Search
Browse By Day
Browse By Time
Browse By Person
Browse By Policy Area
Browse By Session Type
Browse By Keyword
Browse Artificial Intelligence Presentations
Program Calendar
Sign In
Search Tips
Teacher quality is among the most important determinants of students’ academic achievement and long-term outcomes (Araujo et al. 2016; Chetty et al. 2014). In many countries, licensure tests serve as gatekeeping tools to screen out potentially low-performing teachers before they enter the profession (Jackson et al. 2014). Not all educational systems require such assessments, however, making it difficult for policymakers to forecast teacher sorting and performance. In these contexts, reliable predictions that link pre-entry skills to career outcomes are critical for policy (Harris & Sass 2011; Neilson et al. 2022).
This study examines Chile’s Evaluación Nacional Diagnóstica de la Formación Inicial Docente (END), a mandatory exit exam for prospective teachers, to address two central questions. First, we assess the relationship between candidates’ content knowledge and pedagogical skills on early-career outcomes, including on-time graduation, retention, and the characteristics of the schools and students they serve. Second, we quantify the contribution of teaching preparation programs to the skills measured in these exit exams.
Our analysis draws on a comprehensive panel of approximately 40,000 high school students who applied to university programs in Chile between 2013 and 2020 and enrolled into a teacher education program. We link these data to higher education trajectories, END scores, graduation, and teaching employment, supplemented by school-level indicators of student composition, standardized test performance, and school climate.
To address the first question, we regress early-career outcomes on exam performance, controlling for teachers’ observed characteristics, including age, gender, and college admission exam scores. Program and cohort fixed effects account for unobserved differences across teacher preparation programs. For the second question, we estimate an education production function in which teaching skills are modeled as a linear function of students’ pre-college skills, cohort effects, and program effects.
Our analysis reveals that END scores predict early-career outcomes, but the two exam dimensions matter in distinct ways. Subject knowledge scores predict a higher probability of on-time graduation and sorting into private schools and those serving higher-performing students. Pedagogical skills, by contrast, are more strongly linked to labor supply outcomes. Program fixed effects account for 13% and 25% of the variance in pedagogical skills and subject knowledge, respectively. Moving a student from the average program to one at the 84th percentile improves performance by 0.3 and 0.49 standard deviations on each dimension.
These findings carry important policy implications. First, exit exams for prospective teachers effectively predict teacher trajectories during a critical window for professional development and career decisions. Second, the modest share of variance in teaching skills explained by programs underscores the need for stronger accountability policies in teacher education to improve prospective teachers’ skills before they enter the classroom.