June 2017: Systematic Review of Language and Problem Behavior

Chow, J. C., & Wehby, J. H. (2016). Associations between language and problem behavior: A systematic review and correlational meta-analysis. Educational Psychology Review. Advance online publication. doi:10.1007/s10648-016-9385-z

Summary by Dr. Garrett Roberts


There is a well-documented high co-occurrence rate of language and behavioral deficits for school-age children. For students with an emotional/behavioral disorder, 81% have an unidentified language deficit (Hollo, Wehby, & Oliver, 2014), and students with a specific language impairment are twice as likely to develop a behavior or attention-related difficulty (Yew & O’Kearney, 2013). Both problem behavior and language deficits also lead to long-term negative outcomes such as difficulties in employment and postsecondary education. Therefore, a better understanding of the relationships and constructs underlying both language and behavior is important when developing and implementing interventions to improve student outcomes.

Research supports a negative relationship between oral language ability (i.e., expressive, receptive) and problem behaviors, meaning that as language abilities increase, problem behaviors decrease. Yet the causal mechanisms underlying this relationship have not yet been determined. According to Hinshaw, Han, Erhardt, and Huber (1992), there are four possible scenarios for the nature of the causal relationship between problem behavior and oral language deficits:

  1. Oral language deficits lead to problem behavior.
  2. Problem behaviors lead to oral language deficits.
  3. A bidirectional relationship exists between problem behaviors and oral language deficits, wherein oral language and behavior affect each other.
  4. There is an undefined mediating or underlying variable contributing to both oral language deficits and problem behaviors.

In the reviewed article, Chow and Wehby (2016) use Hinshaw et al.’s (1992) framework as a basis for conceptualizing and meta-analyzing the correlations between oral language and problem behavior.

Theoretical Framework

In this meta-analysis, the authors posit that the underlying construct for learning and social development is both expressive and receptive oral language ability. In this model, language influences academic and behavioral skills, followed by academic and behavioral skills interacting and influencing each other.

In the classroom context, language and effective communication skills are essential for success. For teachers, language is the primary mode of communication, and for students, effective communication skills are necessary to demonstrate understanding of academic concepts.

Students with problem behaviors often have expressive and receptive language deficits. These deficits can lead to a negative cycle, in which students engage in problem behaviors to avoid instruction and academic demands, falling further behind and thus reinforcing the underlying cause of problem behavior.

Study Purpose and Methodology

To better support our students and teachers, it is beneficial to understand the relationship between language ability and problem behavior. Chow and Wehby (2016) conducted the first review to quantify and test the correlation between oral language and problem behavior. They tested the correlations in two separate meta-analyses. In the first analysis, they explored the concurrent correlations, in which oral language and problem behavior were measured once and at the same time. In the second analysis, they explored studies in which oral language abilities were assessed prior to the assessment of problem behavior.

Additionally, the authors investigated whether the strength of the correlation would change if they looked at receptive language compared to expressive language, internalizing behavior compared to externalizing behavior, or different ages or risk statuses.

To complete their meta-analysis, the authors included studies that included the following:

  • Individuals between the ages of 3 and 21 who were normally developing or had a high-incidence disability (e.g., learning disability). Students with low-incidence disabilities (e.g., intellectual disability) were not included in this meta-analysis.
  • Measures of oral language and externalizing problem behavior with a norm-referenced, standardized, psychometrically sound assessment. Internalizing behavior (e.g., anxiety, depression) was analyzed only if the study also included an externalizing behavior measure.

A total of 19 studies met the inclusion criteria for the concurrent meta-analysis, and 8 studies met the inclusion criteria for the predictive meta-analysis. 

Key Findings

  • The overall finding was that oral language ability was significantly and negatively associated with problem behavior. This relationship was found to extend beyond students with identified deficits to students with and without deficits in oral language abilities and behavior skills.
  • In the 13 studies that included measures of expressive and receptive oral language, the mean correlation was negative, indicating that as expressive and receptive oral language abilities decreased, problem behavior increased.
  • In the nine studies that included measures of externalizing and internalizing behavior, the mean correlation was negative, indicating that as externalizing and internalizing behavior increased, oral language decreased.
  • The findings did not vary according to student age or risk status.
  • Due to a limited number of studies, the authors were unable to assess whether early oral language abilities predict future problem behaviors.

