Kathryn S. Todd and Dr. Martin Fujiki, Audiology and Speech Language Pathology
Children diagnosed with Specific Language Impairment (SLI) “exhibit significant limitations in language functioning that cannot be attributed to deficits in hearing, oral structure and function, or general intelligence” (Leonard, 1987). Recently, research revealed children with SLI also demonstrate difficulty in social interaction. As part of an ongoing research program at Brigham Young University over the last decade, Dr. Bonnie Brinton and Dr. Martin Fujiki evaluated the social skills of children with SLI. Yet to be investigated, this past year I researched the nonconforming behavior of children with SLI directed towards adults. Nonconformity or noncompliance is the failure to act and behave in agreement with expected social norms.
My study of noncompliant behavior in children with SLI involved comparing noncompliant specific language impaired behavior to noncompliant behavior of typically developing children. I analyzed teacher responses from previously collected teacher behavior rating scales for 120 children with SLI and 120 of their typically developing peers, ranging from 5 to 10 years of age. After reviewing current literature to sharpen the construct of noncompliant behavior, 19 questions were chosen from the questionnaire which best defined noncompliance. I entered the responses of my 19 questions of the 161 questionnaire questions into an Excel spreadsheet; however, this method did not facilitate analysis using an Mplus statistical analysis program.
After consulting a statistician, we decided to change the method of data entry. Two graduate students and I entered the entire questionnaire for each subject and not only the questions needed for my study. The questionnaire responses were then transferred from an MS Word document to the Mplus program by means of assigned codes representing name, gender, age, language diagnosis and responses to the 19 questions. The Mplus program analyzed the data as a confirmatory factor analysis. The factor analysis involved one dependent variable, nonconforming behavior, and the independent variable, the group of questions best predicting nonconforming behavior. The tightest fitting group of the 19 questions was indicative of the construct of nonconformance while the other variables did not show any predictive power. This method allowed for easy analysis of my study, other behavioral studies conducted by the graduate students and possible future studies.
According to the data analysis in terms of nonconformance, behaviors which best represent nonconformance include dawdling when required to do something, unwilling to do chores or assignments unless threatened, not following the teacher’s instructions, not finishing class assignments within time limits, and not putting work materials or school property away.
Upon reviewing the results with a statistician, the analysis showed language skills differ significantly between the responses of typical subjects and SLI subjects (p > 0.05). Disordered subjects exhibit significantly more noncompliant behavior than normal subjects. Behavior also differed significantly between males and females. Male subjects exhibit significantly more noncompliant behavior than female subjects. My research provided some insight into the social interactive behaviors of children with SLI, although additional analysis and detailed explanation of the results are needed.
This research project, an invaluable experience, was beneficial as I enter a master’s program this fall. Working with BYU professors and graduate students, I gained an understanding of the involved research process and recognized the importance of statistics. Learning statistical programs and developing interpretive skills are essential, as is the assistance of a statistician.