The importance of teacher quality in student academic achievement is well-established (Chetty et al., 2014), but countries across and beyond Europe are struggling to improve the quality of their teacher workforce amid chronic teacher shortages, economic reorganization, and a surge in globalization that demands new teaching competencies (OECD, 2024). While efforts to affect the quality of the teacher workforce at later points in the teacher pipeline- via improved pre-service teacher education, hiring practices, and professional development- continue to grow globally, identifying and attracting higher-quality prospective teachers to the profession remains a comparatively efficient approach to improving teacher quality (Bartanen & Kwok, 2023). Considering the substantial evidence that individuals’ pre-career characteristics are effective signals of teaching quality (Jacob et al., 2018; Rockoff et al., 2011), and given that schools in many European countries struggle to provide students with sufficient numbers of effective teachers (OECD, 2025), research-based, early-timepoint teacher recruitment strategies appear to be an underutilized lever to improve teacher quality.

Despite this, research examining the job-choice processes and preferences of young people considering teaching   is relatively scarce. The present paper seeks to fill this gap, asking broadly: (1) How do preferences for specific job traits affect young people’s consideration of a career in teaching? and (2) What are the implications of these preferences for crafting recruitment programs that contribute to building a higher quality teacher workforce? To answer these questions, we conduct an original discrete choice survey experiment with a national sample of early-stage university students in the United States. This approach allows us to capture the causal effects of specific job attributes (such as salary, professional autonomy, and job security) on the probability that a student prefers a teaching job over their ideal post-graduation occupation (Hainmueller et al., 2013).

This study extends recent methodologically similar work on preferences for job traits in teaching in several key ways. First, teacher recruitment strategies are most effectively implemented before individuals have invested considerable time, study, or finances into another field of work; interventions after such investments have been made are likely to be less effective, as students fall prey to the sunk-cost fallacy and the path dependence of occupational choice (Dlouhy & Bliemann, 2018; Redekopp, 2016). However, existing work in this area relies on samples consisting primarily of individuals who have already begun specialized studies in their post-graduation occupational roles (e.g., De Cort & De Witte, 2025; Sims & Routledge, 2025). In considering alternative job options, students at this stage in their studies must weigh potentially attractive attributes of a theoretical teaching job against the financial and temporal resources they have already invested in their planned career path. This fact potentially limits the external validity of these studies’ findings to students’ decision-making processes at slightly earlier timepoints, when students have yet to invest significant time or resources into a career path and when recruitment strategies are likely to be most effective. By limiting our sample to students currently in the first 2-4 semesters of their university enrollment, the present study provides new direct evidence on the job trait preferences of students during this pivotal early period.

This study also builds on prior work by De Cort and De Witte (2025) to assess job attribute preference heterogeneity along individual characteristics (e.g., cognitive ability) that are predictive of teaching quality (Rockoff et al., 2011). By assessing preferences for job traits among subgroups of respondents with highest expected teaching quality, we make recommendations for teacher recruitment policies most likely to increase the average teacher quality of the teacher workforce.

We conduct a discrete choice experiment, administered through electronic survey, to a national sample of domestic students in their second to fourth semester in bachelor’s degree programs in the United States. Respondents are presented with a series of “choice tasks”, each of which presents two job profiles with a list of attributes (such as salary, professional autonomy, and job security) describing each job and are asked to select the job they would prefer. Job attributes were chosen to represent a mix of incentives with significant evidence of effectiveness in recruiting individuals to teaching, as well as those that have not yet been empirically tested. These attributes draw on work in the field of teacher motivation (Mancenido, 2021; Watt & Richardson, 2007)), economics literature on occupational choice (e.g., Wiswall & Zafar, 2017), and the relatively small literature empirically examining incentives into teaching (e.g., Christian et al., 2024). By 1) ensuring that every choice task contains one job profile labelled a teaching role and one profile labelled with the respondent’s preferred post-graduation job field and 2) randomly varying the values of job attributes, we estimate the “average marginal component effect” (AMCE) of each attribute on the probability that a respondent chooses the teaching job over their preferred post-graduation job field (Hainmueller et al., 2014). Estimates are calculated by regressing a binary variable indicating whether the respondent selected the teaching role in a given choice task on the vector of included job attributes. In line with Lovison & Mo (2024), we use ordinary least squares regression and cluster standard errors at the respondent level to account for correlation between multiple choice tasks completed by a single respondent. The coefficient on each job attribute, or AMCE is our coefficient of interest and is an estimate of that attribute’s causal effect on the probability that a respondent prefers the teaching job profile over their preferred post-graduation field. By dividing each AMCE by the coefficient on the “salary” job attribute, we also calculate a “willingness-to-pay” for each attribute, which can be interpreted as the salary amount in USD that participants are willing to forgo in order to work in a job with a specific attribute (for example, a job with high professional autonomy). To estimate preferences among those with high predicted teaching quality, we perform subgroup analyses by students’ academic characteristics using the same model specified above.

