Reminders to parents can improve student outcomes

Editor’s note: For a broader synthesis of themes covered in this article, check out our VoxDevLit on Education Technology.

Governments and development organisations frequently rely on informational interventions to change behaviour. From sending reminders to taxpayers to providing farmers with price data or parents with school performance reports, the idea is simple: better information leads to better decisions.

A growing body of research shows these interventions can be remarkably effective (Jensen 2010, Rogers and Feller 2016, Bergman 2021). But an important question remains unresolved: Do these policies work because they change what people know, or because they change what people pay attention to?

This distinction matters for policy design. If information itself is the key ingredient, interventions must transmit accurate, individualised data – which is often expensive to collect, especially at high frequency. If, conversely, it is salience – making an issue top-of-mind – doing much of the work, then much simpler (and cheaper) interventions may achieve similar or even larger impacts.

Information versus salience: why the difference matters

Economists have long recognised that people face attentional constraints: we cannot monitor everything that matters in our lives at once (Gabaix and Graeber 2024, Bordalo et al. 2012). As a result, behaviour often responds to whichever issue is most salient at a given moment.

Informational interventions may therefore influence behaviour through two different channels:

1. Belief updating: people learn new facts and adjust their actions accordingly.
2. Salience: the intervention simply reallocates attention to a different decision domain, potentially leading them to adjust their actions within that domain (which might have otherwise fallen through the cracks).

In practice, separating these channels is difficult because providing information almost inevitably attracts attention as well.

In Lichand et al. (forthcoming), we study the mechanisms behind the impacts of an informational intervention to school parents in Brazil, experimentally comparing messages that provide child-specific information with messages that merely make school attendance salient, without conveying child-specific data. 

A natural testing ground: Parents and school engagement

Communication between schools and parents provides an ideal setting to study informational interventions. Parents care about their children's education but often have limited visibility into daily school effort, particularly as children grow older. This creates a classic monitoring problem: children may skip classes or neglect homework without parents realising.

Previous studies have shown that providing parents with information about their children's school behaviour – attendance, grades, or homework – can improve educational outcomes (Bergman 2021, Dizon-Ross 2019, Berlinski et al. 2016).

But do these interventions work because parents learn new information, or simply because the messages remind them to pay attention to school?

Testing the mechanism: A large field experiment

To answer this question, we conducted a large randomised experiment across 287 public schools in São Paulo, Brazil, involving roughly 19,300 ninth-grade students. Teachers reported weekly information about students’ behaviour in math classes – attendance, punctuality, and homework completion. We randomly assigned parents to receive one of three types of messages via SMS:

1. Child-specific information: These contained child-specific information about school effort. For example: “Nina missed between 3 and 5 math classes over the last three weeks.”
2. Salience: These contained no child-specific information but emphasised the importance of monitoring behaviour. For example: “It is important that Nina attends every math class.”
3. Control:Parents received no messages.

Crucially, this design allowed researchers to isolate the additional effects of information beyond attention.

Most parents were initially uninformed

Like previous research, before the intervention began, parents had very inaccurate beliefs about their children's school attendance. Only about one-third of parents correctly estimated how often their child missed classes. Many were overly optimistic: while more than 40% believed their child had missed no classes in the previous three weeks, the true figure was closer to 15%. This suggests that providing accurate information should, in principle, improve parental monitoring and student outcomes.

And indeed, it did.

Information improved outcomes, but so did salience

Receiving child-specific information significantly improved several educational outcomes:

• Attendance increased by about 2 percentage points, equivalent to two or three additional classes attended.
• Standardised test scores increased by roughly 0.1 standard deviations, comparable to skipping a quarter of a school year ahead of control students.
• Grade repetition decreased by about 1/3.

These results align with earlier findings in the evidence base. But the most striking result was as follows: salience messages improved outcomes by nearly the same amount—even though they contained no child-specific information.

Across outcomes, the effects of salience were 89–126% – as large as those of child-specific information. In other words, most of the impact of the informational intervention appears to have come from making school engagement more salient to parents.

Information changed beliefs, but salience changed behaviour

Our experiment also measured how accurately parents understood their children's attendance. Here the two interventions differed dramatically:

• Information messages improved accuracy substantially.
• Salience messages did not improve accuracy at all.

Yet student outcomes improved under both treatments. This suggests that better knowledge was not the primary driver of behaviour change. Instead, both interventions appear to have triggered increased parental monitoring.

Parents who received messages – whether informative or not – reported:

• asking children more often about school
• encouraging studying more frequently
• discussing grades and classes more regularly

Students also reported spending more time on academic activities. In other words, the messages mobilised parents to pay attention.

