Introduction: Interventions on behalf of the marginalized in society can assume many formats. In an urban slum the Government of Delhi built one-room houses for some of the residents in what is termed a ‘plot area’. Not all residents could be accommodated in the project and the remainder continued to live next door in shanty houses of the slum. Nineteen years later, young children who had migrated with their parents, have grown up and have children of their own. We looked at the development of the children living in the two types of accommodation.

Methods: A total of 373 children were studied. All children (n = 200) between the ages of 3.5 and 5.5 years in a cluster of five residential blocks in the plot area were studied. As a control, children in two large clusters of shanty houses (n = 173) were also studied. For development assessment the Central Institute of Education (CIE) Test was performed. This is an Indian adaptation of the Standford–Binet Test. Multiple regression analysis was utilized to determine the factors that influenced IQ most.

Results: The mean IQ of the children in the plot area was 92.5 (s.d. 13.38) and in the shanty houses 89.5 (s.d. 12.9) (p = 0.05). Analysis showed that the most significant factors affecting IQ were malnutrition in the first 6 months of life and attendance of the child at pre-school. For nutrition in the first 6 months, there was no difference between the groups. For attendance at pre-school, 110 of 200 in the plot area and 47 of 173 in the shanty houses were attending pre-school (p < 0.01).

Conclusion: We find that children living in the permanent houses had a significantly better IQ than those in shanty houses. A review of the literature did not reveal a comparable study.