Assessing Parent Perceptions of School Fit: The Development and Measurement Qualities of a Survey Scale

Abstract

Students whose school environment fits their developmental needs also typically experience academic success and increased motivation. Most investigations of school fit, which focus on teachers’ and students’ perceptions, have found a general decline in fit across the transition from elementary to middle school. However, little research has examined the school-child fit from the parent perspective. In this article, we first detail the development process behind a new survey measure of parent perceptions of school fit. Second, using three online panel-based samples of parents from across the country (n₁ = 323, n₂ = 188, n₃ = 1,033), we evaluate the scale's measurement properties and conduct exploratory analyses examining grade-level and income-based differences on reported school fit. Finally, in line with previous research, we find that parents of middle school children perceived statistically significantly worse fit than parents of elementary school children. Among parents of high school students, we found that, on average, high-income parents perceive statistically significantly better fit than low-income parents.

ACKNOWLEDGMENTS

We thank all of the participating schools, parents, and expert scale reviewers, as well as Karen Mapp, Richard Weissbourd, and Lauren Capotosto for their contributions to this article. We take full responsibility for any shortcomings.

Portions of the findings from this article were presented at the 2013 American Education Research Association Conference in San Francisco, California.

Notes

¹We use “parent” throughout to describe the broad range of primary caregivers, such as legal guardians, grandparents, and aunts or uncles.

²SurveyMonkey is online survey development cloud-based company.

³While we view the indicators as ordinal and treat them as such in our models, we understand that it is also common for others, such as school leaders, to treat them as interval variables so we have reported descriptive statistics including alpha, means, and standard deviations.

⁴Here and elsewhere, d refers to the standardized mean differences, or effect sizes, between the groups (using a pooled standard deviation).

⁵Sample sizes are as follows: for lower-income parents (n = 379) and higher-income parents (n = 623); and for lower-income parents of elementary school children (n = 90), middle school children (n = 61), and high school children (n = 88), and higher-income parents of elementary school children (n = 126), middle school children (n = 70), and high school children (n = 178).

⁶Due to sample size considerations, we did not test for measurement invariance using this six-group categorical variable.