Why did they lie? Socio-economic bias in reporting menarcheal age

Primary objective: To show that socio-economically dependent cultural bias distorts results of the status quo method of estimating age at first menstruation. Methods: Questionnaires asking for menstrual status and the recalled age at menarche were distributed to approximately 1000 Javanese girls who attended junior and senior high schools in Malang. Age of participants ranged from 11.98 to 18.89 years. Probit analysis was applied to the status quo data while average menarcheal age recalled by girls ≥ 16 year old was also calculated. t-test, F-test, ANOVA and χ² tests were applied to test significance of differences between groups. Main outcome and results: Girls of low occupational status fathers (Group 1) had a probit average menarcheal age of 13.99 years (SD = 1.33, n = 524). Girls of medium occupational status fathers (Group 2), had a lower probit average menarcheal age (13.06 SD = 1.38, n = 315). Girls of fathers with the highest occupational score (Group 3) rarely reported premenarcheal status (less than 10% in all age groups studied) and the probit analysis of their reports yielded an unbelievable average of 9.61 (SD = 3.41, n = 157) years. Group 3 girls tend to report their menarcheal status incorrectly, probably due to a 'fashion' of appearing mature or to 'contagion' during filling out of a questionnaire. The recalled average menarcheal age of Group 3 females questioned when they were aged 16 years and above, yields an average that seems to be more reliable (12.74 years, SD = 1.41, n = 7), because at a fully postmenarcheal age there is no need to enhance one's status by falsely claiming maturity.