Associations of Mail Survey Length and Layout With Response Rates

We assess the association between survey layout and response rates (RRs) in the 2017 Medicare Advantage Consumer Assessment of Healthcare Providers and Systems mail survey. Among 438 Medicare Advantage plans surveyed by six vendors, there was latitude in survey layout, and plans could add up to 12 s...

Full description

Saved in:
Bibliographic Details
Published in:Medical care research and review 2021-08, Vol.78 (4), p.441-448
Main Authors: Burkhart, Q, Orr, Nate, Brown, Julie A., Hays, Ron D., Cleary, Paul D., Beckett, Megan K., Perry, Suzanne E., Gaillot, Sarah, Elliott, Marc N.
Format: Article
Language:English
Subjects:
Age
Beneficiaries
Health care industry
Layout
Layouts
Mail surveys
Medicare
Polls & surveys
Regression analysis
Regression models
Response rates
Sociodemographics
Vendors
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We assess the association between survey layout and response rates (RRs) in the 2017 Medicare Advantage Consumer Assessment of Healthcare Providers and Systems mail survey. Among 438 Medicare Advantage plans surveyed by six vendors, there was latitude in survey layout, and plans could add up to 12 supplemental items. Regression models predicted survey response from survey characteristics (page count, number of supplemental items, and survey attractiveness), and beneficiary sociodemographics. Beneficiary-age-by-survey-characteristic interactions assessed whether survey characteristics were more strongly related to RRs among older beneficiaries. We found that surveys with more supplemental items and less attractive layouts had lower adjusted odds of response. RRs were more sensitive to format among older beneficiaries. The difference in adjusted RRs for the most favorable versus the least favorable survey design was 14.5%. For a 65-year-old, this difference was 13.6%; for an 80-year-old, it was 21.0%. These findings suggest that even within a relatively standardized survey, formatting can substantially influence RRs.
ISSN:1077-5587
1552-6801
DOI:10.1177/1077558719888407