doing less, or is it currently doing the right amount? 1. Doing more about climate change
2.
3.
4. Doing the right amount
5.
6.
7. Doing less about climate change
8. Don’t Know
9. I haven’t thought much about it
Waves 2 – Wave 26

fourth (holiday) I find it important to celebrate the 4th of July. 1. Fully disagree
2.
3.
4. Neither agree nor disagree
5.
6.
7. Fully agree
Wave 4 (some participants),
Wave 9

ideo (identification) When it comes to politics, do you think of yourself as a liberal, conservative, moderate, or haven’t you thought much about this? 1. Strongly liberal
2. Liberal
3. Slightly liberal
4. Moderate, middle of the road
5. Slightly conservative
6. Conservative
7. Strongly conservative
8. Don’t know
9. I haven’t thought much about it
All Waves

partyid (identification) Do you think of yourself as a Republican, a Democrat, an Independent, or haven’t you thought much about this? 1. Strongly Democrat
2. Democrat
3. Independent, lean Democrat
4. Independent
5. Independent, lean Republican
6. Republican
7. Strongly Republican
8. Don’t know
9. I haven’t thought much about it
All Waves

check (attention check) We are using this question to check your attention. Please select “Don’t know”. 1. Strongly Democrat
2. Democrat
3. Independent, lean Democrat
4. Independent
5. Independent, lean Republican
6. Republican
7. Strongly Republican
8. Don’t know
9. I haven’t thought much about it
Wave 12

check (attention check) This question is checking your attention. Please select “I haven’t thought much about it”. 1. Strongly Democrat
2. Democrat
3. Independent, lean Democrat
4. Independent
5. Independent, lean Republican
6. Republican
7. Strongly Republican
8. Don’t know
9. I haven’t thought much about it
Wave 15

beh_att_1 How likely would you be to sign a petition in support of the following issues?
…increase aid to the poor
1. Very unlikely to sign petition
2.
3.
4. Neither likely nor unlikely to sign the petition
5.
6.
7. Very likely to sign the petition.
8. Don’t Know
Waves 17, 20, 23, 26

beh_att_2 How likely would you be to sign a petition in support of the following issues?
…increase spending to stimulate the economy
1. Very unlikely to sign petition
2.
3.
4. Neither likely nor unlikely to sign the petition
5.
6.
7. Very likely to sign the petition.
8. Don’t Know
Waves 17, 20, 23, 26

beh_att_3 How likely would you be to sign a petition in support of the following issues?
… increase spending on healthcare
1. Very unlikely to sign petition
2.
3.
4. Neither likely nor unlikely to sign the petition
5.
6.
7. Very likely to sign the petition.
8. Don’t Know
Waves 17, 20, 23, 26

beh_identity In the next Presidential election, how likely are you to vote for the candidate from the Democratic Party or the Republican Party? 1. Very likely to vote for the Democratic candidate
2. Likely to vote for the Democratic candidate
3. Somewhat likely to vote for the Democratic candidate
4. Equally likely to vote for the Democratic as the Republican candidate
5. Somewhat likely to vote for the Republican candidate
6. Likely to vote for the Republican candidate
7. Very likely to vote for the Republican candidate
8. I would vote for another party, namely: ….
9. Don’t know
Waves 17, 20, 23, 26

virusthreat How concerned are you about the coronavirus (COVID-19)? 1. Not at all concerned
2.
3.
4. Somewhat concerned
5.
6.
7. Very concerned
Waves 23–26

quarantine Do you favor or oppose laws that prohibit travel to and from regions in the United States with coronavirus (COVID-19) outbreaks (i.e. a quarantine)? 1. Favor strongly
2.
3.
4. Neither favor nor oppose
5.
6.
7. Oppose strongly
8. Don’t Know
9. I haven’t thought much about it
Waves 23–26

sickleave Do you favor or oppose laws that would require all businesses to pay for their employee’s sick leave? 1. Favor strongly
2.
3.
4. Neither favor nor oppose
5.
6.
7. Oppose strongly
8. Don’t Know
9. I haven’t thought much about it
Waves 23–26

votereport (voting) Who did you vote for in the 2016 presidential election? 1. Donald Trump
2. Hillary Clinton
3. A different candidate
4. I did not vote
5. I planned to vote, but I forgot
6. I do not remember who I voted for
Wave 1

gender (demographics) To which gender identity do you most identify? 1. Female
2. Male
3. Not Listed
4. Prefer not to answer
Wave 1

ethnic (demographics) I identify my ethnicity as: (select all that apply) 1. Asian
2. Black/African
3. Caucasian
4. Hispanic/Latinx
5. Native American
6. Pacific Islander
7. Not listed
8. Prefer not to answer
Wave 1

edu (demographics) What is the highest degree or level of school you have completed? If currently enrolled, highest degree received. 1. No schooling completed
2. Nursery school to 8th grade
3. Some high school, no diploma
4. High school graduate, diploma or the equivalent (for example: GED)
5. Some college credit, no degree
6. Trade/technical/vocational training
7. Associate degree
8. Bachelor’s degree
9. Master’s degree
10. Professional degree
11. Doctorate degree
Wave 1

inc (demographics) What is your household income? That is, the total income of everyone living in your residence. 1. Less than $20,000
2. $20,000 to $34,999
3. $35,000 to $49,999
4. $50,000 to $74,999
5. $75,000 to $99,999
6. Over $100,000
Wave 1

age (demographics) What is your age? Wave 1

state (demographics) What state do you currently live in? Wave 1

relig (demographics) What is your present religion, if any? 1. Protestant
2. Roman Catholic
3. Mormon
4. Orthodox such as Greek or Russian Orthodox
5. Jewish
6. Muslim
7. Buddhist
8. Hindu
9. atheist
10. agnostic
11. something else
12. nothing in particular
Wave 1

