Search

In prior research, we used 4 stability platform tests as a measurement of core stability and found that scores on the third and fourth days of testing were essentially the same for each of the 4 tests. Lafayette Instrument Co. subsequently made us a prototype stability platform to enhance this research. The purpose of the present research was to determine the effect of changes in (1) equipment and (2) number of test administrations on test reliability. We also increased the number of test administrations on each day from 5 to 10 but only used the quadruped arm raise test. The subjects were 25 university students; each was tested on 10 trials of 30-seconds duration on 4 different days to enable us to study the learning effect. All 10 trials for the first day, as well as the first trial on the 3 subsequent days of testing, were used as practice trials. Trials 2 to 6 on testing days 2 to 4 were chosen for the data analysis. With 5 trials, the maximum score attainable was 150 seconds.; the means were 125 for day 2 and 132 for both days 3 and 4. Internal consistency intraclass reliability coefficients based on a 1-way analysis of variance model and a criterion score, which was the sum or mean of trials 2 to 6 for days 2 to 4, were 0.89, 0.95, and 0.92, respectively. Stability reliabilities were 0.76 and 0.92 for day 2 versus day 3 and day 3 versus day 4, respectively. Although we had hoped to show that only 2 days of testing would be required in future research, because of learning effect, 3 will be needed.

In this study, a 4-item battery of core stability (CS) tests modeled on core stabilization activities used in training and rehabilitation research was developed, and a measurement schedule was established to maximize internal consistency and stability reliabilities. Specifically, we found that 4 test administrations on each of 4 days produced intraclass correlation coefficients that in most instances exceeded 0.90 and stability reliability coefficients on the third and fourth days of testing that exceeded 0.90 for 2 of the tests and 0.80 for the other 2. Thus, it is recommended that in future research, examiners administer the battery for at least 3 days and consider the data collected on day 3 as the best estimate of participant CS.