Practices that cross disciplines? : a closer look at instruction in elementary math and English language arts
Abstract/Contents
- Abstract
- Current policies require that states develop comprehensive teacher evaluation systems that include both value-added measures (VAMs) based on student achievement data and classroom observations. Though many of the nation's largest districts are using these composite metrics, there is neither a large body of empirical research about the relationship among these measures nor carefully delineated theory about the facets of teacher quality each measure is designed to capture. In particular, we know little about how content moderates the relations among teaching practices and student achievement gains. The question of content-specific versus generic practices is particularly important for elementary educators who are generalists and typically assessed as more or less "effective" without regard to content. We know little of whether and how elementary teachers enact similar practices in multiple content domains, and or how the content influences the association of those practices with student achievement. I focus on three practices-- modeling, strategy instruction, and orchestrating discussions-- that research suggests are used in multiple content areas and may be associated with student achievement gains in both math and language arts. My research questions are: 1) How are the instructional practices of modeling, strategy instruction, and orchestrating classroom discussions used in 4th grade English Language Arts (ELA) and math classrooms?, and 2) How do the relations among each of these classroom practices and student achievement vary by content area and assessment type? The sample for this study, drawn from the Bill and Melinda Gates Foundation's Measures of Effective Teaching (MET) project, includes videos of math and ELA lessons from all the participating 4th grade teachers, 120 teachers total, in a single district. Using data from the MET study also allowed me to analyze student achievement gains on multiple assessments: the state's assessments in math and language arts as well as well as two supplemental assessments given across the MET districts, the Balanced Assessment of Mathematics that assesses conceptual understanding of mathematics through open-ended problem solving, and the open-ended version of the Stanford Achievement Test (9th Edition for reading) that asks for written responses to reading passages. I first modified a structured classroom observation protocol designed for scoring language arts teaching, PLATO, so that it could be used to reliably score math instruction. I coded the teaching practices of teachers in both math and ELA, examined descriptive differences in the ways in which teachers used these practices in the two subjects (i.e. when, how, and to what effect), and analyzed the relationships with teacher value-added, computed using multiple student achievement measures. Across the 637 lessons scored, there was little rigorous, extended classroom discourse in either subject. During whole class discussions, teachers tended to evaluate student responses rather than press them to elaborate, clarify, or justify their contributions. There was, however, statistically significantly more modeling and procedural strategy instruction in mathematics instruction than in language arts. In mathematics lessons, teacher modeling was often coupled with procedural strategy instruction. The teacher would provide a formula or algorithm, and students would model solving problems using the formula on the board. Rarely did teachers engage in conceptual exploration of solution methods or ask students to explain why or how a procedure worked. In ELA, students were often asked to complete more open-ended tasks (i.e. writing a persuasive letter) with no strategic instruction or modeling. Interestingly, teachers did not demonstrate the same teaching practices when teaching different content. The within-teacher cross-subject correlations ranged from 0.04 for modeling to 0.29 for orchestrating discourse. Even teachers who displayed strong use of these practices in one subject, did not necessarily employ them when teaching another subject. There were also differential relationships with teacher value-added in the two content areas, and those relationships also seemed to be contingent upon the student outcome measure used to calculate a "teacher effect". When value-added was computed using the state test, comprised entirely of multiple choice questions and focused on procedural tasks, scores on modeling and strategy instruction only differentiate teachers at the very bottom of the value-added distribution for ELA. In math, scores on modeling and strategy instruction differentiate teachers at both the top and bottom of the value-added distribution. However, there are few discernible relationships between scores on teaching practices and student achievement gains in either subject when I run the same analyses using value-added based on the supplemental assessments, which target conceptual or "higher-order" reasoning. I hypothesize these differential patterns are driven by strong district control over classroom instruction and tight coupling between mandated curricular materials and the state assessment. What material is modeled and what strategies are taught, and the extent to which they align with material covered on a given assessment, likely influence the relationship between teaching practices and student achievement gains. Like modeling and strategy instruction, classroom discourse can take varied forms in math and ELA classrooms. I show how one might draw different conclusions about the cross-subject consistency of discourse patterns in math and ELA classrooms based on quantitative and qualitative data. The descriptive statistics and distribution of discourse scores suggest that teachers orchestrate discourse to a similar degree in math and ELA. However, more detailed descriptions of the ways in which teachers orchestrate discussions reveal notable differences both within and across subjects, particularly among lessons that score higher. Our understanding of the extent to which the subject matters in orchestrating discourse depends heavily on the method of analysis and on the quality of the discussion being analyzed. Based on these findings, I argue that teacher education provides a vital opportunity to make cross-content linkages and promote flexible use of instructional practices across the elementary school curriculum. My data suggests that teachers may benefit from support in when and how to model language arts concepts and that more frequent and higher quality modeling may be associated with student achievement gains. The extent to which teachers already model mathematical material provides a potentially useful bridge for incorporating modeling in language arts instruction. Given the limited duration of teacher preparation, we are well served to identify practices that could be targeted for instruction in multiple content areas. This study also suggests the importance of student outcome measures in understanding the complex domain of teacher quality. Different assessments privilege different content and skills, and as a result, they are likely differentially sensitive to specific teaching practices. As states rush to develop assessments that align with the Common Core standards, it will be important to build a theoretical and empirical base about how these new assessments can and should factor into modifications of existing teacher evaluation systems.
Description
| Type of resource | text |
|---|---|
| Form | electronic; electronic resource; remote |
| Extent | 1 online resource. |
| Publication date | 2013 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Associated with | Cohen, Julia Jackson |
|---|---|
| Associated with | Stanford University, Graduate School of Education. |
| Primary advisor | Grossman, Pamela L. (Pamela Lynn), 1953- |
| Thesis advisor | Grossman, Pamela L. (Pamela Lynn), 1953- |
| Thesis advisor | Borko, Hilda |
| Thesis advisor | Loeb, Susanna |
| Thesis advisor | Shavelson, Richard J, 1942- |
| Advisor | Borko, Hilda |
| Advisor | Loeb, Susanna |
| Advisor | Shavelson, Richard J, 1942- |
Subjects
| Genre | Theses |
|---|
Bibliographic information
| Statement of responsibility | Julia Jackson Cohen. |
|---|---|
| Note | Submitted to the Graduate School of Education. |
| Thesis | Thesis (Ph.D.)--Stanford University, 2013. |
| Location | electronic resource |
Access conditions
- Copyright
- © 2013 by Julia Jackson Cohen
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...