Self-concept in poor readers: a systematic review and meta-analysis

Abstract

Background: The aims of this systematic review and meta-analyses were to determine if there is a statistically reliable association between poor reading and poor self-concept, and if such an association is moderated by domain of self-concept, type of reading impairment, or contextual factors including age, gender, reading instruction, and school environment.

Methodology: We searched 10 key databases for published and unpublished studies, as well as reference lists of included studies, and studies that cited included studies. We calculated standardised mean differences (SMDs) and 95% confidence intervals for one primary outcome (average self-concept) and 10 secondary outcomes (10 domains of self-concept). We assessed the data for risk of bias, heterogeneity, sensitivity, reporting bias, and quality of evidence.

Results: Thirteen studies with 3,348 participants met our selection criteria. Meta-analyses revealed statistically significant SMDs for average self-concept (-0.57) and five domains of self-concept (reading/writing/spelling: -1.03; academic: -0.67; math: -0.64; behaviour: -0.32; physical appearance: -0.28). The quality of evidence for the primary outcome was moderate, and for secondary outcomes was low, due to lack of data.

Conclusions: These outcomes suggest a probable moderate association between poor reading and average self-concept; a possible strong association between poor reading and reading-writing-spelling self-concept; and possible moderate associations between poor reading and self-concept in the self-concept domains of academia, mathematics, behaviour, and physical appearance.

Keywords: Dyslexia; Emotional health; Emotional problems; Meta-analyses; Poor reading; Reading impairment; Self-concept; Systematic review.

Figure 1. Flow diagram of the literature search resulting in 13 included studies.

Figure 2. Forest plot of data for the primary outcome (average self-concept).

Figure 3. Funnel plot of the relationship between standard mean differences (SMDs) and standard errors (SEs) for studies contributing to the primary outcome (average self-concept).

Vertical dotted line represents mean SMD for all studies.

Figure 4. SMDs (with mean confidence interval; CI) for each secondary outcome in decreasing order of strength (left to right).