Off-campus UNL users: To download campus access dissertations, please use the following link to log into our proxy server with your NU ID and password. When you are done browsing please remember to return to this page and log out.

Non-UNL users: Please talk to your librarian about requesting this dissertation through interlibrary loan.

Describing state level mathematical growth using the Student Growth Percentile (SGP) methodology

Darin C Kelberlau, University of Nebraska - Lincoln


The purpose of this study was to describe growth in student mathematics performance at the student and group levels as measured by the statewide mathematics test in Nebraska. Student groupings analyzed for differences in growth patterns were cohorts (elementary and middle school aged students), growth categories, grade levels, gender, ethnicity, students receiving special education (SPED) services, students receiving support services due to English not being their native language (ELL). Building differences were defined by average number of students per grade level (size), levels of student performance, levels of students receiving SPED services, and levels of students receiving ELL services. The Student Growth Percentile (SGP) methodology was the growth model employed to quantify learning growth in mathematics as measured by the Nebraska State Accountability (NeSA) mathematics assessment. This methodology meets federal requirements for accountability, provided student level growth scores, and compared students with similar testing history. The NeSA-M results (2010-2014) were analyzed using the SGP approach for grades 3-8. Analyses were run utilizing the open source SGP Package within the R Studio software application. The SGP assumptions were verified prior to testing for growth differences between student groups. Median SGPs matched expectations, and a variety of perspectives ensured that performance and growth scores were uncorrelated. These approaches confirmed that no relationship existed between performance and growth. Scatterplots were generated to evaluate possible ceiling and floor effects. The plots demonstrated the possibility of experiencing all levels of growth regardless of performance. Comparisons between groups of learners revealed significance due to the power of the study. Effect sizes were all small, which suggests little practical significance. The availability of more longitudinal data offers an opportunity for further investigation.

Subject Area

Educational tests & measurements|Education|Educational psychology

Recommended Citation

Kelberlau, Darin C, "Describing state level mathematical growth using the Student Growth Percentile (SGP) methodology" (2015). ETD collection for University of Nebraska - Lincoln. AAI3712513.