Human Development

How teachers’ beliefs can shape motivation and student learning

  • Blog Post Date 06 May, 2024
  • Articles
  • Print Page
Author Image

Jalnidh Kaur

Columbia University

While teacher effort is crucial for student learning, evidence suggests that this is not what teachers themselves believe. This article presents findings from a randomised experiment involving a psycho-social intervention targeted at teachers. It shows that teachers exposed to the intervention exhibit greater confidence in their ability to raise student learning and exert more effort within and outside the classroom – leading to improved student performance. 

Teachers play a pivotal role in shaping the future. Research has shown that teacher effort is an important determinant of how much students learn (Leaver et al. 2021, Muralidharan and Sundararaman 2011). However, do teachers believe the same about themselves and their efforts? 

High proportions of teachers across multiple low- and middle-income countries echo a sentiment of helplessness in self-reported surveys. For example, over 40% of teachers across nine developing countries surveyed by the World Bank report that there is little they can do to help students learn if students come from disadvantaged backgrounds characterised by financial constraints or low levels of parental education (Sabarwal et al. 2022). Given that these settings primarily comprise first-generation learners with no parental support, these beliefs generate worrying implications for teaching and learning in the classroom. 

The concept of ‘perceived control’

Such beliefs have been extensively studied by psychologists under the broad construct of ‘perceived control’ – self-beliefs about one’s ability to influence outcomes. Beliefs about perceived control have been shown to be key drivers of motivation and behaviour (Skinner 1985, Weiner 1985). When people perceive greater control in situations, they are more likely to take interest, initiate action, and persist in the face of failures. On the contrary, when people perceive situations as out of their control, they are more likely to withdraw, retreat, and give up. It is possible that the presence of multiple constraints outside the control of teachers such as the presence of disadvantaged learners, and large variation in learning levels within the classroom, could be leading teachers to internalise low levels of control in influencing learning.

In my research (Kaur 2024), I ask two questions: (i) What is the role of these beliefs in shaping teacher effort and student learning? (ii) Are these beliefs malleable, and can these change through a targeted intervention?

The study

I leverage insights and approaches from positive psychology, to design an intervention that targets teachers’ beliefs about perceived control in everyday situations. This intervention is in collaboration with WorldBeing, an organisation that creates evidence-based programmes in psychology for adults and youth in developing countries. Through interactive lectures, worksheets, and group sessions, the programme trained participants to recognise differences between controllable and uncontrollable situations and aimed at building skills to navigate each of these. Partnering with a local non-governmental organisation, the programme curriculum was adapted to the context such that it could be delivered to teachers. Importantly, the curriculum had no references to teaching or pedagogy. This was an intentional design feature to minimise the scope for experimenter demand effects (that is, teachers reporting answers that they believed surveyors wanted to hear). 

The study involves a randomised controlled trial across a rural school chain in northern India during the academic year 2022-23. Teachers are randomised to receivethe psycho-social intervention (‘treatment group’) or an informational intervention unrelated to personal development (‘control group’) – in a similar format over the same duration. The training is imparted online over Zoom, in 10 sessions spread across five weeks. All sessions are conducted after school hours to avoid crowding out of teacher time on other tasks. 

I measure teacher and student-level outcomes across multiple follow-up rounds. A first-order problem with measuring beliefs is that self-reports may not be credible as people may not have incentives to report truthfully. To tackle this concern, I elicited beliefs using a ‘revealed preference’ approach that lets teachers have skin in the game. The conceptual idea behind this is as follows: if you believe your effort can change outcomes, you would be more willing to prefer an incentive-based contract over a flat-pay contract. I present teachers with a menu of choices between a flat bonus and a performance-linked one, which they would receive on top of their salary. Based on the choices that teachers make in successive rounds with increasing stakes, I elicit teachers’ beliefs about their perceived control over education production.  

Given that effort is multidimensional, I captured teacher effort using a range of measures, including attendance records, classroom observations, and reviews of homework notebooks of students assessing the quality of feedback provided by teachers – in addition to self-reports of time-use. I measure student learning using scores in the end-of-year Math exams, which were centralised across schools and externally graded.                                                           


I find that exposure to psycho-social intervention leads teachers to positively update their beliefs about how much their effort translates into learning. Compared to the control group, treatment-group teachers exhibit a 14% increase in confidence in their abilities to raise student learning – as deduced by their choices in the incentivised belief-elicitation task. In other words, teachers in the treatment group were more likely to forego a monetary bonus received with certainty in exchange for a higher-paying monetary bonus that was conditional on the test-score improvement of low-performing students in their classroom – revealing higher confidence in their ability to raise student scores by their effort. It may be noted here that this is a setting where all exams are externally graded, so teachers are unable to manipulate test scores in their favour. This shift in beliefs persists six months after the intervention and is not driven by changes in risk attitudes1

Irrespective of how effort is measured, I find that teachers in the treatment group exert more effort both inside and outside the classroom. This is reflected in classroom observations conducted by treatment-blind observers where these teachers are seen putting in more work in facilitating engagement and are also more likely to adopt good pedagogical practices. I also find that teachers in the treatment group put in more effort outside the classroom, as reflected in the quality of feedback (level of detail) provided to students in graded work. The effects are substantive in terms of magnitude. 

