For both the CPM/CVS and SPM+/MHV, the individual’s raw scores can be converted to standard scores with a mean of 100 and a standard deviation of 15. However, it should be noted that these scores are in steps of five points. The relatively small number of items within the tests meant that a finer degree of precision could not be realistically obtained.
Percentile ranks are given which correspond to the standard scores.
Confidence Intervals can be derived around all scores.
Age Equivalents can be obtained.
A discrepancy analysis can be completed between the matrices (CPM and SPM+) and their relevant vocabulary scales (CVS and MHV).
back to top
Evidence of Reliability
The split-half reliability of the CPM was found to be .97.
An equating study between the CPM and the CPM-Parallel found the correlation between the two forms to be .87.
The standard error of measurement for the CPM standardised score is 2.62.
The split-half reliability of the CVS was .96.
The test-retest reliability was .88.
The standard error of measurement was 3.00.
back to top
105 children were administered the SPM+ individually as well as in groups. There was a significant interaction between order and method of administration, such that administering the SPM+ in a group has no effect on subsequent scores when the test is administered individually, while a previous individual administration of the SPM+ does increase scores by 2 raw score points when the test is subsequently administered in groups.
The split-half reliability of the SPM+, calculated from the UK standardisation sample, was found to be .94.
The standard error of measurement for the SPM+ standardised scores is 3.79.
171 children were administered the MHV on two occasions. Of these, 90 were administered Form 1 on the first occasion and Form 2 on the second occasion, while 81 were administered Form 2 on the first occasion and Form 1 on the second. This study showed that Form 1 and 2 were indeed parallel and that a previous administration of the MHV had little effect on the subsequent score resulting from a retest.
These data also generated the test-retest reliability for the MHV of .916, and the parallel forms reliability between Form 1 and Form 2, which was .929.
The standard error of measurement for the MHV standardised scores is 3.99.
Evidence of validity
A wealth of studies conducted with the CPM, CVS, SPM+ and MHV have demonstrated its content, construct and criterion-related validity and these are detailed in the manual.
back to top