Back to Top

Do Financial Incentives Crowd Out Intrinsic Motivation to Perform on Standardised Tests?

By: John A. List, Jeffrey A. Livingston, and Susanne Neckermann

Economics of Education Review (October 2018)

Published version

Manuscript version (free)

Commentary by Gabriel Heller-Sahlgren

There is considerable debate in the education research and policy communities regarding the extent to which the use of extrinsic incentives, such as financial rewards, could induce higher pupil performance. One of the arguments in favour of providing such incentives is that they could help increase pupils’ motivation to perform and in this way help them reach their true potential. Normally, the returns to education take time to materialise, which may have negative consequences for pupil motivation, either because pupils do not possess accurate information regarding the long-term returns or because they have high discount rates – that is, they may discount future rewards in their decision making regarding the level of effort they choose to exert on tasks today. By providing external incentives to do well on certain tasks, pupils are given more immediate rewards to their effort and the returns to higher performance are made more salient. If incentives instead are provided to other actors, such as parents and teachers, these actors will have more motivated to help pupils improve their performance (potentially by improving pupil motivation).

Yet others argue that external incentives may very well backfire. This is because such incentives could crowd out actors’ internal motivation, thereby either having no impact or even a net negative effect on performance, both on the incentivised task and similar future tasks. If this holds true, once the external incentives are removed, the actors may well very well have lower motivation to perform than had been the case without the introduction of incentives in the first place.

In this paper, the authors analyse the results of a randomised experiment conducted among pre-school, primary, and lower-secondary pupils in Chicago, testing the impact of financial incentives on pupil performance in an otherwise low-stakes test. Simultaneously, the pupils also take another high-stakes test designed to measure the same content but without any financial incentives attached. Importantly, the pupils sit the second, non-incentivised test twice, first in conjunction with the test to which the incentives are attached, and then again one year afterwards. This allows the authors to test whether or not the incentives attached to the low-stakes test decrease performance on the non-incentivised test in a short-term and longer-term perspective.

In the experiment, pupils were allocated to either a control group or to one of five separate treatment groups. In the first three treatments, either the pupil, their parents, or their tutor were paid $90 if the pupil improved their performance on the low-stakes test relative to their baseline performance, maintained their course grade in the subject, as well as avoided unexcused absences and suspensions. In the fourth treatment, both pupils and parents were given $45 each if the pupils reached these targets, and in the fifth treatment, pupils, parents, and the tutor were paid $30 to do so.

The authors find a large positive impact on the low-stakes test to which incentives were attached, with similar effect sizes regardless of the specific treatment group to which pupils were assigned. The results indicate that pupils performed the equivalent of about 30-37 PISA points better on this test as a result of the financial incentives. At the same time, there is no impact at all on the high-stakes test to which there were no financial incentives attached in the short term. If anything, the estimates are mostly negative, although they are not statistically significant. This suggests that the financial incentives may in fact have decreased the actors’ internal motivation to do well.

Yet when analysing the effects on the high-stakes test in the follow up a year later, the authors find large positive gains. In other words, the incentives attached to the otherwise low-stakes test appear to have had a long-term impact on learning without incentives, indicating that the negative impact on internal motivation was only temporary. In the longer-term perspective, the financial incentives provided a net gain in terms of performance on tests to which no incentives were attached.

In addition, the authors find suggestive evidence that the effect of incentives differs depending on whether or not pupils have high or low internal motivation at baseline. While financial incentives appear to have no or even negative effects on internal motivation among all pupils in the short term, the long-term benefits appear concentrated among pupils with high performance at baseline. Still, there is no negative impact among pupils with lower performance at baseline in the longer-term perspective, indicating that the negative effect on their internal motivation was temporary as well.

Overall, the paper is an important contribution to an expanding literature on the effects of financial incentives in the education system. Certainly, as other research is mixed and tends to suggest that the effects of external incentives depend on how they are  framed and structured, more research is necessary before drawing strong specific conclusions for policy and practice. Nevertheless, the paper does suggest that such incentives at least are likely to have some role to play, a finding that also chimes well with recent randomised research in English schools.

--

Gabriel Heller-Sahlgren is Lead Economist at CfEE and Editor of its Monthly Research Digest.                                                                                     

Subscribe here to get monthly email notifications of the release of each issue.

You can download free copies of back issues of the Digest here.

Blog Category: 
About the author