View profile

Franklin & Marshall Poll Release: October 2021

Issue #15 • View online
Franklin & Marshall College Poll
Dear Subscribers,
Today’s newsletter contains information from the October 2021 Franklin & Marshall College Poll, conducted October 18-24 among a sample of 522 registered Pennsylvania voters. The survey asked about a wide range of topics, including voters’ general feelings about the direction of the state and economy, support for another audit of the 2020 election, COVID-19 mitigation efforts, and job approval ratings for President Biden and Governor Wolf. You can see the full set of questions asked as well as the overall responses to each question in the attached topline summary. I’ve highlighted a few of the notable findings in today’s newsletter and you can find many more details in our Summary of Findings.
Sincerely,
Berwood Yost, Director

Key Findings
The October 2021 Franklin & Marshall College Poll finds President Biden’s job approval ratings and personal popularity have declined in Pennsylvania since August. About one in three (32%) registered voters in Pennsylvania believes President Biden is doing an “excellent” or “good” job as president, which is a sizable decline from his June rating of 44 percent and his August rating of 41 percent. Nearly two in five (39%) voters say their feelings about President Biden’s job performance have changed over the past few months, with most (90%) of these voters feeling the President is performing “worse” today. Voters who feel the President is doing “worse” most often say that the Afghanistan withdrawal (31%), the border crisis and immigration (15%), the economy and inflation (15%), and COVID-19 (14%) are the reasons for their change of mind.
Comparatively, the President’s positive approval ratings among Democrats and independent voters are well below the Governor’s positive ratings among those groups. Governor Wolf’s positive job approval rating is 42 percent, consistent with prior surveys.
Concerns about government and politicians (20%) and the economy (17%), including unemployment and personal finances, are the most important problems facing Pennsylvania right now, according to the state’s registered voters. Pennsylvania voters’ optimism about conditions in the state has declined since August and there is also no notable improvement in voters’ assessments of their personal finances.
A majority (55%) of registered voters opposes the state legislature’s plans to conduct a forensic investigation of the 2020 election. Of course, there are significant partisan and ideological differences related to support for the investigation, with Democrats (78%), independents (64%), liberals (88%), and moderates (69%) opposed and Republicans (70%) and conservatives (76%) in support. 
More voters favor than oppose requiring masks in public schools (64%), employers requiring COVID-19 vaccinations of their employees (56%), and businesses requiring customers to provide proof of a vaccination (51%). Despite each measure having majority support, each also has a relatively sizable group that “strongly opposes” it. For example, Republicans (41%) and conservatives (44%) strongly oppose mask mandates in public schools while Democrats (71%) and liberals (86%) strongly favor them.
Methodology
The survey findings presented in this release are based on the results of interviews conducted October 18 - 24, 2021. The interviews were conducted at the Center for Opinion Research at Franklin & Marshall. The data included in this release represent the responses of 522 registered Pennsylvania voters, including 244 Democrats, 205 Republicans, and 73 independents.[i] The sample of voters was obtained from Marketing Systems Group. All sampled respondents were notified by mail about the survey. Interviews were completed over the phone and online depending on each respondent’s preference. Survey results were weighted (age, gender, education, geography, and party registration) using an iterative weighting algorithm to reflect the known distribution of those characteristics. Estimates for age, geography, and party registration are based on active voters within the PA Department of State’s voter registration data. Gender and education is estimated using data from the November 2018 CPS Voter Registration Supplement.[ii]
The sample error for this survey is +/- 5.2 percentage points when the design effects from weighting are considered. [iii] In addition to sampling error, this poll is also subject to other sources of non-sampling error. Generally speaking, two sources of error concern researchers most. Non-response bias is created when selected participants either choose not to participate in the survey or are unavailable for interviewing. Response errors are the product of the question and answer process. Surveys that rely on self-reported behaviors and attitudes are susceptible to biases related to the way respondents process and respond to survey questions.
References & Resources
[i] The question reported here is self-reported voter REGISTRATION and is consistent with past reporting practices. The survey also asked about self-reported voter IDENTIFICATION, which shows a slightly different partisan split: 45% identify as Republican (n=222), 10% as Independent (n=52), and 41% as Democrat (n=201), with the balance not offering a response. Partisan comparisons in this summary show self-reported registration.
[ii] Data downloaded from IPUMS-CPS, University of Minnesota, www.ipums.org, accessed 12/31/2019.
[iii] An alternative means of calculating the variation in a sample is to take a series of bootstrap samples from the original sample and to use those bootstrapped samples to produce an estimate of sampling error (see Canty, Angelo. 2002. “Resampling Methods in R: The boot Package.” R News 2/3 (December): 2-7). The procedure involves resampling a data set, calculating a statistic for each bootstrapped sample, accumulating the results of these samples and calculating a sample distribution. The standard deviation of the mean of 10,000 bootstrapped samples for the estimated positive job approval for Governor Wolf was 2.2% and 95% of the samples fell within a range of 39% and 48%.
Did you enjoy this issue?