Name File Type Size Last Modified
TreasureState2020_Codebook_PublicRelease_FINAL_21Jan2022.pdf application/pdf 551.8 KB 02/04/2022 11:41:AM
TreasureState2020_Data_PublicRelease_FINAL_21Jan2022.dta application/x-stata-dta 485.4 KB 01/21/2022 01:06:PM
TreasureState2020_Data_PublicRelease_FINAL_21Jan2022.xlsx application/vnd.openxmlformats-officedocument.spreadsheetml.sheet 1.3 MB 02/05/2022 11:09:AM
TreasureState2020_Frequencies_PublicRelease_FINAL_21Jan2022.pdf application/pdf 10.9 MB 01/21/2022 01:11:PM
TreasureState2020_Methods_PublicRelease_FINAL_04Feb2022.pdf application/pdf 118.7 KB 01/28/2022 11:02:AM

Project Citation: 

Parker, David C. W., Raile, Eric D., and Shanahan, Elizabeth A. 2020 Treasure State (Montana) Election Surveys. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2022-02-05.

Project Description

Summary:  View help for Summary The Treasure State 2020 Pre-Election Survey was conducted by the HELPS Lab of Montana State University-Bozeman between September 14 and October 2, 2020. The original population for the survey was Montana voters who registered by August 20, 2020, and were deemed active by the Montana Secretary of State. We stratified the sample by state house districts, and then drew a random sample of 9,000 voters proportionally from these strata. Sampled individuals received a questionnaire by mail and were asked to return the questionnaire via a prepaid business reply envelope. Respondents returned 1,789 surveys, a response rate of 20.2% based on 8,836 deliverable addresses. The Treasure State 2020 Post-Election Survey, also conducted by the HELPS Lab, used a panel design that returned to the individuals who responded to the Pre-Election Survey. This survey was in the field from November 4-25, 2020. Of the 1,789 original respondents, 1,069 completed the post-election questionnaire, for a response rate of 59.8%. The procedures for the Post-Election Survey otherwise paralleled those for the Pre-Election Survey.

The number of individuals voting in the 2020 general election in Montana was 612,075, according to information from the Montana Secretary of State. We have adjusted the overall survey margin of error (MoE) for design effect owing to our subsequent use of weights in calculating results. This adjustment has the consequence of making the margin of error substantially larger than it otherwise would be. For the pre-election sample, this results in a MoE of +/- 3.7 percentage points, using a confidence level of 95% and a response distribution of 50% (the most conservative estimate) with the given population size. For the post-election sample, the corresponding MoE is +/- 6.0 percentage points. We note that these figures are overall approximations given sampling stratification and different response options and response patterns for specific questions.

This project was funded by over 100 individual donors to the Montana State University Alumni Foundation and by Montana State University.
Funding Sources:  View help for Funding Sources Montana State University

Scope of Project

Subject Terms:  View help for Subject Terms Elections; US Congress; Montana; Policies; Politics
Geographic Coverage:  View help for Geographic Coverage Montana
Time Period(s):  View help for Time Period(s) 9/14/2020 – 10/2/2020; 11/4/2020 – 11/25/2020
Collection Date(s):  View help for Collection Date(s) 9/14/2020 – 10/2/2020; 11/4/2020 – 11/25/2020
Universe:  View help for Universe Montana voters in 2020 general election
Data Type(s):  View help for Data Type(s) survey data
Collection Notes:  View help for Collection Notes Data collected via mail surveys in two waves -- one prior to the election and another following the election


Response Rate:  View help for Response Rate Pre-election survey:  20.2%
Post-election survey (panel):  59.8%
Sampling:  View help for Sampling The original universe for the poll was Montana voters who registered by August 20, 2020, and were deemed active by the Montana Secretary of State. For the pre-election survey, the sample was stratified by state House districts, and then a random sample of 9,000 voters was drawn proportionally from these strata. For the post-election survey, researchers used a panel design, meaning they re-surveyed individuals who had completed the pre-election questionnaire. The data posted here have been filtered and weighted to represent individuals who actually voted in the 2020 general election in Montana. Therefore, the final universe is actual Montana 2020 voters.
Data Source:  View help for Data Source Surveys of Montana registered voters
Collection Mode(s):  View help for Collection Mode(s) mail questionnaire
Weights:  View help for Weights To mitigate non-response bias in the sample, the researchers chose to weight the sample using iterative proportional fitting, or raking. Raking generates weights that adjust the sample in subsequent analyses so that the sample more closely resembles the target population, in this case individuals who voted in the 2020 Montana elections. Weights are generated by forcing sample margins to approximate population margins for key demographic characteristics. The researchers generated weights for individuals using age, education, gender, media market, self-reported 2020 presidential vote choice, and self-reported vote on the state marijuana measure I-190. Population margins for education and gender were obtained using data from the U.S. Census Bureau’s 2018 Current Population Survey Voter Supplement. Population margins for 2020 presidential vote choice and vote on the state marijuana measure were based on vote returns retrieved from the Montana Secretary of State’s website. Population margins for media market and age came from public data also available from the Montana Secretary of State.

The pre-election weight used in the current data release differs slightly from the weight used by the investigators prior to the 2020 election. Consequently, the full set of results differs from earlier-published results in minor ways. The updated weight incorporates better information about “likely” voters by including individuals who actually voted in the election. Therefore, all results are meant to represent the views of Montana voters. Survey respondents were filtered prior to the creation of survey weights to retain only respondents who actually voted in the 2020 general election. This left 1,765 respondents for the pre-election survey and 1,066 respondents for the post-election survey. The weights used in the presented frequencies and marginal percentages are prewgt2 and postwgt2.
Unit(s) of Observation:  View help for Unit(s) of Observation Individual persons
Geographic Unit:  View help for Geographic Unit State

Related Publications

Published Versions

Export Metadata

Report a Problem

Found a serious problem with the data, such as disclosure risk or copyrighted content? Let us know.

This material is distributed exactly as it arrived from the data depositor. ICPSR has not checked or processed this material. Users should consult the investigator(s) if further information is desired.