Finally, I show that students taught by teachers who receive the treatment scored 0.09 SD (standard deviations)2 higher in the end-of-year exams. The gains in learning are broad-based, that is, students across the ability distribution achieve an increase in test scores. In particular, gains are driven by students who are taught by teachers with low levels of control beliefs at baseline, in other words, those with the most room for improvement. Investigating the mechanisms, I find suggestive evidence that the effects are driven by teachers’ beliefs about perceived control, as opposed to other channels, such as risk preferences, beliefs about the malleability of intelligence (growth mindset), or mental health. While I find no effect on growth mindset beliefs, I see a positive but statistically insignificant effect on mental health, which cannot fully explain the effects on teacher behaviour. On the other hand, I find strong positive effects on teachers’ beliefs about the perceived importance of teaching inputs relative to other inputs, as well as teachers’ locus of control as captured by standard psychological scales, which provide suggestive evidence that the intervention worked by influencing teachers’ beliefs about the mapping between their effort and outcomes.                             

Discussion and policy implications

These findings are striking, especially given the fact that the training did not outline any expectations from teachers to exert more effort, and nor did it provide any skills related to teaching. This was not an intervention about what to teach or how to teach. Yet, we see these behavioural effects on teacher effort suggesting that self-beliefs about effort had a role to play. 

These findings are relevant for policy, especially in the context of teachers’ professional development in developing countries. While countries undertake massive spending to invest in teacher capacity-building through in-service training, a wide range of evidence suggests that these programmes fail to produce systematic improvements in instructional practice and student achievement. A possible reason suggested for this failure is that existing programmes are traditionally focused on targeting skill deficits by providing content and pedagogical knowledge using overly theoretical content (Loyalka et al. 2019). In contrast, there is a limited focus on teacher motivation and building skills around enabling teachers to effectively deal with the unique challenges of their classrooms. 

My results on teacher effort using a purely psychological intervention also suggest that teachers might have innate content and pedagogical knowledge to begin with but lack the drive to use these skills in the status quo. Thus, using targeted psychological content that enables teachers to perceive themselves as active agents who can influence learning may provide a promising option to influence teacher productivity. Additionally, given the low costs of training, these results hold important policy implications for scale-up by embedding these modules into regular professional development programmes. 

In sum, the big-picture takeaway is that teacher beliefs matter and that these beliefs are malleable. While prior approaches that target teacher effort have largely focused on extrinsic motivation using monetary incentives and monitoring, this work shows that alleviating internal, psychological constraints can be a powerful lever for raising teacher effort and student learning. Teachers operating in complex and challenging environments are constrained by self-limiting beliefs about their ability to raise student learning. It is critically important to empower teachers and support them in their work by building psychological reserves and increasing perceived control. 

I4I is now on  Substack. Please click here (@Ideas for India) to subscribe to our channel for quick updates on our content


  1. I measure risk preferences, both at baseline and endline, using an experimental task (Charness et al. 2013). The task provides a menu of options and asks teachers to choose between a safe payment and a lottery.
  2. Standard deviation is a measure used to quantify the amount of variation or dispersion of a set of values from the mean (average) value of that set. 

Further Reading

  • Charness, Gary, Uri Gneezy and Alex Imas (2013), “Experimental methods: Eliciting risk preferences”, Journal of Economic Behavior & Organization, 87: 43-51. Available here.
  • Kaur, J (2024), ‘How much do I matter? Teacher self-beliefs, effort, and student learning’, Job Market Paper. Available here.
  • Leaver, Clare, Owen Ozier, Pieter Serneels and Andrew Zeitlin (2021), “Recruitment, Effort, and Retention Effects of Performance Contracts for Civil Servants: Experimental Evidence from Rwandan Primary Schools”, American Economic Review, 111(7): 2213-2246.
  • Loyalka, Prashant, Anna Popova, Guirong Li and Zhaolei Shi (2019), “Does Teacher Training Actually Work? Evidence from a Large-Scale Randomized Evaluation of a National Teacher Training Program”, American Economic Journal: Applied Economics, 11(3): 128-154.
  • Muralidharan, Karthik and Venkatesh Sundararaman (2011), “Teacher Performance Pay: Experimental Evidence from India”, Journal of Political Economy, 119(1): 39-77. A version of the article is available here.
  • Sabarwal, Shwetlena, Malek Abu-Jawdeh and Radhika Kapoor (2022), “Teacher Beliefs: Why They Matter and What They Are”, The World Bank Research Observer, 37(1): 73-106.
  • Skinner, Ellen A (1996), “A guide to constructs of control”, Journal of Personality and Social Psychology, 71: 549-570.
  • Weiner, Bernard (1985), “An attributional theory of achievement motivation and emotion”, Psychological Review, 92(4): 548-573.
  • Weiner, Bernard (2000), “Intrapersonal and Interpersonal Theories of Motivation from an Attributional Perspective”, Educational Psychology Review, 12(1): 1-14.


No comments yet
Join the conversation
Captcha Captcha Reload

Comments will be held for moderation. Your contact information will not be made public.

Related content

Sign up to our newsletter