Methodology › 2015 Methodology

2015 Methodology Download PDF

The 2015 American Values Atlas (AVA) is a project of Public Religion Research Institute (PRRI). Results for all demographic, religious affiliation, and political affiliation questions were based on 81,970 bilingual telephone interviews (including 40,968 cell phone interviews) conducted between January 7, 2015 and January 7, 2016 by professional interviewers under the direction of SSRS. Results for questions on specific issues (e.g. immigration and LGBT issues) are based on a subset of 42,586 telephone interviews conducted between April 29, 2015 and January 7, 2016. The AVA was made possible by generous grants from The Ford Foundation, The Carnegie Corporation of New York, the Arcus Foundation, the Gill Foundation and the Nathan Cummings Foundation.

Throughout 2015, at least 1,000 interviews were completed each week, with about 500 interviews conducted among respondents on their cell phones. Each week, interviewing occurred over a five-day period, from Wednesday through Sunday or from Thursday through Monday. The selection of respondents within households was accomplished by randomly requesting to speak with the youngest adult male or female currently living in the household.

Data collection was based on stratified, single-stage, random-digit-dialing (RDD) of landline telephone households and randomly generated cell phone numbers. The sample was designed to represent the total U.S. adult population from all 50 states, including Hawaii and Alaska. The landline and cell phone samples were provided by Marketing Systems Group.

The weighting was accomplished in two separate stages. The first stage of weighting corrects for different probabilities of selection associated with the number of adults in each household and each respondent’s telephone usage patterns. In the second stage, sample demographics were balanced to match target population parameters for gender, age, education, race and Hispanic ethnicity, region (U.S. Census definitions), population density, and telephone usage. The population density parameter was derived from 2010 Census data. The telephone usage parameter came from an analysis of the July-December 2014 National Health Interview Survey. All other weighting parameters were derived from an analysis of the U.S. Census Bureau’s March 2015 Current Population Survey.

The sample weighting was accomplished using iterative proportional fitting (IFP), a process that simultaneously balances the distributions of all variables. Weights are trimmed so that they do not exceed 4.0 or fall below 0.25 to prevent individual interviews from having too much influence on the final results. The use of these weights in statistical analysis ensures that the demographic characteristics of the sample closely approximate the demographic characteristics of the target populations.

The table below shows the sample sizes for all 50 states and 30 metro areas for both the total sample and issue subsample. The margin of error for total sample is +/- 0.4 percentage points at the 95% level of confidence. The design effect for total sample is 1.4. In addition to sampling error, surveys may also be subject to error or bias due to question wording, context, and order effects.

Sample Sizes for States and Metro Areas

State Total Sample Issue Subsample
United States 81,970 42,586
Alabama 1,308 657
Alaska 710 379
Arizona 1,560 829
Arkansas 782 407
California 7,671 3,926
Colorado 1,346 712
Connecticut 872 475
Delaware 239 148
Florida 4,917 2,572
Georgia 2,393 1,240
Hawaii 407 202
Idaho 471 230
Illinois 2,936 1,514
Indiana 1,938 978
Iowa 1,103 587
Kansas 876 465
Kentucky 1,289 690
Louisiana 1,170 602
Maine 460 226
Maryland 1,349 721
Massachusetts 1,521 750
Michigan 2,379 1,225
Minnesota 1,496 769
Mississippi 753 423
Missouri 1,761 894
State Total Sample Issue Subsample
Montana 465 244
Nebraska 587 331
Nevada 690 352
New Hampshire 369 189
New Jersey 2,184 1,048
New Mexico 589 307
New York 5,515 2,813
North Carolina 2,855 1,477
North Dakota 276 155
Ohio 3,349 1,773
Oklahoma 1,038 557
Oregon 1,296 694
Pennsylvania 3,753 1,924
Rhode Island 279 154
South Carolina 1,327 702
South Dakota 278 155
Tennessee 1,895 979
Texas 5,342 2,782
Utah 813 427
Vermont 220 148
Virginia 2,260 1,168
Washington 1,923 1,027
West Virginia 640 311
Wisconsin 1,900 986
Wyoming 213 139
Metro Area Total Sample Issue Subsample
United States 81,970 42,586
Atlanta 1,240 621
Boston 1,041 515
Charlotte 631 309
Chicago 2,000 1,014
Cincinnati 615 355
Cleveland 549 286
Columbus 531 275
Dallas 1,280 685
Denver 649 350
Detroit 965 495
Houston 1,177 632
Indianapolis 503 256
Kansas City 560 295
Las Vegas 467 247
Los Angeles 2,287 1,152
Miami 1,207 633
Milwaukee 478 243
Minneapolis-St. Paul 863 450
Nashville 433 241
New York 5,183 2,609
Orlando 519 273
Philadelphia 1,477 793
Phoenix 972 524
Pittsburgh 779 398
Portland 684 369
San Francisco 900 463
Seattle 791 415
St. Louis 740 386
Tampa-St. Petersburg 751 390
Washington, DC 1,421 759