الملخص الإنجليزي
Reliability meta-analysis is a systematic approach to find out the typical estimate of the reliability of the tests' scores and the factors that could increase or decrease the reliability coefficient. The present study aimed at meta-analyzing the reliability of the scores of the Raven's Progressive Matrices Tests (RPMT) and their corresponding standard errors of measurement (SEM). The researcher were analyzed fifty six studies which used the RPMT. The studies were selected using a convenience sampling method. A total of 143 reliability coefficients (117 internal consistency coefficients and 26 test-retest coefficients) and 126 SEM were extracted from the selected studies.
Using Vacha-Haase's (1998) approach of reliability meta-analysis, the results showed that the typical estimates of the score reliability of
the RPMT were 0.85 for internal consistency and 0.76 for test-retest. The typical estimates of the SEM were 2.51 for internal consistency and
3.43 for test-retest.
Also, the results revealed no statistically significant differences (p > .05) in the internal consistency coefficients and SEM due to the gender. However, there were statistically significant differences (p<.05) between the studies of the RPMT in the internal consistency coefficients and SEM due to the environment of the test administration, age category, and type of the RPMT.
Further, the results showed that the size of the reliability coefficient is not influenced by the size of the sample. Yet, there was a statistically significant (p<.05) positive relationship between sample size and SEM. Also, a statistically significant (p<.05) negative relationship was found between the size of the reliability coefficient and the SEM after controlling for standard deviation of the test scores. Finally, the results revealed no statistically significant (p > .05) relationship between the size of the reliability coefficient and the time interval when using test-retest method in RPMT.
The results were discussed in light of the study theoretical framework and the results of the previous studies. Based on these results, some recommendations and suggestions were presented.
