Using the High School Longitudinal Study to Assess Trends and Outcomes of Teacher Collaboration

• In Event: Excellence in Education Research: Early-Career Scholars and Their Work

Abstract

Research on instructional improvement and teacher professional learning reveals that teachers are most likely to improve their skills and increase student learning when they collaborate with colleagues. Yet, such collaboration requires schools to invest significant resources – e.g., time for teachers to collaborate, scheduling to accommodate collaboration, and forgoing other possible forms of professional development, among many other tradeoffs. In this regard, instructional collaboration is a sizable investment for schools. Yet, if instructional collaboration does positively impact instruction, then it should be widespread. This is particularly true in math and science, as national initiatives are focusing strongly in these areas because of the need for more high school graduates with strengths and interests in the STEM fields of Science, Technology, Engineering, and Math. However, we do not yet know whether, where, and in what ways the trend toward teacher collaboration is playing out in schools, how it is related to the distribution of teachers and students across schools, and whether it is having its intended impact on student engagement and learning. In other words, although the research supports collaboration, are teachers actually collaborating? If so, how? Who? And to what end?

This study begins to answer these questions by assessing these practices in a national sample of high school math and science departments using the first two waves of data from the High School Longitudinal Study: 2009 (HSLS:09). Analyses begin by examining the nature and prevalence of collaboration in high school math and science departments, using principal components analysis to examine sub-constructs within the HSLS items on teacher collaboration, and identifying the characteristics of high schools with the highest levels of collaboration. Further, this study attempts to linking teacher collaboration to the critical outcomes of student engagement in math and science classes, students’ senses of identity and efficacy in math and science, and students’ levels of algebraic understanding. This study looks longitudinally at these relationships among a cohort of 21,444 ninth-grade students, assessing the relationship between teacher collaboration in the baseline year when students were in ninth grade and changes in student outcomes two years later when most students were in eleventh grade. Analyses include propensity score matching to estimate a causal impact of teacher collaboration on increases in student engagement, efficacy, identity, and learning over time.

Preliminary findings reveal that the 12 items measuring professional learning community on the teacher survey in HSLS represent three dimensions of practice related to teacher collaboration within math and science departments in high schools: (a) the extent to which teachers share and discuss their practice, (b) the level of cohesion within the department, and (c) the extent to which teachers collaborate in regards to supporting high-needs students. Correlations among these practices range from 0.59 to 0.68, revealing that they often occur together but do not necessarily do so. Additionally, findings show that science teachers are significantly more likely than math teachers to report that their departments share research on effective teaching methods, particularly for supporting English Language Learners – practices that we would expect to support quality instruction. However, math departments are more likely to coordinate content and feel supported by department chairs.

In terms of where teachers are most likely to collaborate, findings show few significant differences for math teachers based on the characteristics of students they serve – yet significant variation among science teachers. Collaboration in the form of sharing and discussing teaching practice is less common among science teachers serving greater proportions of black, Latino, and American Indian students and among science teachers serving more students receiving special education. However, controlling for other student characteristics, science teachers working with greater proportions of students receiving free or reduced-price lunch were more likely to share and discuss their teaching than those working with wealthier students. Among math teachers, there were no significant differences along any of these dimensions. Additionally, science teachers were more likely to share and discus their teaching in larger schools, whereas school size was not related to collaboration among math teachers. Interestingly, math teachers in alternative schools were much more likely to collaborate than those in regular schools. This was the predictor that was significantly related to math department collaboration.

Preliminary findings also reveal five types – or clusters – of schools in terms of collaboration among math and science teachers: (1) high collaboration schools in which math and science teachers all report collaborating at relatively high levels, (2) schools in which only science teachers report high levels of collaboration, (3) schools in which only math teachers report high levels of collaboration, (4) schools in which science teachers report extreme disconnection, and (5) schools in which math teachers report extreme disconnection. Findings show that schools with collaborative math teachers (where science teacher collaboration is at the mean) tend to be in the most academically oriented communities, having the highest percentage of residents with a Bachelors degree or higher, the highest levels of college enrollment for the class of 2008, a relatively high percentage of students taking Advanced Placement (AP) classes, the highest eighth-grade math and science grades among incoming ninth-graders, and the highest levels of baseline algebraic understanding among ninth-graders. By contrast, schools with the most disconnected math teachers (where science collaboration is also relatively close to the mean) tend to be in communities that are much less academic in nature, with the lowest percentage of adults in the community having a college degree, with the fewest students taking AP classes, with students coming in with the lowest eighth-grade grades in math and science, and with the lowest baseline levels of algebraic understanding among ninth-graders. These findings reveal clear patterns regarding the types of schools and communities in which math teachers are more or less likely to collaborate, holding science teacher collaboration relatively constant.

In regards to the student benefits of teacher collaboration, preliminary findings show that teacher collaboration is not strongly predictive of enhanced student outcomes from the ninth to eleventh grades (including increased engagement in math and science, increased feelings of efficacy in math and science, an increased sense of identity as a math or science person, and increased algebraic understanding). This is true of all three dimensions of collaboration in HSLS: sharing and discussing teaching, having a cohesive department, and focusing on high-needs students. Additionally, analyses using propensity score matching (to reduce some of the selection bias) also found very little impact of teacher collaboration on student outcomes, although there is some evidence that particular collaborative practices among math teachers – including discussing student work and discussing lessons that were not successful – can supported increased feelings of math efficacy among students.

Collectively, these preliminary findings suggest that teachers are collaborating to various levels and in various ways around the country – and that such collaboration tends to be particularly high among math teachers working in academically oriented schools and communities. However, we are not yet seeing extensive outcomes from teacher collaboration broadly.

Authors

• Kristy S. Cooper, Michigan State University

• Amy Auletto, Michigan State University