Recently, computerized tests have been successfully applied in many areas such as education, medicine, or technology. While computerized tests have several advantages: (a) scheduling convenience; (b) instant scoring capability; and (c) reduction of testing time in an adaptive testing situation, little effort has been made applying computerized tests in physical education (PF.) settings. The purpose of this study K to develop a Computerized Dynamic Assessment for Basketball Knowledge (CDABK). For this study, a total of 158 junior high school students (88 boys and 70 girls; age 15 years ± 0.82) were recruited from Korea. The CDABK was built using Microsoft Excel as a platform, and programming for the various functions and interface was carried out using Microsoft Visual Basic. A digital camera and video image editing were used to develop multimedia contents (e.g., full motion video). Psychometric evidences including validity and reliability for the CDABK were investigated, and item analyses (i.e., item difficulty and discrimination) were performed provide Information on the effectiveness of the CDABK. First, the validity evidence based on fest content was established by creating a table of specification. The authors identified three content categories (i.e., basic skills, rules, game strategy) and linked items under each category. In order to establish the validity evidence based on relations to other variables, the CDABK was compared to an existing paper-and-pencil (P&P) basket-ball knowledge test. Participants' scores from CDABK were positively related to P&P basketball test scores (r= .62), which supported the evidence of convergent validity. The internal consistency (i.e., Cronbach's alpha coefficient) was estimated to see how consistently participants responded to the items within a test. The 10-item CDABK had a Cronbach's alpha of .57. Item difficulty (P) of CDABK ranged from .52 to .92, and item discrimination (D) ranged from .53 to .66, showing an appropriate level of item statistics, Compared to item statistics for P&P Lest (P - ranged from .47 to .76; D - ranged from .20 to .51), the CDABK has a wider range of item difficulty and higher item discrimination, indicating that the test contains better quality items. Traditional P&P test items in PK setting often require a long passage to describe psychomotor skills or knowledge. Developing computerized tests using multimedia contents, such as CDABK, may address this shortcoming and lead to more practical assessments.
© Copyright 2008 Research Quarterly for Exercise and Sport. American Alliance for Health, Physical Education, Recreation and Dance (AAHPERD). Alle Rechte vorbehalten.