How Valid are Self-report Measures for Evaluating Relationships between Women's Health and Labor Force Participation?

Abstract

For a sample of white women aged 45-64, women who were out of the labor force had poorer self-reported health and higher mortality than women who were in the labor force. It has been hypothesized that women who are out of the labor force may tend to exaggerate their poor health in self-report data. However, no evidence of bias of this type was found in an analysis of the relationships between self-reported health and subsequent mortality. The validity of self-reports of illness as a reason for not seeking work has been assessed using data for a sample of 30-44 year old women who were out of the labor force. Over 90% of the women who gave illness or disability as their main reason for not seeking work had previous or contemporaneous independent, self-report evidence of poor health. The findings of this study and previous evidence indicate that poor health reduces the likelihood that a woman will join the labor force, and this is a major reason why women who are not in the labor force have poorer health than those who are in the labor force.

Psychosocial predictors of health status

“…This permitted an examination of the incremental variance attributed to the major psychosocial variables to physical health status, while minimizing the possible confounds of negative affectivity (NA) on self-reported health status (cf. Watson and Clark 1984, Smithlet at. 1989, Costa et al 1989, Watson and Pennebaker 1989.…”

“…Watson and Clark 1984, Smithlet at. 1989, Costa et al 1989, Watson and Pennebaker 1989. Psychological well-being entered significantly on the first step of the first regression analysis: illness frequency.…”

“…Respondents are asked to rate how frequently they practice these specific health habits on a 1-5 scale (e.g., 'never', 'rarely', 'sometimes', 'often', 'always'). This scale is conceptually based on earlier research on preventative health behaviours including Harris and Guten (1 979), Belloc and Breslow (1972), Wiley and Camacho (1980), and Pardine et al (1982).…”

“…Physical illness was measured by a modified version of the seriousness of illness scale (Wyler et al 1968(Wyler et al , 1970. This scale contains a list of 94 physical health conditions, illnesses, and diseases ranked from least to most severe (e.g., warts, colds, flu, gallstones, ulcer, high blood pressure, heart attack, cancer, etc.).…”

The employment of mothers and the outcomes of their pregnancies: an australian study

“…Various sources suggest that work histories may be obtained with a reasonable level of accuracy and where inaccuracies are found these are not systematically related to age, education or socioeconomic variables. [24] Similarly, morbidity [25] and health service utilisation reports have reasonable validity. Nevertheless, it may be the case that employed women and housewives differ in their perception of and response to symptoms (perhaps as a consequence of the different time constraints operating for these women).…”