split-half method

split-half method

[′split ‚haf ‚meth·əd]
(statistics)
A method used to gage the reliability of a test; two sets of scores are obtained from the same test, one set from odd items and one set from even items, and the scores of the two sets are correlated.
References in periodicals archive ?
Specifically, the reliability estimated by Cronbach's alpha, omega, and split-half method was good.
Second, we examined the reliability (Cronbach's alpha, omega, and split-half method) of ASBS-E scores.
Blanc & Rojas (under review) reported high internal consistency reliability with two separate Spanish samples as indicated by both Cronbach's alpha ([alpha] = .92 and .90) and split-half method reliability (.96 and .95).
Cronbach's alpha coefficient and split-half method were employed in order to check the reliability of the inventory, which were 0.94 and 0.90 for the entire inventory, respectively.
In the present investigation, Cronbach's alpha coefficient and split-half method were employed in order to check the reliability of the scale, which were 0.84 and 0.85 for the entire scale, respectively.
We calculated reliability estimates using both the split-half method and the ratio of observed to true variance (Nunnally, 1970).
For the sixth aim, the reliability coefficient obtained by the split-half method revealed the reliability to be .87 (p < 0.01).
The item-total correlation coefficient and correlation coefficient obtained by the test-retest and split-half methods were found to be reasonable.
While the reliability of the instrument was further ascertained through the split-half method, which yielded a correlation coefficient of .81
For those familiar with the split-half method for assessing reliability, alpha has an additional interpretation worth noting.
Other models to determine reliability are the test-retest procedure and the split-half method. Test-retest is a procedure used to examine characteristics that remain stable over time.