Questions for the 91女神/YouGov Poll can be broken down into three overlapping categories.
First, there is a 鈥渃ore鈥 set of questions that will be asked on every survey in order shed light on how Missourians鈥 opinions changed over time (e.g. their perceptions of the economy or crime).
Second, there are questions that directly contribute to 91女神 faculty鈥檚 research agendas.
Third, there are questions aimed to be relevant to current events or the immediate needs of the Missouri community.
In recent polls, this includes questions on the COVID-19 pandemic and race relations. To build these sets of questions, the 91女神 Poll directors actively seek input and feedback from across 91女神 and encourage the 91女神 community to approach the directors with suggestions.
The 91女神 Poll team welcomes recommendations for questions to ask on future polls. To suggest a question, contact 91女神 Poll Director Dr. Steven Rogers. Please note, space may be limited on polls.
Within six months of each survey, individual level responses will be posted to the 91女神 Poll web site 91女神 Poll web site data archive.
Respondents to the 91女神 Poll are self-reported likely voters in the state of Missouri drawn from a sample of respondents recruited by YouGov. Most YouGov panelists are individuals who have made an account on yougov.com to receive periodic notifications about new surveys. Others are recruited live from online advertisements or other survey providers. YouGov respondents are compensated by points for taking each survey. Respondents can exchange accumulated points for gift cards and other prizes.
For the July 2021 survey, YouGov conducted a benchmarking survey of Missouri likely voters using a census click methodology, which involves sampling in such a way that all starts to a particular survey are representative of the known demographic characteristics of a specific target group. In this case, YouGov sampled so all starts were representative of Missouri registered voters population, as defined by the 2020 Current Population Survey (CPS) November. The sample of all starts was matched from 1125 respondents to 950 completes and weighted to a population frame on age, gender, race and education.
YouGov used propensity score weighting. In the propensity score weighting process, the matched cases and frame were combined and a logistic regression was estimated for inclusion in the frame. The propensity score function included age, gender, race/ethnicity and years of education. The propensity scores were grouped into deciles of the estimated propensity score in the frame and post-stratified according to these deciles.
The weights were then post-stratified on 2016 and 2020 presidential vote choice, and a four-way stratification of gender, age ( four categories), race (four categories) and education (four categories). The weights were then subsetted on completes and re-centered to produce the final weight for the target population.
In 2020, 91女神/YouGov polls were funded by a grant from the Saint Louis University Research Institute鈥檚 Big Idea鈥檚 competition. The 2021 91女神/YouGov Poll was additionally funded through support from the 91女神 School of Education. The 91女神/YouGov Poll also receives support from individual 91女神 researchers who lease time on the survey instrument for their independent research projects.