Gender trouble: Item 110 on the child behavior checklist and youth self-report

When the Child Behavior Checklist and the Youth Self-Report were first published, they both contained two items pertaining to gender identity (Item 5: “Behaves like opposite sex”; Item 110: “Wishes to be of opposite... [ view full abstract ]

When the Child Behavior Checklist and the Youth Self-Report were first published, they both contained two items pertaining to gender identity (Item 5: “Behaves like opposite sex”; Item 110: “Wishes to be of opposite sex”). In subsequent versions of the CBCL and YSR, only Item 110 was retained. Both Items 5 and 110 or Item 110 alone have been used to gauge the prevalence, in a crude way, of gender-variant behavior or even as a rough proxy for Gender Identity Disorder (the diagnostic term in DSM-III, III-R, and IV) in both the referred and non-referred samples in the standardization studies (e.g., Steensma et al., 2013; Zucker et al., 1997) and in more specific populations, such as twins (van Beijsterveldt et al., 2006) or specific diagnostic groups, such as children with an autism spectrum disorder or attention deficit hyperactivity disorder (Strang et al. , 2014). In the standardization samples, Item 5 was endorsed more frequently than Item 110 and both items were endorsed more frequently in girls than in boys. As of yet, it is unclear if endorsement of Item 110 might be associated with degree of general behavioral and emotional problems in referred and non-referred children in general.

The aim of the present study was two-fold: (1) to report on the prevalence of Item 110 in two new standardized samples: the 2001 U.S. version of the CBCL and YSR (Achenbach & Rescorla, 2001) and the 2013 Dutch version of the... [ view full abstract ]

The aim of the present study was two-fold: (1) to report on the prevalence of Item 110 in two new standardized samples: the 2001 U.S. version of the CBCL and YSR (Achenbach & Rescorla, 2001) and the 2013 Dutch version of the CBCL and YSR (Verhulst & van der Ende, 2013); (2) to see whether or not children and youth for whom this item was endorsed differed from those for whom it was not with regard to the probands’ birth-assigned sex, the sum of all other items that were scored positive, suicidality, and a metric of poor peer relations.

Four samples were used: For the CBCL, the U.S. and Dutch samples of referred and non-referred children (ages 6-18 years); for the YSR, the U.S. and Dutch samples of referred and non-referred youth (ages 11-18 years). In total,... [ view full abstract ]

Four samples were used: For the CBCL, the U.S. and Dutch samples of referred and non-referred children (ages 6-18 years); for the YSR, the U.S. and Dutch samples of referred and non-referred youth (ages 11-18 years). In total, there were 12,216 children and youth. Across the four samples, Item 110 was dichotomized as present or absent. The sum of all behavioral and emotional problems (less Item 110) was recorded as was the sum of the two suicidality items (Items 18 and 91) and a 3-item metric of poor peer relations (Items, 25, 38, and 48). Because we were analyzing an anonymous set of data bases supplied by the authors of the standardization samples, formal ethical approval to conduct this research was not required.

The dependent measures of total behavior problems, suicidality, and poor peer relations were each analyzed in a 2 (Gender) x 2 (Referral Status: Referred vs. Non-Referred) x 2 (Item 110: Present vs. Absent) analysis of... [ view full abstract ]

The dependent measures of total behavior problems, suicidality, and poor peer relations were each analyzed in a 2 (Gender) x 2 (Referral Status: Referred vs. Non-Referred) x 2 (Item 110: Present vs. Absent) analysis of variance (ANOVA).

Across referral status, country (U.S. vs. Holland), and form (CBCL vs. YSR), Item 110 was coded as present in less than 1% of cases (e.g., in Dutch referred boys on the CBCL) to 16.0% of U.S. referred girls on the YSR. Across... [ view full abstract ]

Across referral status, country (U.S. vs. Holland), and form (CBCL vs. YSR), Item 110 was coded as present in less than 1% of cases (e.g., in Dutch referred boys on the CBCL) to 16.0% of U.S. referred girls on the YSR. Across the entire sample, the prevalence was 3.7%. In all samples, Item 110 was endorsed more frequently for girls than for boys. In each of the four samples, the ANOVAS showed significant main effects for Item 110 for the sum of the total behavior problems, the sum of the two suicidality items, and the sum of the metric of poor peer relations, with most of the p values < .001. In all instances, children and youth for whom Item 110 was coded as present had more behavior problems, more suicidality, and poorer peer relations. Effect sizes using Cohen's d ranged from .40 to 3.43.

In both clinic-referred and non-referred children and youth, Item 110 proved to be a powerful correlate of behavioral and emotional problems in general and suicidality and poor peer relations in particular. It is rather... [ view full abstract ]

In both clinic-referred and non-referred children and youth, Item 110 proved to be a powerful correlate of behavioral and emotional problems in general and suicidality and poor peer relations in particular. It is rather remarkable that a single item measure that might be construed as a crude proxy for the presence of gender dysphoria was able to detect such strong effects on general psychologic well-being. In clinic-referred populations at large, this finding suggests that whenever Item 110 is coded as present, it is quite likely the case that these children and youth will show evidence of a more general psychologic vulnerability than children and youth for whom Item 110 is coded as absent. It may well serve as a useful gateway for a more thorough evaluation of the felt gender identity of these children and youth.