Although sampling techniques have been used effectively in education research and practice it is not clear how stratified random sampling techniques apply to high-stakes testing in the current educational environment. The present study focused on representative sampling as a possible means for reducing the quantity of state-administered tests in Texas public education. The purpose of this study was two-fold: (1) to determine if stratified random sampling is a viable option for reducing the number of students participating in Texas state assessments, and (2) to determine which sampling rate provides consistent estimates of the actual test results among the population of students. The study examined students’ scaled scores, percent of students passing, and student growth over a three-year period on state-mandated assessments in reading, mathematics, science, and social studies. Four sampling rates were considered (10%, 15%, 20%, & 25%) when analyzing student performance across demographic variables, including population estimates by socioeconomic status, limited English proficiency, and placement in special education classes. The data set for this study included five school districts and 68,641 students. Factorial ANOVAs were used initially to examine the effects of sampling rate on bias in reading and mathematics scores and bias in percentage of students passing these tests. Also 95% confidence intervals (CIs) and effect sizes for each model were examined to aid in the interpretation of the results. The results showed main effects for sampling rate and campus as well as a two-way interaction between these variables. The results indicated that a 20% sampling rate would closely approximate the parameter values regarding the mean TAKS reading and mathematics scale scores and the percentage of students passing these assessments. However, as population size decreases, sampling rate may have to be increased. For example, in populations with 30 or fewer students in a subgroup it is recommended that all students be included in the testing program. This study situated in one state contributes to the growing body of research being conducted on an international basis in sample-based educational assessments.
| Identifer | oai:union.ndltd.org:unt.edu/info:ark/67531/metadc407797 |
| Date | 12 1900 |
| Creators | Brown, Stephanie N. |
| Contributors | Byrd, Jimmy, Middlemiss, Wendy, Brooks, John, Stromberg, Linda |
| Publisher | University of North Texas |
| Source Sets | University of North Texas |
| Language | English |
| Detected Language | English |
| Type | Thesis or Dissertation |
| Format | Text |
| Rights | Public, Brown, Stephanie N., Copyright, Copyright is held by the author, unless otherwise noted. All rights Reserved. |
Page generated in 0.0178 seconds