This study investigated the usefulness of the bifactor model in the investigation of score equivalence from computerized and paper-and-pencil formats of the same reading tests. Concerns about the equivalence of the paper-and-pencil and computerized formats were warranted because of the use of reading passages, computer unfamiliarity of primary…

Studied the factorial invariance of scores from male and female students when reading comprehension was assessed with narrative and expository passages related to war and military content. Narrative and expository item factor loadings and intercepts were invariant for both genders, a finding that supports the use of passages about war for reading…

This study investigated factors related to score differences on computerized and paper-and-pencil versions of a series of primary K-3 reading tests. Factors studied included item and student characteristics. The results suggest that the score differences were more related to student than item characteristics. These student characteristics include…

Examines variables including handwriting, English mechanics, number of words written, average T-unit length, type of correct answers, and unrelated responses on which male and female adolescents in grades 7 and 10 were hypothesized to differ when producing constructed responses. Results suggest that further studies need to define and separate the…

In this study, we investigated possible context effects when students chose to defer items and answer those items later during a computerized test. In 4 primary school reading tests, 126 items were studied. Logistic regression analyses identified 4 items across 4 grade levels as statistically significant. However, follow-up analyses indicated that…

This study investigated the equivalence of scores from computerized and paper-and-pencil formats of a series of K-3 reading screening tests. Concerns about score equivalence on the computerized formats were warranted because of the use of reading passages, computer unfamiliarity of primary school students, and teacher versus computer…

Studied the factorial invariance across students from different ethnic groups of a seventh-grade state reading assessment with a narrative passage selected for its Hispanic representativeness. Results for the 35,324 Kansas students support the factorial invariance of the scores, but invariance of the error variances across groups was not…

Studied the factorial invariance of a seventh grade reading assessment across groups of 600 general education and 1,128 limited English proficient students. Fit statistics generally supported invariance across all groups for factor loadings and intercepts, but invariance was not supported for the error variances. (SLD)

Gender differences in answers to constructed-response reading items from a state assessment program were studied with four raters rating approximately 500 papers at two grade levels. Results indicate that number of words written and number of unrelated responses show significant gender differences and are related to holistic scores. (SLD)

This study investigated the statistical and practical significance of context effects for items randomized within testlets for administration during a series of computerized non-adaptive tests. One hundred and twenty-five items from four primary school reading tests were studied. Logistic regression analyses identified from one to four items for…

Investigated the factorial invariance of scores from a seventh grade state reading assessment across general education students and selected groups of students with disabilities. Assessed the fit of a two-factor model and five levels of constraint. Results generally support the score comparability of the reading assessment, but more research is…

Rubric-related and rubric-unrelated features used by teacher raters to assign scores to constructed response items for mathematics and reading assessment were studied by looking at 1000 papers from Kansas seventh and tenth graders. Results validated the influence of correctness of answer in both mathematics and reading holistic scores. (SLD)