Investigating Potential Biases in Self-Evaluations of Reading and Math Performance By Individuals with ASD, ADHD, and Typical Development

Background:  Self-evaluative abilities for academic competence are critical for student success. Biased self-evaluations have the potential to prevent children and adolescents from knowing when they need to work harder, and or get help from their teachers. Researchers have examined overly optimistic self-ratings of competencies, referred to as positive illusory biases (PIB), by comparing self-rated ability to performance. There is some evidence youth with autism spectrum disorder (ASD) and Attention Deficit Hyperactivity Disorder (ADHD) demonstrate PIB when assessing abilities on laboratory tasks. However, it is unclear if PIB occurs when children with ASD rate their abilities more generally on perceptions of their academic abilities in reading and math, and whether these biases are associated with self-esteem, implying they serve a self-protective function.

Objectives:  To answer these questions, we: (1) examined the extent to which youth with ASD, ADHD and typical development (TYP) displayed PIB when making self-assessments of their reading and math achievement; and (2) tested whether there was a positive association between PIB and self-esteem, suggesting the biases are self-protective. 

Methods:  98 youth, aged 9 to 17 (see Table 1), were administered assessments of their perceptions of their reading, and math abilities, and their general self-esteem using subscales from the Marsh SDQ (Marsh, 1992). Actual math ability was assessed with the Numerical Operations and Problem Solving subtests from the Wechsler Individual Achievement Test, 3rd edition (WIAT-III; Pearson, 2009). Reading fluency and comprehension was assessed using the Gray Oral Reading Tests, 5th Edition (GORT-V; Wiederholt & Bryant, 2012). Subtracting z-scored performance scores from z-scored self-rated abilities, we quantified the amount of bias to examine if positive biases were related to self-esteem. Bivariate correlations were performed using SPSS 22.

Results:  All groups showed significant positive associations between self-rated ability on math and performance on WIAT-III Numerical Operations (Pearson’s r ranged from .36 to .65). The ASD group did not, however, exhibit a similar relationship between their math self-ratings and WIAT-III Problem Solving (r(35)=.16, p=.34). Additionally, the ASD group did not show a positive association between self-rated reading ability and performance on fluency (r(37)=.06, p=.70) or comprehension (r(37)=.09, p=.58), and the ADHD group had an association between self-rated reading ability and fluency, but not comprehension (r(23)=.23, p=.26). On the contrary, the TYP group showed significant associations between self-rated abilities and performances in each domain. Finally, there was a significant positive association with self-esteem and positive bias scores in the reading domain in the ASD group (r(37)=.33, p=.04), suggesting that the function of positive biases may be self-protective for this group.

Conclusions:  These results indicate that youth with ASD may be evaluating their math abilities in relation to arithmetic rather than problem solving ability, suggesting that they may need extra help identifying all abilities relevant to math. In addition, youth with ASD, and to some extent youth with ADHD, are not making accurate appraisals of their reading ability, but there is evidence that these biases are self-protective. Future work should develop self-assessments that help support these abilities for individuals with ASD.