Academic Journals Database
Disseminating quality controlled scientific knowledge

Repeatability of the VMI Supplemental Developmental Test

Author(s): Marjean Taylor Kulp, OD, MS, FAAO | Michael J. Earley, OD, PhD, FAAO

Journal: Optometry & Vision Development
ISSN 1557-4113

Volume: 39;
Issue: 2;
Start page: 76;
Date: 2008;
VIEW PDF   PDF DOWNLOAD PDF   Download PDF Original page

Keywords: children | repeatability | visual perception

ABSTRACTVisual perceptual ability has been found to berelated to academic achievement. Therefore, thescreening of perceptual skills in children should providevaluable information. However, the time requiredto administer and/or score many visual perceptualtests makes them difficult to use as a screening test.The VMI Supplemental Developmental Test ofVisual Perception (VP) test 1) has been shown to berelated to academic performance, 2) has an objectivescoring system and 3) can be administered and scoredquickly and easily. Therefore, the VP test may havethe potential to be used as a stand-alone screeningtest of motor-reduced visual perception. However, itsrepeatability as a stand-alone screening test has notbeen evaluated.Method: The Ohio State University institutionalreview board approved the study protocol, advertisements,and informed consent forms. Informedconsent was obtained prior to administering anytesting. A Modified Clinical Technique visionscreening was administered on a different day prior tothe perceptual testing. The VP test was administeredto second, fourth and sixth grade children (n=171,mean age=10.08 years) from a middle class, primarilywhite, elementary school near Columbus, Ohio. Children (n=136, mean age=10.10 years) were thenretested within one month. Repeatability analysisincluded McNemar analyses, repeated measures, andplotting the difference versus the mean of the scoresobtained at the initial test and retest.Results: The mean +/- standard deviation VPstandard score was +115.5 +/- 14.5. For McNemaranalyses, failure was defined as a score below onestandard deviation below the mean. Analyses usingboth sample means and published norms revealedthat children were referred similarly on test and retest.The mean difference between test and retest scoreswas similar to zero. The 95% limits of agreement werefound to be –3.74 to 3.99.Conclusion: No consistent learning effect appearedto be present upon retest.Supported by COVD, T35-EY07151, OLERF,and EF Wildermuth Fdn grants to MTK and MJE.

Tango Rapperswil
Tango Rapperswil

RPA Switzerland

Robotic Process Automation Switzerland