Psychometric properties of the Turkish version of the behavior assessment for children (BAC) scale

Abstract

Behavioral problems are seen in one out of four school-age children. There are difficulties in evaluating children's behavior in Turkish culture and therefore a valid and reliable measuring tool is needed for assessing behavioral problems in this age group. The aim of this study was to evaluate the psychometric properties of the Turkish version of the Behavior Assessment for Children (BAC). The study was conducted using the cross-sectional data collection method to examine the content validity, factor structure, measurement invariance, reliability of BAC based on data from 495 parents and 14 teachers. The Content Validity Index (CVI) was used for testing content validity. Reliability analysis was carried out with Cronbach's alpha, MacDonald's omega test, item total correlations, Spearman Brown's and Guttman's split-half reliability coefficients, percent agreement and intraclass correlation coefficients (ICC). Construct validity was evaluated by confirmatory factor analysis (CFA). The measurement invariance was examined by multiple-group CFA. The CVI of the BAC was found to be .92. Three models were tested with CFA, and a 17-item, three-factor (attention, emotion, self-control) structure of the BAC was finally supported; this dimensional structure proved to be invariant across gender. Factor loadings varied between .56 and .83. Cronbach's alpha was .93; ICC was .60. Percent agreement varied between 41.6%-63.1%. It was determined that the BAC scale is a valid and reliable instrument that can be used to assess the behavioral problems of Turkish children. Its results can moreover contribute to the cross-cultural study of behavior problems and add to worldwide discussions.

Development of Hypoglycemia Management Scale for Teachers

“…Split-half method refers to splitting a body of supposedly homogeneous data into two halves and calculating the results separately for each to assess their reliability by using Spearman-Brown formula and Guttmann split-half formula (27). Reaching a value of .70 and above represents a reliable measurement for the scale (26). In our study, Spearman Brown split-half correlation was found 0,702 and Guttman Split-half coefficient was calculated as .70 (Table 3).…”

“…Higher item correlation means the items exemplify similar behaviours and high level of internal consistency as well. Literature suggests on this issue that item total score correlation's being .30 and above represents that items distinguish the participants well (24,25); .20-.30 can remain in the test if necessary; .20 and less means that those items should be removed (26). At present study, even though item total score correlation of the 6th item was .27, it was decided to remain in the test since its factor load was .57 (Table 2).…”

“…According to test-retest result performed to test the reliability of the scale, a positive high level relation was identified for the overall scale (r=.863; p<.01) and it was determined as .865 for hypoglycemia knowledge sub-dimension and .865 for hypoglycemia management sub-dimension (Table 4). Test-retest method, used to measure scale's do not change in time, is expected to have a value over .70 (26). When those values are taken into account, this scale has an appropriate reliability to be implemented.…”