Search

Is daily data collection of sexual and other health-related behaviors reactive? This question was examined by exploring trends in daily reports of sexual intercourse, condom use, drinking, smoking, and vegetable and fruit consumption over a 2-month period. The trends were examined for evidence of fatigue effects and socially desirable responding. The effects of data collection method and gender were also examined. Results from a hierarchical linear modeling analysis showed a significant trend toward decreased reporting of drinking, smoking, and sexual intercourse, but no significant trends in fruit and vegetable consumption or condom use. These patterns are not consistent with fatigue effects or socially desirable responding. The results are discussed in relation to choosing data collection methods that minimize error. [ABSTRACT FROM AUTHOR]

This study examined the agreement between daily and retrospective reports of vaginal sex over a two-month period in a sample of 376 heterosexually active men and women. We also examined whether gender, age, or method of daily data collection (self-administered vs. interviewer administered) were related to agreement between daily and retrospective reports. Both counts and categorical measures of frequency of the behaviors were examined. There were no gender, age, or data collection method effects. When measured as a count, participants reported more instances of vaginal intercourse in the retrospective reports than on the daily reports. In contrast, comparison of retrospective categorical measures of frequency to daily reports showed considerable variability. Possible reasons for the over-reporting of counts of vaginal sex in retrospective reports are explored.