The prevalence of elder care responsibilities among the work force population: Response bias among a group of cross-sectional surveys

This review synthesizes the findings of 17 independent studies dealing with the prevalence of elder care responsibilities among the work force population. Across-study, summative findings were: (a) approximately one fifth (M = 21.1%) to one quarter (Md = 23.1%) of employees provide care for an elderly dependent; primary study findings varied by a factor of nearly 25, ranging from a high estimate of 46.0% to a low of 1.9%; (b) the average response rate was fairly low (M = 45.0% and Md = 41. 1%), indicating that the studies captured only slightly more than one third, but less than half of all eligible in-sample assignments; (c) the correlation of prevalence and response rates was found to be r = -.69, p \u3c .01; (d) the partial correlation of prevalence with response rate, adjusted for the breadth of the elder care operational definition, remained significant, r = -.50,p \u3c .05; and (e) these two methodological characteristics together accounted for half (R2 = .505) of the variability in reported prevalence, response rate accounting for nearly all (95.4%) of this explained criterion variation. Bias due to nonresponse thus represents a potent threat to the validity of the mean prevalence estimate found in this body of research (21.1%). The implementation of statistical controls for nonresponse and definitional inconsistencies resulted in an estimated prevalence of 7.4% to 11.8%; however, this review outcome is tentative at best and must be tested with future, better controlled primary studies