Test of Gross Motor Development-2 (TGMD-2)
All information found in the previous posts regarding the Test of Gross Motor Development – Second Edition (TGMD-2) and the development of the Test of Gross Motor Development – Third Edition (TGMD-3). The official publication of the TGMD-3 is expected to be released in the near future.
Maeng, H., Webster, E. K., Pitchford, E. A., & Ulrich, D. A. (2017). Inter- and Intrarater Reliabilities of the Test of Gross Motor Development—Third Edition Among Experienced TGMD-2 Raters. Adapted Physical Activity Quarterly,34(4), 442-455. doi:10.1123/apaq.2016-0026
One study was located that researched the inter- and intra-rater reliabilities of the TGMD-3 and was published in October of 2017.
The purpose of this study was to examine the inter- and intra-rater reliabilities of the Test of Gross Motor Development – third edition (TGMD-3). The study population included ten typically developing children from the Midwestern region that were digitally recorded performing the TGMD-3. The age of participants ranged from 3 years, 7 months to 10 years, 9 months with a mean age of 6.57 years old. There was somewhat of a balance to the sample related to sex and race, with 60% participants being male, 40% participants being female, 70% participants classified as white, and 30% participants classified as African American. Informed consent by the parent or legal guardian was obtained for each child, as well as child assent on the day of testing.
Methods included the recruitment of five raters by the principal investigator to score the digitally recorded performances of the children using the TGMD-3. Eligibility requirements to determine if a rater could participate included: a) a graduate degree in motor development or adapted physical activity; b) experience administering a scoring the TGMD-2; and c) experience teaching physical education, adapted physical education, or physical activity for children of at least 2 years. Standardized procedures outlined in the TGMD-3 were followed. During test administration, one assessor ran the assessment including providing instruction, visual demonstration, and classroom management to the participant. The second assessor digitally recorded the child’s performance.
Each rater independently scored the digital recordings of the 10 recordings to provide data to analyze inter-rater reliability. Each rater then rescored the same digital recordings of the performances after an interval of 2 weeks to provide data to analyze intra-rater reliability.
Descriptive statistics, including mean and standard deviation of scores of the 10 participants, were analyzed for each rater and all of the TGMD-3 skills, locomotor and ball skills subtests, and gross motor total score. To determine the intra-rater and inter-rater reliability of the examiners, intraclass correlation coefficients (ICC) were calculated. This type of ICC was selected to account for systematic and random variance between and within raters. Coefficients less than 0.40 were interpreted as poor reliability, 0.40–0.59 as fair reliability, 0.60–0.74 as good reliability, and 0.75 and above as excellent reliability.
After statistical analysis, inter-rater reliability for the total score, locomotor subscale, and ball skills subscale were all excellent with an ICC between 0.77 and 0.98. However, individual skills had fair to excellent inter-rater reliability with an ICC between 0.52 and 0.93. Overall intra-rater reliability across all raters was excellent with an ICC between (0.77-0.98). However, intra-rater reliability varied greatly for individual raters with an ICC between 0.28 and 1.00, with multiple raters having poor intra-rater reliability.
The strengths of this study include a large age range that represents the spectrum of appropriate ages for the test, qualified raters, and easily applicable information. Some limitations of the study include space limitations, angle of the camera related to the performer, and needing to pan the camera to keep the performer in the frame.
Overall, the initial results suggest that the TGMD-3 has excellent inter- and intra-rater reliability when analyzing total score and subscales. However, reliability between individual skills is less reliable. With this information, it will be important for administrators of the test to be adequately trained to accurately and reliably score the TGMD-3 as it becomes publicly available. More research is needed to test the inter- and intra-reliability of test administration involving children with special needs.
One response to “Test of Gross Motor Development-2 (TGMD-2)”
Leave a Reply
You must be logged in to post a comment.
I liked that this study looked at the different aspects of reliability, and that they identified that this is a test that a person needs to be trained in to administer accurately and reliably. It is important to know when deciding on a test, if it is one you can administer in your clinic after reviewing the forms or if it is one that requires formal training to administer well.