Measuring The Gaps Between The Research And Statistical Skills And The Training Needs Of Graduate Students
Loading...
Date
2010-11
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Universiti Sains Malaysia
Abstract
This study was initiated by the lack of research-based study to provide reliable and valid information regarding graduate students’ training needs in the area of research and statistics. The primary purpose of this study was to examine the graduate students’ state of preparedness in conducting and evaluating research literature, their knowledge base and their training needs in the area of research and statistics. The secondary purpose was to identify the graduate students’ demographic and situational variables that may explain or predict the overall training needs. This study modified the Borich (1980) model to prioritize graduate students’ training needs and utilized quadrant analysis to provide informative list of training for master students and doctoral students respectively. The target population was graduate students who currently pursuing their master degree or doctoral degree in the four public universities (i.e., UKM, UM, UPM, & USM). The questionnaire template was designed and data collection was accomplished through the subscription of web-based survey. A total of 240 fully completed surveys were returned and analyzed.
Graduate students reported low to moderate preparedness level in evaluating and conducting research. This result was not surprising since their knowledge base were confined to introductory level research and statistics. Graduate students rated high training needs in the topics listed in data interpretation stage which included interpret statistical output, interpret p-values, interpret statistical and practical significance, and make generalization. In addition, the measurement topics such as construct instruments, establishing validity evidence and reliability estimates, and factor analysis were also reported as having
high training needs. Therefore, it is crucial that these topics or skills should form the basis for developing, implementing, evaluating, and improving research and statistics training program. Graduate students who have exposure to statistical software packages rated significantly higher overall preparedness level in evaluating and conducting research, higher overall knowledge and overall performance in research and statistics topics/skills. A worth noting area is that none of the selected demographic and situational variables except exposure to statistical package(s) found to be significant predictor of graduate students’ overall training needs. The low variance explained by the exposure to statistical package(s) indicated the insufficiency of students’ exposure to statistical package(s) and needed to be considered when training curriculum is structured and developed. Finally, the use of Borich model in determining training needs based on the mean weighted discrepancy score (MWDS) was further validated.
Description
Keywords
The lack of research-based study to provide reliable , and valid information regarding graduate students