Umeed in Punjab: How Play Can Change Emotions

Between December 2025 and February 2026, nine government and community schools in Punjab state ran the first two Umeed modules with their own teachers. This is what happened, and what it is reasonable to conclude from it.

What we were testing

Social and emotional learning is largely absent from Pakistan's public school curriculum. Children are rarely given concrete, teachable ways to recognise and handle difficult feelings like rejection and loneliness.

Umeed teaches those skills through short, story-based activities built on cognitive behavioural therapy, run by the classroom teacher. Where there are devices, it runs as a digital activity. Where there are not, which is most Punjab government schools, the identical content runs as a paper worksheet and a teacher script.

The pilot tested two modules out of ten. Rejection, taught through cognitive restructuring, and Loneliness, taught through behavioural activation.

How it was run, and the limits of that

The District Education Authority assigned the nine schools. We did not choose them. Within each school, sections were split between a group that received the modules and a group that carried on with the normal curriculum, and both groups answered the same twenty-item survey before and after.

That design is a difference-in-differences comparison, which measures the change in one group against the change in the other over the same period, so that anything affecting both groups equally is netted out.

What it showed

Children who did the modules improved 10.62 points more on the survey than children who did not, over roughly five weeks, with a p-value below 0.001. The two groups started from statistically indistinguishable baselines, and the comparison group's scores did not meaningfully move over the same period.

Three things support that number rather than repeating it.

The paper version works. Split by delivery mode, paper-based delivery accounts for most of the effect on its own, with the digital activity adding a smaller additional benefit. For scaling into schools without devices, that is the most useful finding in the report.

Session-by-session data points the same way. Across the pilot, 85 percent of students showed an overall improvement in emotional state. 73 percent could recall the coping strategy immediately after a session, and 91 percent could apply it correctly in a situation different from the one they learned it in.

Parents and teachers described the same change independently. In focus groups, parents reported children naming feelings instead of going quiet, and teachers reported more openness and less conflict. Teachers did not report equivalent changes in the comparison group over the same weeks.

What it does not show

Two modules out of ten, five weeks, one district, one survey, and no measurement after the pilot window closed.

So this is early evidence that the approach moves something, in a setting it is hardest to move anything. It is not evidence about the full programme, it does not tell us whether the effect lasts past a school term, and it does not tell us what happens when the other eight modules are added.

The closest comparable local study is worth knowing about. A randomized trial of the World Health Organization's School Mental Health Manual, also in Lahore, significantly improved teachers' mental health literacy but found no significant change in students at three months. That is the gap a student-facing programme has to clear, and it is why the next step is a properly randomized trial rather than a bigger version of this one.

The full report

It has the regression specification, the tables, the survey construct mapping, and the international literature this sits inside.

If you are evaluating Umeed for a school or a ministry, we would rather have you read the study than a summary of it.