Our intervention aimed to create a cost-effective, scalable, evidence-based solution that addressed some of the key behavioural barriers to better school management practices.

We also aimed to create an innovative intervention to make it easier for principals in resource-constrained environments to better support their teachers.

The intervention was inspired by the work of Roland Fryer, which shows that school management can have a crucial effect on student performance i.e. a 300-hour training for principals in some basic management practices (including training on lesson planning, data-driven instruction, and teacher observation and coaching) can improve scores on standardised maths and tests by at least 0.08 standard deviations.

The promise of this evidence excited us and creating a low-intensity, scalable version of this training, complemented by the behavioural science literature and qualitative work to make it context-led, motivated us.

This innovation has been prototyped and tested rigorously through a randomised control trial. We are now looking for opportunities to scale, further adapt and replicate these results.

We thank the Global Innovation Fund for their support and the grant that made this work possible as part of a wider capacity-building effort in Guatemala, Bangladesh and Indonesia

We also would like to thank the team that work to design and deliver this project including Alan Palala and Luisa Muller in DIGEDUCA Ministry of Education, DIGEMOCA for their support with data collection and school visits, and BIT Stewart Kettle, Johannes Lohmann, Monica Wills Silva, Daniela Mendez, Joseph Cole

The main beneficiaries of the intervention were principals and teachers in Guatemala. Principals in the treatment condition conducted important school management procedures more often. The intervention significantly increases the frequency of formal planning sessions and class observations conducted by principals (as reported by teachers). For instance, The guidance resulted in an increase of 2.3 percentage points or 10% in principals conducting formal planning sessions with teachers.

We surveyed teachers from each school at the end of the school year about management practices in their school (i.e. by their principal). The two primary outcomes were the frequency of formal planning sessions and frequency of classroom observations. Secondary outcomes included whether principals spoke to teachers about their teaching, helped them to set class objectives.

The guidance resulted in an increase of 2.3 percentage points in the proportion of principals who were reported to conduct formal planning sessions every two weeks or more over a baseline of 24.1% (a 10% increase). We also find an increase of 2.0 percentage points in the proportion of principals who observed classrooms every two weeks over a 36.1% baseline (a 6% increase). Teachers in the treatment groups were more likely to be spoken to about teaching, be helped to set class objectives, adapt their teaching to the level of students, provide support to students with difficulties and reported higher job satisfaction.

It was expected that during the course of the year, some principals would change schools or professions. We were unable to capture which principals changed schools, and thus, principals that received the training and the ‘rules of thumb’ and then changed school may affect our results. We would expect this to make the project look less effective than it actually was, if anything. We randomised at the principal level because a number of principals oversee more than one school.

Finally, some principals from treatment groups did not attend the workshop where the poster and implementation checklist were given and explained. For these principals (16.7% of those in the treatment group) the materials were sent directly to the school address of the principal. There is a risk that these principals did not understand the ‘rules of thumb’ as well as those who attended the sessions. These principals were still considered for analysis. Again, this would bias the results downward, if anything

The introductory session to present the content to staff were crucial to mitigate some of the risks mentioned above. Representatives from DIGEDUCA in the Ministry of Education convened principals in groups of about 50 to distribute the above materials and led a session on how to use the materials and why implementing the ‘rules of thumb’ is important, as well as answering principals’ questions.

Making sure the posters and letters followed the design line and logos of the Ministry also help build trust and legitimacy around the intervention. Introductory letter from the Minister of Education: an introductory letter from the ministry explaining how to use the poster and implementation checklist, emphasizing the importance of doing so. The importance of the choice of messenger to inspire a desired set of actions has been well documented. In this case, the letter came from the Minister of Education and was personalised with the principal and school name.

The intervention has not yet been replicated but it has helped strengthen the evidence for interventions that help principals learn and introduce better management practices in their schools

Beyond Guatemala, we are looking for potential partners and opportunities to adapt the content of this intervention to other contexts and rigorously test its replicability.

We find that the ‘rules of thumb’ intervention increased the frequency and quality of school management practices by principals across a broad range of indicators. Teachers who were in schools in the treatment group reported more planning sessions and class observations conducted by their principals. These teachers were also more likely to be spoken to about teaching, be helped to set class objectives, adapt their teaching to the level of students, and provide support to students with difficulties. These changes in turn increased the job satisfaction of teachers. However, the intervention did not translate into an impact on student attainment in maths or reading.

We recommend that the intervention is scaled across the country and that the Ministry test additional higher-intensity interventions for principals on school management practices. Principals themselves would likely benefit from more planning sessions, observation, and feedback from more planning, observation, and feedback.

Sample: There is a total of 4,124 schools in Guatemala that are categorised as Diversificado (High School), with a total of 2,892 principals responsible for one or more of these schools. The principals of 11 top performing schools were excluded from the sample at the request of the Ministry.

This trial’s sample consisted of principals from schools at the Diversificado level, including both urban and rural areas throughout the country. Around 90% of these schools were urban while a 10% were rural. 74% of schools were private schools, 21% public schools, and 5% quasi-public (municipal or cooperative).

Principals in the sample were automatically enrolled into the treatment and control groups. Principals that were assigned to the treatment group received the materials and introductory session from the Ministry.