Limitations of the Study

  • Behavior is difficult to measure, and the conclusions of this study are limited by how behavior was assessed in the available studies. The authors noted that adults or parents completed all the standardized behavior measures; these reports may not be sensitive to behavioral changes over time. Further, no study included student- or peer-rated behavior forms.
  • Another limitation of standardized behavior measures was a reliance on composite scores of behavior. Composite scores do not allow for an investigation of whether the outcomes varied by specific behavioral topographies (e.g., talking out, aggression). Further, composite scores may be confounded due to the inclusion of items indicating hyperactivity, inattention, and internalizing behavior.
  • The authors noted that they were unable to take into account how socioeconomic disadvantages may have played a role in students’ problem behaviors or oral language abilities.

Implications for Research and Practice

  • Future research is warranted to further investigate how the associations between oral language skills and problem behavior may vary by expressive and receptive oral language skills as well as topographies of behavior such as anxiety, aggression, inattention, or other characteristics. Future research may also explore differences in problem behaviors through both indirect reports, such as surveys, and direct observations, as previous research has suggested these two types of measures may yield different findings.
  • Researchers can also focus on improving student outcomes through developing student-level interventions and supporting teachers’ instruction. To accomplish this aim, future research can investigate how best to provide pull-out services, scaffold classroom instruction, or provide effective student feedback. This research also could explore differential forms of instruction, such as using visuals or increasing opportunities for students to respond. As researchers continue to more precisely identify the key mechanism in the relationships between oral language skills and problem behaviors, they can use this information to develop tailored and targeted interventions to improve students’ social, behavioral, and academic outcomes.
  • Of interest to practitioners, the authors found that the association between oral language abilities and problem behaviors did not vary across age groups or disability (or deficit) status. Even though the authors acknowledged that their study may not have been sufficiently powered to detect these differences, this finding suggests that regardless of risk factor or grade level, students may benefit from developing their knowledge and strategies in language and communication, which could improve their understanding of teacher directions and their conveying of the knowledge gathered and could increase the likelihood of positive teacher-student relationships. Overall, across all students, improving language may improve student academics and behavior, allowing for greater access to future instruction.


This meta-analysis found that as oral language abilities decrease, problem behaviors increase. This association was small but statistically significant. This finding was consistent when investigating the relationship for both expressive and receptive oral language skills and for externalizing and internalizing problem behaviors. Additionally, the findings were consistent across age groups and risk factors, yet the authors note that the meta-analysis may not have included sufficient studies to detect such differences.

One of the primary limitations of the study was in analyzing the problem behavior outcomes. Some behavior outcomes were problem behavior composite scores, which may have been confounded by including items of hyperactivity, inattention, and internalizing behavior, thus influencing the findings. This study’s analysis was also not able to differentiate teacher and parent reports on behavioral outcomes.

Overall, this study and previous studies support the notion that oral language deficits and problem behaviors are related and that continued research is required to support practitioners and students and improve academic outcomes.


Hinshaw, S. P., Han, S. S., Erhardt, D., & Huber, A. (1992). Internalizing and externalizing behavior problems in preschool children: Correspondence among parent and teacher ratings and behavior observations. Journal of Clinical Child Psychology, 21, 143–150. doi:10.1207/s15374424jccp2102_6 

Hollo, A., Wehby, J. H., & Oliver, R. M. (2014). Unidentified language deficits in children with emotional and behavioral disorders: A meta-analysis. Exceptional Children, 80, 169–186. doi:10.1177/001440291408000203

Yew, S. G. K., & O’Kearney, R. (2013). Emotional and behavioural outcomes in childhood and adolescence with specific language impairments: Meta-analyses of controlled prospective studies. Journal of Child Psychology and Psychiatry, 54, 516–524. doi:10.1111/jcpp.12009