Melanie Cobabe and Dr. Martin Fujiki, Communication Disorders
Teachers and speech pathologists have noted that children with Language Impairment (LI) frequently struggle in social interactions when compared with their typical peers. They have difficulty making friends, entering on-going interactions, and are often viewed negatively by others in their own age group. The purpose of this study was to co-vary intelligence in order to eliminate it as a variable that might impact the differences in social behavior between typical and language impaired children.
To conduct this study, I used the Teacher Behavioral Rating Scale or, TBRS form. It is an observational measure completed by teachers who rate children on behavior from one to three, one being that they never are observed to exhibit the given behavior, two that they sometimes exhibit it, and three that they often exhibit the behavior. This study focused on three behaviors included on the TBRS; one type of withdrawn behavior: reticence, and two sociable behaviors: likeability and prosocial. Teachers were unaware which items referred to which category of behavior in order to reduce bias.
The first behavior on the TBRS, reticence, measures how often a child is anxious to interact with other children. They may watch other children play, but will not join in themselves, perhaps driven by a fear of rejection. The second behavior, likeability, is measured by how the child receives criticism and controls emotional impulses. A child who displays this behavior will be able to engage in rough play and assert leadership skills. The third behavior, prosocial, measures the child’s ability to comfort, cooperate with, offer help and share with other peers.
Taking the TBRS scores for the typical and LI children, I averaged the scores within each behavior. Included below is a sample of data from this step:
Sample Data for Group 1 (Typical)
Subject
Typical/SLI
IQ
Withdrawal
Sociability
Reticence
Prosocial
Likeability
Individual 1
1
113
0.5
1.6
1.8
Individual 2
1
101
0.5
2
1.8
Individual 3
1
88
0.5
1
2
By combining this data, I calculated the overall mean for behavioral scores and the standard deviation within the groups:
Combined Data
LI
Typical
Mean
Standard Deviation
Mean
Standard Deviation
IQ
89.04
12.55
IQ
106.86
12.67
Reticence
0.97
0.63
Reticence
0.32
0.25
Likeability
1.07
0.49
Likeability
1.88
0.17
Prosocial
1.28
0.58
Prosocial
1.67
0.51
To co-vary the IQ, we performed an analysis of co-variance (ANOCOVA). The fixed factors were the two groups, either Typical or LI, the dependent variable being one of the behaviors, reticence, likeability or prosocial. Each of these behaviors were analyzed separately to determine their individual relationship to the two groups.
After the analysis of co-variance, our findings indicated that there was still a significant gap between the typical and language impaired groups and their social behaviors, even without differences in intelligence. For reticence, the level of significance was .004, meaning that there is a .004 chance that the difference between the two groups occurred randomly. For likeability, the level of significance was even lower at .000. The level of significance was .013 for Prosocial. Because we had set our alpha level at .05, the level of significance needed to be below .016 in order to be statistically significant, a requirement which all of the behavioral categories met. Intelligence was not a statistically relevant cause of the difference between any of the behaviors within the two groups.
While the results of the study were close to both Dr. Fujiki’s and my expectations, I expected intelligence to play a larger role in the behaviors than it did. The most difficult part of this study was interpreting the analysis of co-variance and arranging the variables correctly. Although I had recently excelled in a statistics class, I still found myself bewildered by the computer program’s calculations. It was frustrating after working so hard in statistics to find that I still needed help inputting the data and interpreting it. Although I would have liked to have been more self-sufficient, Dr. Fujiki’s help in this area was invaluable and I understand the process better than I did before.
I am excited to be entering the field of speech-pathology at a time when there is an increased focus on the social aspect of language impairment. Language is most valuable as it is applied to relating to others, and focusing solely on language as a speech therapist would be an enormous setback to any therapy program. As I enter graduate school this fall, I am looking forward to finding ways to apply these ideas in my clinical experience and in the future as a certified clinician.