Benefits of the Lion Poll Methodology

The Center for Survey Research previously conducted the Penn State Poll, which utilized a dual frame (cell phone and landline) methodology to reach 600 adult Pennsylvanians.  Why change the data collection strategy from telephone to a web panel?

 
People are not answering their phones.
The proportion of cases in which someone in the household is reached continues to fall.  For example, in fall 2012, 14.4% of our calls resulted in speaking to a person compared to only 4.5% in fall 2017.  These statistics demonstrate that telephone surveys have been experiencing unprecedented levels of non-response.  With more and more potential respondents opting to not answer their phones, non-response bias remains a growing problem.  
 
Get responses quickly!
This is an improvement to our methodology.  Clients can obtain their data much faster using this online data collection method, which allows you to get a snapshot of a specific (shorter) amount of time.  In other words, there is a smaller likelihood that outside factors (i.e., events) could occur during this smaller fielding period.
 
Obtain a larger sample size - only $1 per completed survey!
The Lion Poll gathers 1,000 completed surveys, compared to the Penn State Poll, which only collected 600 completed surveys.
 
No weighting necessary
We utilize quotas to ensure that the results will be representative by age/sex and separately by region, which means that we do not have to weight the data.  This reduces error in the analysis process.
 
Lots of demographic data
The difference in survey mode allows us to provide five additional demographic questions as part of the per question cost.  The following demographics are included in the Lion Poll dataset: age, sex, ethnicity, race, education, income, political affiliation, political ideology, geography (zip code, county, region, rural/urban status), number of children in the household, number of adults in the household, and household size.
 
High quality results
We completed two bridge surveys to identify the impact of changing modes.  Because age/sex and region quotas are used, the Lion Poll matches the Census data.  Also, the Lion Poll gets us better representation from those with lower education (high school diploma or less).  Further, we found that results from the survey questions were within the margin of error.  Learn more about how Penn State Poll results compare to Lion Poll results and Census data (opens in a new window). 
 
The Center for Survey Research utilizes a variety of strategies to ensure that our datasets contain high quality respondents.  First, we utilize screening methods that do not reveal the type of respondents that we are looking to survey.  For example, instead of confirming that they are a Pennsylvania resident, we ask them to select their state from a list of eight states.  Next, we build in checks to identify bots (short for "robot") and people who do not read the questions.  If respondents do not meet our eligibility requirements, appropriately answer quality check questions, etc., they are removed from the survey and are not counted towards the 1,000 completed interviews. 
 
The Lion Poll is a self-administered survey, which means that there are no trained interviewers to probe or clarify responses.  However, this methodology sees better response rates when surveying about sensitive topics because respondents do not have to verbalize their answers to another person. 
 
Flexibility with the survey's geography and/or target population
The Lion Poll utilizes a statewide sample of adult Pennsylvanians (18 years of age or older).  However, this methodology provides opportunities to poll nationwide or other geographies - or even specific subgroups (race, ethnicity, parents, specific age ranges, etc.). Availability of survey participants determines whether the option is viable. Contact us csr@psu.edu to inquire about this option.