Attention constraints explain the pattern

Additional evidence from the experiment reinforces the idea that attention is the key mechanism.

Frequency matters. In a separate experiment, parents who received more frequent engagement messages (none of which conveyed child-specific information) showed substantially larger impacts – consistent with the idea that repeated prompts keep issues salient.

General reminders have broader effects. Messages focused specifically on math behaviour mainly affected math outcomes. In contrast, more general engagement messages improved outcomes across subjects. Again, this pattern fits an attention-based mechanism.

Why salience can outperform information

If attention drives behaviour, salience interventions may sometimes outperform informational ones for two reasons.

1. Lower data requirements: Information interventions require collecting and updating individual-level data – attendance records, grades, performance metrics. In many contexts, particularly in low- and middle-income countries like Brazil, this data are often not available at high frequency. Salience interventions, by contrast, can be implemented without detailed information systems.
2. Higher frequency: Information can only be transmitted as often as it is measured. Salience messages can be sent as frequently as desired, allowing policymakers to repeatedly draw attention to important behaviours. In the Brazilian experiment, higher-frequency messages produced even larger effects.

Implications beyond education – and an important caveat

The insights from our research may extend well beyond school communication. Many policy interventions rely on providing information: reminders about savings, energy consumption feedback, health advice, tax notices, or credit repayment warnings. But if attention is the true mechanism, policymakers should reconsider how they design these interventions.

For example:

• Financial behaviour: reminders about savings goals (as in Allcott and Taubinsky 2015) may work mainly by keeping long-term planning salient.
• Workplace performance: performance reports to managers (as in Rockoff et al. 2012) may ultimately affect hiring/firing decisions because they keep monitoring top-of-mind.
• Credit repayment: warnings about the consequences of credit default (as in Bursztyn et al. 2019) may work rather because they draw attention to enforcement risks.

In many of these contexts, simpler nudges might achieve similar results without costly data collection. This raises, however, a deeper point: if attention reallocation is the key mechanism behind the impacts of these interventions, we are no longer guaranteed that these interventions ultimately improve welfare.

Benkert and Netzer (2018) make that point sharply: whether a nudge improves the quality of decision-making or not depends on the underlying decision process. Ultimately, what is it that it displaces from the top of the mind?

Designing better informational interventions

The broader lesson is not that information is unimportant. In many settings, accurate data are essential. But policymakers should recognise that information policies often work partly because they act as reminders.

Designing effective interventions therefore requires thinking about:

• message frequency
• simplicity and timing
• what the intervention reallocates attention away from

References

Allcott, H, and D Taubinsky (2015), "Evaluating behaviorally motivated policy: Experimental evidence from the lightbulb market," American Economic Review, 105(8): 2501–2538.

Benkert, J-M, and N Netzer (2018), "Informational requirements of nudging," Journal of Political Economy, 126(6).

Bergman, P (2021), "Parent–child information frictions and human capital investment: Evidence from a field experiment," Journal of Political Economy, 129(1): 286–322.

Berlinski, S, M Busso, T Dinkelman, and C Martinez (2016), "Reducing parent-school information gaps and improving education outcomes," Journal of Development Economics, 121: 1–14.

Bordalo, P, N Gennaioli, and A Shleifer (2012), "Salience theory of choice under risk," Quarterly Journal of Economics, 127(3): 1243–1285.

Bursztyn, L, T Fiorin, D Gottlieb, and M Kanz (2019), "Moral incentives in credit card debt repayment," Journal of Political Economy, 127(4): 1641–1683.

Dizon-Ross, R (2019), "Parents' beliefs about their children's academic ability," American Economic Review, 109(8): 2728–2765.

Gabaix, X, and T Graeber (2024), "The complexity of economic decisions," Unpublished manuscript.

Jensen, R (2010), "The perceived returns to education and the demand for schooling," Quarterly Journal of Economics, 125(2): 515–548.

Lichand, G, N Cunha, R A Madeira, and E Bettinger (forthcoming), "When the effects of informational interventions are driven by salience – evidence from school parents in Brazil," American Economic Journal: Economic Policy.

Rockoff, J, D Staiger, T Kane, and E Taylor (2012), "Information and employee evaluation: Evidence from randomized intervention in public schools," American Economic Review, 102(7): 3184–3213.

Rogers, T, and A Feller (2016), "Reducing student absences at scale by targeting parents' misbeliefs," Nature Human Behaviour, 2: 335–342.

Singh, A, L Navarro-Sola, and P Oreopoulos (2025), “Education Technology”, VoxDevLit, 20(1).