Beyond question wording: How survey design and administration shape estimates of disability.
Heide Jackson, Natalie A E Young, Danielle Taylor
PMID 34154971WHAT IT FOUND
A sharp rise in disability estimates between 2011 and 2014 was largely due to the sampled population being in poorer health, not survey errors.
Data processing did not inflate the numbers, but a large portion of the increase remains unexplained.
Key findings
01Data processing methods like imputation and weighting did not cause the increase in disability estimates; they actually reduced the difference between panels.
02Changes in sample composition, particularly an increase in the number of adults reporting poor self-rated health, explained about a quarter of the rise in disability prevalence.
03Higher rates of functional limitations and ADL difficulties in the 2014 panel occurred even without priming questions, suggesting the rise in disability reflected actual health differences rather than survey context effects.
STILL TO COME
How it was doneWhat they found
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
The decomposition model could only include variables measured in both panels, so important factors like obesity were not included, meaning the true impact of sample characteristics may be larger. The analysis of context effects for functional limitations was restricted to adults 65 and older due to non-response issues in the younger group, so it does not confirm whether priming affected the 40-64 age group. The study cannot rule out other context effects that might have influenced both the standardized disability questions and the functional limitation modules. A large portion of the increase in disability estimates remains unexplained, so the causes are not fully identified.
Declared interests
The research was supported by the National Institutes of Health (N.I.H., Extramural).
The easy way to misread this
Do not assume that a sudden rise in disability prevalence in survey data means the population has become significantly more disabled or that the survey instrument is broken. In this case, the rise was largely due to the specific group sampled being in poorer health, not errors in how the questions were asked or processed.