Survey Methodology: PPBS 2016



The following describes the survey methodology of the 2016 iteration of The Psychology Political Behavior Studies (PPBS). This report is meant to detail the utilized survey methodology and relevant, general purpose characteristics of both exploratory and confirmatory samples, as well as any extra samples and its combined version. The text is in APA Style and is akin to an APA-like ‘Methods Section’.

PPBS datasets are designed to have an exploratory, quota-based Nationally Representative sample (on Age, Education, Income and Sex), and a Confirmatory (Replication) convenience sample, from the same data source to avoid false positives. Some PPBS studies also have recontacts or extra-samples to answer additional specific research questions (e.g., 2nd Sample in 2016). All data’s coodbooks can be found at the bottom of the page, with a searchable feature allowing to find metadata within and between columns.

Taken together, there were three independent samples collected in the months preceding the 2016 U.S. Presidential Election (from August 16th to September 19th, 2016) and amount to 4039 online interviews of American adults, lasting ~55 minutes (Median) for those successfully passing the variety of data quality checks (87%). The 2016 attrition rate was pretty constant at ~22%.

For more details, see PBBS’s Motivation.


Metadata of PBBS 2016 Samples
Sample 1 Sample 2 Sample 3
Election cycle Pre-Election 2016 Pre-Election 2016 Pre-Election 2016
Type Nationally Representative Confirmatory (Replication) Extra (2nd Study)
Survey Period (Start) August 16th, 2016 August 20th, 2016 September 15th, 2016
Survey Period (End) September 16th, 2016 September 13th, 2016 September 19th, 2016
Country United States United States United States
Attention Checks Yes (11; various types) Yes (11; various types) Yes (11; various types)
Time Checks Yes Yes Yes
CAPTCHA Yes (begining of survey) Yes (begining of survey) Yes (begining of survey)
Sample Size (N) 1500 2119 420
Length of Interview (MD) 51.28 57.77 55.54
Attrition (%) 0.22 0.22 0.23
Data Quality checks (%) 0.16 0.16 0.07
Note. The Overall column displays the simple, non-weighted on sample size, average between the preceeding columns. If interested in the weighted averages, see section 'Combined Samples' below.


Nationally Representative Sample (N=1500)

Sample Description


We hired Survey Sampling Incorporated (SSI; www.surveysampling.com), a market research firm that recruits participants from a panel of over 7 million U.S. citizens, to recruit a nationally representative sample of 1500 Americans (50.67% women) in the months preceding the 2016 U.S. Presidential Election (from August 16 to September 16, 2016).

The quotas were designed to match the 2014 American Community Survey (ACS)1 from the US Census Bureau on age, income, education and gender. The representativeness of the collected sample is presented in the table below, which shows an average absolute deviation of 11.85% points (MD = 8.03% points) from the desired quotas, with age showing an an average absolute deviation of 0.85% points (MD = 0.72% points), gender 1.73% points (MD = 1.73% points), income 14.27% points (MD = 12.44% points), and education, 27.32% points (MD = 13.12% points), which had the largest spread on the bracket “Less than high school/No high school diploma”, for which we could only arrive at ~25% of the expected frequency, indicating the sample has achieved a moderate level of national representativeness.

In addition to administering a much greater number and variety of political and psychological instruments (including full scales) than in other nationally representative surveys (such as ANES, GSS, and WVS), we took a number of steps to insure that the quality of the data would be especially high. These included following professional recommendations to minimize problems of careless responding and satisficing behavior in online survey studies (Meade & Craig, 2012). Specifically, we employed 11 random attention questions, as well as page-time, survey-total, and click count controls. A total of 1500 participants were directed to the survey, and 1500 of them finished the survey (attrition rate 0%). There were 0 (0%) participants who failed more than two attention checks or finished the survey in under ~22 minutes and were therefore excluded. For the final sample of 1500, participants who successfully completed all study materials had a completion time of 69.29 minutes on average (MD: 51.28min).

The age distribution of our sample was as follows: 18–24 years (12.87%), 25–34 (17.6%), 35–44 (17.53%), 45–54 (19.47%), 55–65 (15.6%), and older than 65 (16.93%). The ethnic breakdown was: White (82.47%), Black/African American (7.67%), Latino (5.87%), Asian/Pacific Islander (1.93%), Native American (0.87%), Middle Eastern (1.2%). In terms of religion, (67.6% identified as Christian, 0.6% as Muslim, 3.47% as Jewish,15.33% as either Atheist or Agnostic, 13% and responded they are not sure, religion not listed, or refused to answer. With respect to education, 3.4% declared their highest educational achievement to be less high school, 31.67% indicated High school graduate (includes equivalency), 31.4% chose some college ( or associate degree) or no higher-education degree, 20.67% indicated having a Bachelor’s degree, and finally, 12.87% indicated having received a Graduate or professional degree. The median income category was $35,000 to $49,999. The exact distribution of Income is as follows: Less $15,000 (11.87%), $15,000 to $24,999 (12%), $25,000 to $34,999 (11.73%), $35,000 to $49,999 (15.13%), $50,000 to $74,999 (19.47%), $75,000 to $99,999 (12.8%), $100,000 to $149,999 (10.67%) and $150,000 more (6.33%).


