Self-reported honesty among middle and high school students responding to a sexual behavior questionnaire.

PURPOSE: To determine self-reported honesty in completing a sexual and other risk behavior questionnaire among middle and high school students, and to relate honesty scores to sexual behavior item responses as a method to detect bias in reporting.
METHODS: A self-administered questionnaire measuring overall honesty (7-point rating scale), sexual honesty (5-category scale), and selected sex behaviors was used. Urban, predominantly minority middle and high schools (Grades 7-12) were examined, and participants were 3144 male and female students in middle (mean age = 13.7 +/- 2.0) and high (17.3 +/- 1.6) school health classes.
RESULTS: The majority of students stated that they had been very or completely honest in responding to items on the questionnaire. Seventy-eight percent of middle school males (lowest rate), and 94% of high school females (highest rate) reported honesty. Middle school males were most likely to declare dishonesty regarding sexual behavior items, overstating their actual behavior (14%), while middle school girls were most likely to understate (8%) their behavior. Self-reported sexual honesty and reports of behavior were most consistent for understaters. That is, those subjects who answered that their questionnaire responses underreported their true sexual behavior did, in fact, report lower sexual activity on selected survey items.
CONCLUSION: Middle and late adolescents reported high levels of honesty in responding to a sexuality-related questionnaire. When interpreting such questionnaire data, correction for the tendency to overstate among middle school males and understate among middle school females should be considered; conclusions about self-reports of sexual behavior among young adolescents need to take into account degree of honesty. However, the presence of some overreported and some underreported behavior does not invalidate interpretation of the overall survey findings.