Although we were gratified to achieve a 78 % retention rate in the study during 3 years, there remain significant differences in the demographic composition
of nonrespondents in the intervention group compared with the UC group, and within the randomized treatment groups as well.
I bet it's 50 % — which accounts for all but 3 %
of your nonrespondents!
Not exact matches
And those
nonrespondents are harder to find, thanks in part to economic disruptions and a growing population
of immigrants.
Finally, to account for the minor differences between respondents and
nonrespondents that we did observe, the test scores
of children who, based on their demographic characteristics, were more likely to attend follow - up sessions were weighted less heavily, while the test scores
of children who were less likely to attend follow - up sessions, but nevertheless did, were weighted more heavily.
Finally, toward the end
of the survey administration period, CAP mailed a hard copy version
of the survey — including a self - addressed, stamped envelope — to all
nonrespondents and asked that they return the survey to PSA as soon as possible.
In wave 1, the HAC questionnaire data were abstracted for respondents and
nonrespondents to the ACE Study questionnaire, enabling a detailed assessment
of the representativeness
of respondents in terms
of demographic characteristics and health - related issues.
Response rates did not differ by sex, education, cigarette smoking behavior, or history
of childhood sexual abuse as recorded in the clinic's medical record; however, respondents were slightly older (56.7 years vs 49.3 years) and more likely to be white (84 % vs 75 %) than were
nonrespondents.
Thus, there was no evidence that the general health
of respondents and
nonrespondents differed.
Comparisons between the respondents and
nonrespondents indicated no statistically significant differences in terms
of age, sex, and whether they attended city or rural schools.
Results
of this analysis have been published elsewhere.29 Briefly,
nonrespondents tended to be younger, less educated, or from racial or ethnic minority groups.
Although there was a lower response among families
of minority groups (70 %) than among white families (84 %), (p <.01), t - tests and chi - square analyses indicated that minority - group respondents did not differ from minority - group
nonrespondents on any birth status or sociodemographic variable (p >.05).
The adjustments reduce potential bias to the extent that the survey respondents and
nonrespondents (noncontacts, refusals, etc.) with similar geographic, demographic, and socioeconomic characteristics are also similar with respect to the survey statistics
of interest.