National Representativeness of sample
Demographic Brackets Census ACS % Expected Sample Frequencies Observed Frequencies Expected vs. Observed Frequencies Expected vs. Observed % Expected vs. Observed % [Census 2010]*
Age 18 to 24 0.13 192 193 1 0.52 -1.53
Age 25 to 34 0.18 269 264 -5 -1.86 0.38
Age 35 to 44 0.18 265 263 -2 -0.75 0.00
Age 45 to 54 0.19 290 292 2 0.69 1.39
Age 55 to 64 0.15 232 234 2 0.86 0.43
Age 65 to 80+ 0.17 253 254 1 0.40 -1.55
Education Less than high school 0.14 204 51 -153 -75.00 -75.71
Education High school graduate (includes equivalency) 0.28 420 475 55 13.10 9.20
Education Some college (associate degree) or no degree 0.29 436 471 35 8.03 8.28
Education Bachelor's degree 0.18 274 310 36 13.14 14.81
Education Graduate or professional degree 0.11 165 193 28 16.97 28.67
Gender Female 0.52 773 760 -13 -1.68 2.98
Gender Male 0.48 727 740 13 1.79 -2.89
Income Less than $15,000 0.12 188 178 -10 -5.32 -8.72
Income $15,000 to $24,999 0.11 160 180 20 12.50 0.00
Income $25,000 to $34,999 0.10 153 176 23 15.03 6.67
Income $35,000 to $49,999 0.14 202 227 25 12.38 8.10
Income $50,000 to $74,999 0.18 267 292 25 9.36 8.15
Income $75,000 to $99,999 0.12 183 192 9 4.92 6.67
Income $100,000 to $149,999 0.13 195 160 -35 -17.95 -11.11
Income $150,000 or more 0.10 150 95 -55 -36.67 -20.83

1 Note. An important note is that the US Census Bureau makes minor adjustments to its data over time and is not averse to closing tables and/or redesigning them. This is not particular of the US. Hence, the numbers were correct at date of access, which was Monday 1st August, 2016. Furthermore, due to this, we also provide representation in relation to the 2010 US Census, which is shown in the last column. Specifically, we take Gender and Age from U.S. Census Bureau, 2010 Census and Income and Education from 2010-2014 American Community Survey 5-Year Estimates, which at the time, were the most accurate data available.



Regional Representation


As shown below, the distribution of data points per state tracks well with state population. The date has not been designed to be regionally representative, nor it claims to be, but results are not bad (cf. US Decennial Census Tables). Please note that frequencies add to 1389, not to 1500, due to non-responses.

Regional Representation by US State, PPBS 2016 (N = 1500)
+−
0
50
100
150
200
250
Leaflet | © OpenStreetMap contributors © CARTO



Convenience Confirmatory (Replication) Sample (N=2119)

Sample Description


Also through SSI, we also administered the same survey to a large convenience sample of 2119 American adults (21.47% women) in the months preceding the 2016 U.S. Presidential Election (from August 20 to September 13, 2016).

We applied the same quality-control criteria as explained in the Nationally Representative sample. Specifically, we followed recommendations to minimize the problem of careless responding in online studies (Meade & Craig, 2012). A total of 2119 participants were directed to the survey, and 2119 of them finished the survey (attrition rate 0%). There were 0 (0%) participants who failed more than two attention checks or finished the survey in under ~22 minutes and were therefore excluded. For the final sample of 2119, participants who successfully completed all study materials had a completion time of 92.01 minutes on average (MD: 57.77min).

The age distribution of our sample was as follows: 18–24 years (9.06%), 25–34 (13.83%), 35–44 (11.42%), 45–54 (2.74%), 55–65 (3.63%), and older than 65 (59.32%). The ethnic breakdown was: White (85.89%), Black/African American (5.05%), Latino (4.11%), Asian/Pacific Islander (2.17%), Native American (0.94%), Middle Eastern (1.84%). In terms of religion, (70.65% identified as Christian, 0.52% as Muslim, 5.57% as Jewish,13.69% as either Atheist or Agnostic, 9.58% and responded they are not sure, religion not listed, or refused to answer. With respect to education, 1.04% declared their highest educational achievement to be less high school, 15.15% indicated High school graduate (includes equivalency), 40.4% chose some college ( or associate degree) or no higher-education degree, 23.88% indicated having a Bachelor’s degree, and finally, 19.54% indicated having received a Graduate or professional degree. The median income category was $50,000 to $74,999. The median income category was $50,000 to $74,999. The exact distribution of Income is as follows: Less $15,000 (10.15%), $15,000 to $24,999 (8.07%), $25,000 to $34,999 (11.09%), $35,000 to $49,999 (14.39%), $50,000 to $74,999 (21.24%), $75,000 to $99,999 (14.91%), $100,000 to $149,999 (11.89%) and $150,000 more (11.89%).


