Summary of Findings: No Positive Impacts on Composition of the District Teaching Workforce After One Year

The last major research question concerned the impact of comprehensive teacher induction on the composition of the teaching workforce in the district. For comprehensive teacher induction to affect the composition of the district’s teaching workforce, it has to produce a difference in the types of teachers who decide to return to the district. As teachers leave the district, the average qualifications of the teachers who remain in the district begin to change, perhaps differentially between the treatment and control groups. We tested this hypothesis by comparing the characteristics of district stayers between the treatment and control groups along three dimensions: (1) their observed classroom practices; (2) their effect on student achievement; and (3) their professional characteristics such as SAT/ACT scores and advanced degrees. Classroom practice and student achievement outcomes are regression-adjusted using the same sets of covariates used in the main analysis.

We found that the treatment had no positive impacts on the classroom practices, no positive impacts on student achievement (and one statistically significant negative impact), and no significant impacts on the professional background characteristics. Table 5 presents the impacts on classroom practices and student achievement outcomes for district stayers. Table 6 shows the background characteristics of teachers by mobility status.

Correlational Analyses Explore Relationships between Induction and Outcomes

Because a majority of both treatment and control teachers reported receiving induction support (93 percent of treatment teachers and 75 percent of control teachers reported having an assigned mentor), we looked at the relationship between the types and intensity of support and our three main outcome measures: classroom practices, student achievement, and teacher retention. These nonexperimental analyses investigate whether there was a relationship between induction support and outcomes, regardless of treatment status. The analyses mimic the experimental analyses discussed above, using the same covariates and model specification, but replacing the indicator for assignment to treatment status with a measure of induction services. We re-ran the model once for each of 12 measures of induction services measured from the fall survey and again using the measures from the spring survey.⁹ The results from these analyses should be viewed cautiously. They should be used to generate hypotheses rather than to establish causal inferences because any association may confound effects of the induction services themselves with the pre-existing differences between the types of teachers who receive different levels of services. For example, those who receive the most support may be the most assertive and effective teachers who are most attached to the profession. Due to the number of analyses conducted, we focus upon the relationships that are statistically significant after applying a Benjamini- Hochberg correction for multiple hypothesis testing within each of the three main outcome domains.

After adjusting for multiple hypothesis testing, none of the relationships between the induction variables and classroom practices was statistically significant. Three of the relationships between the induction variables and student test scores and eight of the relationships between the induction variables and retention measures were positive and statistically significant. Specifically, students of teachers who reported meeting with a subject coach in the fall scored higher on math tests by 0.14 of a standard deviation. The students of teachers who reported receiving feedback on teaching during the fall scored higher on both math and reading tests by 0.02 of a standard deviation per instance that the teacher received feedback. Having an assigned mentor in the spring, receiving guidance in math or literacy content in the spring, each hour spent in the fall on professional development related to content area knowledge, and each hour spent in the fall and spring on professional development related to instructional techniques were associated with a 1 to 6 percentage point increase in the likelihood of remaining in the district or in the teaching profession.

Top

---

⁹ The twelve induction measures were: whether the beginning teacher was assigned a mentor (yes/no), whether the beginning teacher met with a literacy or math coach (yes/no), whether the beginning teacher worked with a study group (yes/no), whether the beginning teacher observed others teaching (yes/no), whether the mentor gave the beginning teacher suggestions to improve his/her practices (yes/no), whether the beginning teacher received a “moderate amount” or “a lot” of guidance in math content (yes/no), whether the beginning teacher received a “moderate amount” or “a lot” of guidance in literacy content (yes/no), the frequency with which the beginning teacher received feedback on his/her teaching (number of times in a threemonth period), time the beginning teacher spent in mentoring sessions (hours per week), time the mentor spent observing the beginning teacher teaching (hours per week), time spent on instruction techniques and strategies as part of professional development activities (hours per three-month period), and time spent on content area knowledge as part of professional development activities (hours per three-month period).