Investigating Factors Underlying Variation in Instructional Time Loss Within and Between California High Schools

• In Event: 61.036 - Exploring Learning Time as a Core Resource for Educational Equity and Achievement

Abstract

The Keeping Time survey of high school teachers in California, part of the larger Ford Foundation Keeping Time study, provides a source of valuable information about teachers’ use of time allocated for instruction, and the various factors inside and outside their schools that shape their use of allocated learning time. A distinctive feature of the sample for this study is that the 783 teachers who responded to the Keeping Time survey are nested within the study’s sample of 193 schools. That is, the data have a multilevel structure. This raises several key questions that we address in this paper via multilevel analyses (see, e.g., Raudenbush & Bryk, 2002).

First, how much of the variation in teacher instructional time loss –a key construct of interest in the Keeping Time study -- lies within schools and how much lies between schools? For example, is there a tendency for teachers in the same school to be highly similar in terms of instructional time loss? Put differently, does virtually all of the variation lie between schools? Or are there appreciable differences in time loss among teachers within schools? Secondly, what are the key teacher factors and classroom compositional characteristics that underlie the variation in instructional time loss and in time spent reviewing / re-teaching that we see within schools? What are the key school-level characteristics and policies that account for the variation in these two dimensions of time use that lies between schools?

Preliminary analyses using multilevel models indicate that more than half of the variation in amount of instructional time loss lies within schools. A key within-school factor that we have focused on thus far captures a teacher’s rating of her typical class’s academic achievement level relative to other students in that school. Our initial multilevel analyses indicate that within schools, higher class academic ratings are associated with appreciably lower amounts of instructional time loss.

With respect to between-school variation in time loss, our initial analyses show that the percentage of students in a school who qualify for free or reduced lunch (%FRL) has a strong positive association with amount of time loss, and accounts for nearly half of the variation between schools in instructional time-loss. This finding is readily understandable in view of the fact that the teachers in schools serving high poverty communities report considerably larger numbers of students in their classes who are impacted by an array of economic and social stressors that affect student attendance and focus (authors, 2014).

Further analyses will focus on a variety of other potentially important within-school factors (e.g., additional compositional characteristics of a teacher’s typical class; the subject area of a teacher’s typical class; various teacher practices such as the amount of time spent planning / prepping), as well as additional potentially important between-school factors (e.g., block versus traditional class schedules; %ELL; school size; various chronic problems connected with school conditions).

Authors

• Michael H. Seltzer, University of California - Los Angeles

• Jung-Eun Yung, University of California - Los Angeles