Combined Samples (N=3619)

Sample Description


We combined and analyzed data from two large surveys conducted before the 2016 U.S. General Election (from August 16 to September 16, 2016), including a nationally representative sample (N = 1500) and a large convenience sample (N = 2119; see Azevedo, Jost, & Rothmund, 2017; Azevedo, Jost, Rothmund, & Sterling, 2019). We hired SSI (SSI; www.surveysampling.com), a survey research firm that recruits participants from a pool of over 7 million U.S. citizens. We took a number of steps to insure that the quality of the data would be especially high. These included following professional recommendations to minimize problems of careless responding and satisficing behavior in online survey studies (Meade & Craig, 2012). Specifically, we employed 11 random attention questions, page-time and number of clicks controls. A total of 3619 participants were directed to the survey, and 3619 of them finished the survey (attrition rate 0%). There were 0 (0%) participants who failed more than two attention checks or finished the survey in under ~22 minutes and were therefore excluded. For the final sample of 3619, participants who successfully completed all study materials had a completion time of 82.6 minutes on average (MD: 55.15min).

The age distribution of our sample was as follows: 18–24 years (10.64%), 25–34 (15.39%), 35–44 (13.95%), 45–54 (9.67%), 55–65 (8.59%), and older than 65 (41.75%). The ethnic breakdown was: White (84.47%), Black/African American (6.13%), Latino (4.84%), Asian/Pacific Islander (2.07%), Native American (0.91%), Middle Eastern (1.58%). In terms of religion, (69.38% identified as Christian, 0.55% as Muslim, 4.7% as Jewish,14.37% as either Atheist or Agnostic, 11% and responded they are not sure, religion not listed, or refused to answer. With respect to education, 2.02% declared their highest educational achievement to be less high school, 22% indicated High school graduate (includes equivalency), 36.67% chose some college ( or associate degree) or no higher-education degree, 22.55% indicated having a Bachelor’s degree, and finally, 16.77% indicated having received a Graduate or professional degree. The median income category was $50,000 to $74,999. The exact distribution of Income is as follows: Less $15,000 (10.86%), $15,000 to $24,999 (9.7%), $25,000 to $34,999 (11.36%), $35,000 to $49,999 (14.7%), $50,000 to $74,999 (20.5%), $75,000 to $99,999 (14.04%), $100,000 to $149,999 (11.38%) and $150,000 more (11.38%).


2nd Study (N=420)

Sample Description


Also through SSI, we also administered the same survey to a large convenience sample of 420 American adults (61.43% women) in the months preceding the 2016 US Presidential Election (from September 15 to September 19, 2016).

This study is characterized by having the same questions as PPBS2016 samples, with the inclusion of four additional instruments: Stealth Democracy, Direct Democracy Attitudes and Citizen Participation items, as well as additional Conspiracy theories, all of which had been cut from the survey instrument due to concerns over the length of interview.

We took a number of steps to insure that the quality of the data would be especially high. These included following professional recommendations to minimize problems of careless responding and satisficing behavior in online survey studies (Meade & Craig, 2012). Specifically, we employed 11 random attention questions, as well as page-time, survey-total, and click count controls. A total of 420 participants were directed to the survey, and 420 of them finished the survey (attrition rate 0%). There were 0 (0%) participants who failed more than two attention checks or finished the survey in under ~22 minutes and were therefore excluded. For the final sample of 420, participants who successfully completed all study materials had a completion time of 90.8 minutes on average (MD: 55.54min).

The age distribution of our sample was as follows: 18–24 years (4.29%), 25–34 (16.43%), 35–44 (20%), 45–54 (20%), 55–65 (38.33%), and older than 65 (0.95%). The ethnic breakdown was: White (82.62%), Black/African American (9.29%), Latino (3.33%), Asian/Pacific Islander (2.62%), Native American (0.71%), Middle Eastern (1.43%). In terms of religion, (69.29% identified as Christian, 0.24% as Muslim, 5.71% as Jewish,14.29% as either Atheist or Agnostic, 10.48% and responded they are not sure, religion not listed, or refused to answer. With respect to education, 4.29% declared their highest educational achievement to be less high school, 11.43% indicated High school graduate (includes equivalency), 24.52% chose some college ( or associate degree) or no higher-education degree, 41.19% indicated having a Bachelor’s degree, and finally, 18.57% indicated having received a Graduate or professional degree. The median income category was $50,000 to $74,999. The exact distribution of Income is as follows: Less $15,000 (9.29%), $15,000 to $24,999 (9.05%), $25,000 to $34,999 (10%), $35,000 to $49,999 (12.14%), $50,000 to $74,999 (21.67%), $75,000 to $99,999 (14.52%), $100,000 to $149,999 (16.43%) and $150,000 more (6.9%).



Codebook

Search box on the right finds items in all columns. Search boxes at the column level finds matches in their specific column.


PPBS Head Logo