Data collection has not yet taken place but is scheduled to occur in the coming months. However, we hypothesize that, in line with prior findings such as those by Sims & Routledge (2025), we expect that extrinsic job attributes such as salary will exert significant power over students’ preferences. Additionally, in line with De Corte and De Witte (2025), we anticipate a preference for higher professional autonomy among students with high predicted teaching quality. Finally, in line with Lovison and Mo (2024), we anticipate that students may place monetary value on some job attributes (such as maximum lifetime salary or loan repayment assistance) that is higher than the real cost of these attributes, signaling a potentially cost-effective policy lever. However, due to the novelty of our sample, the consistency of prior work’s findings to the present study remains ambiguous.

Bartanen, B., & Kwok, A. (2023). From interest to entry: The teacher pipeline from college application to initial employment. American Educational Research Journal, 60(5), 941–985. https://doi.org/10.3102/00028312231185733 Christian, A., Ronfeldt, M., & Zafar, B. (2024). College Students and Career Aspirations: Nudging Student Interest in Teaching. NBER Working Paper No. 32641. De Cort, W., & De Witte, K. (2025). To teach or not to teach? Preferences for working conditions and the alternative career opportunities of pre-service teachers. European Journal of Teacher Education, 1–22. https://doi.org/10.1080/02619768.2025.2571930 Dlouhy, K., & Biemann, T. (2018). Path dependence in occupational careers: Understanding occupational mobility development throughout individuals’ careers. Journal of Vocational Behavior, 104, 86-97. https://doi.org/10.1016/j.jvb.2017.10.009 Hainmueller, J., Hopkins, D.J., & T., Yamamoto, (2014). Causal inference in conjoint analysis: Understanding multidimensional choices via stated preference experiments. Political Analysis, 22(1). Pp. 1-30. https://doi.org/10.1093/pan/mpt024. Chetty, R., Friedman, J. N., & Rockoff, J. E. (2014). Measuring the Impacts of Teachers I: Evaluating Bias in Teacher Value-Added Estimates. American Economic Review, 104(9), 2593–2632. https://doi.org/10.1257/aer.104.9.2593 Jacob, B. A., Rockoff, J. E., Taylor, E. S., Lindy, B., & Rosen, R. (2018). Teacher applicant hiring and teacher performance: Evidence from DC public schools. Journal of Public Economics, 166, 81–97. https://doi.org/10.1016/j.jpubeco.2018.08.011 Lovison, V. S., & Hyunjung Mo, C. (2024). Investing in the Teacher Workforce: Experimental Evidence on Teachers’ Preferences. American Educational Research Journal, 61(1), 108–144. https://doi.org/10.3102/00028312231208956 Mancenido, Z. (2021). How high achievers learn that they should not become teachers. Harvard Educational Review, 91(4), 433-547. OECD (2024). Education Policy Outlook 2024: Reshaping Teaching into a Thriving Profession from ABCs to AI. OECD Publishing, Paris, https://doi.org/10.1787/dd5140e4-en. OECD (2025). Unlocking High-Quality Teaching. OECD Publishing, Paris, https://doi.org/10.1787/f5b82176-en. Redekopp, D.E. (2016). Irrational career decision-making: Connecting behavioural economics and career development. British Journal of Guidance and Counseling, 45(4), 441-450, https://doi.org/10.1080/03069885.2016.1264569. Rockoff, J.E., Jacob, B.A., Kane, T.J., & Staiger, D.O. (2011). Can you recognize an effective teacher when you recruit one? Education Finance and Policy, 6(1), 43-74. https://doi.org/10.1162/EDFP_a_00022 Sims, S. & Routledge, C. (2025). Understanding the decision (not) to become a teacher. UCL CEPEO Working Paper No. 25-15. Watt, H. M. G., & Richardson, P. W. (2007). Motivational Factors Influencing Teaching as a Career Choice: Development and Validation of the FIT-Choice Scale. The Journal of Experimental Education, 75(3), 167–202. https://doi.org/10.3200/JEXE.75.3.167-202 Wiswall, M., & Zafar, B. (2015). Determinants of College Major Choice: Identification using an Information Experiment. The Review of Economic Studies, 82(2), 791–824. https://doi.org/10.1093/restud/rdu044