Content area
Abstract
The role of group size and examiner familiarity in obtaining optimal estimates of reading achievement was examined using a sample of junior high remedial readers. The Communication Skills subtest of the Florida State Student Assessment Test, Part I (SSAT-I), a criterion-referenced measure, was employed as the dependent variable.
151 of a possible 188 students enrolled in a remedial program of reading and writing at a large metropolitan junior high school in Dade County, Florida, were assigned to twelve administration groups. The research design was a 2 x 3 factorial design, fixed effects model, with three levels of group size: Large (n = 30); Medium (n = 15); and Small (n = 5) and two levels of examiner familiarity: Familiar (n = 2) and Unfamiliar (n = 4).
Significant main effects were found for both group size and examiner familiarity. The medium size group (n = 15) was most conducive for optimal performance and remedial readers tested by a familiar examiner did much better than those examined by an unfamiliar examiner. There appears to be no combination of group size and examiner familiarity that leads remedial readers to optimal test performances. The results of this study bring in to suspect the commonly employed testing procedures used to assess the achievement of remedial readers and evaluate the effectiveness of remedial reading programs.