The issue positions, political interest, presidential approval, feelings of internal threat, feelings of external threat, and political prejudice items were completed first and in a random order. For the purposes of the randomization, the five political prejudice items were treated as one item and the order of the five targets were randomized within that one item. The two political identification items came next and were presented in a random order. The remaining measures in Wave 1 were presented last and in the order they appear in Table 3. The behavioral intention items were presented last and in a random order. The COVID-19 items were presented in the order they appear in Table 3 and before the block of items containing political identification items.

Procedures

Participants were recruited on Prolific, an online service that facilitates the crowdsourcing of research participants (for an overview see [7]). The first wave of the study started on 8 May 2019. For the first wave, the survey was left open until the three quotas (described below) were fulfilled (1 day). For the 2nd–26th wave the survey was left open for one week. A new survey was posted every 2 weeks. This means that there is somewhere between a one week and two week gap between each wave for each participant. The final wave started data collection on 22 April 2020.

We opened the task on Prolific to 550 participants with the expectation of having between 250 and 500 participants at each individual wave. Prolific allows us to target people based on their self-reported vote in the 2016 election. For the first wave, we collected data from participants in proportion to their vote choice and the population-level outcome of this election (see Table 2). The target Ns and achieved Ns are reported in Table 2. This was done to increase the diversity of political opinions within the sample. The study was additionally limited to people reporting the United States as their nationality and who had an approval rate 90 or higher.

Participants were paid £0.35 for completing the first wave (~4 minutes) and for completing each of the next four follow-up waves (~3 minutes). Every fifth wave, participants were paid £0.02 more per survey, so that the payment at the final survey was £0.45. If they completed all 26 surveys, participants earned £10.20. This is approximately £7.75/hour (~$9.63/hour). The US Federal minimum wage is $7.25/hour and the average state/territory minimum wage (after removing states/territories with no minimum wage) is $9.24/hour. Participants continued participation was incentivized by offering bonuses. For every wave completed (26 possibilities in total), participants received a virtual ticket. After 6 months and then again at the end of the study, we conducted a drawing for five £10 bonuses and one £50 bonus. The total cost of the study was £5151.64.

Quality Control

Prolific data quality is high [8] and they have procedures in place to prevent bots and repeat participants [1]. Participants were paid fair rates given the length of the study and we included incentives for completing as many waves as possible.

In addition, we checked attention rates with attention checks at Wave 12 and 15. We observed high rates of attention. All participants passed the check in Wave 15. In Wave 12, 10 of 416 participants did not pass the check (4 participants had missing data on the attention check item and were not counted as passed nor failed).

We also checked the correspondence between people’s self-reported votes in the 2016 election given to Prolific and given to us. Correspondence was high (91% match). The largest mismatch was from people who told Prolific they had not voted for neither Trump nor Clinton, but who reported to us that they voted for either Clinton (n = 19) or Trump (n = 14). The remaining mismatches were largely people who told Prolific they had either voted for Trump or Clinton, but who reported to us that they voted for a different candidate besides Trump or Clinton (n = 2), did not vote (n = 9), or planned to vote, but forgot (n = 2). Only one participant reported to Prolific that they voted for Clinton and reported to us that they voted for Trump. No participants reported to Prolific that they voted for Trump and reported to us that they voted for Clinton. Given that fallibility of people’s memories, including for their vote choice, these results indicate a relatively high correspondence (e.g., 91% is higher than many reported estimates, e.g., [4])

Ethical issues

Ethical approval was granted by Tilburg University. Data were anonymized by removing potential identifying information, including IP addresses and open-ended responses.

(3) Dataset description

Object name

  • codebook1.html: codebook
  • yllanon.csv: cleaned data file in long format
  • matrix and demographics.R: for computing information in this data paper and creating the histogram
  • response rates.R: for computing information in this data paper
  • Materials – Qualtrics Format (folder): all materials in Qualtrics format
  • Materials – Word Format (folder): all materials in Word format

Data type

Primary data, codebook, materials, code

Format names and versions

.csv, .html, .R, .qsf, .docx

Data Collectors

Mark Brandt, designed study, collected data, Tilburg University

Felicity Turner-Zwinkels, designed study, Tilburg University

Language

English

License

CC-by

Publication date

01/05/2020

(4) Reuse potential

This data can be reused to test hypotheses related to political attitudes and their change overtime. This could include, for example, if political identification is an antecedent or cause of social distance from political groups, if perceived threat is associated with right-wing or left-wing attitudes, and how stable political attitudes are overtime. The items regarding feelings of threat and anxiety may also be of use to researchers studying these topics. We purposely included items and constructs that have been the focus of investigations in the past, making it possible to conceptually replicate such findings. Lastly, given that there are relatively few datasets with such a longitudinal structure, it may also serve as an interesting teaching tool.