Probability of suffering from cancer among men born in Madhya Pradesh
| Prob (reports having cancer) | |||
| (1) Pooled | (2) Non-movers | (3) Movers | |
| 1986–1990 | −0.00373 | 0.00170 | −0.0141 |
| (0.00340) | (0.00147) | (0.0121) | |
| 1985 | 0.0209*** | 0.0624*** | −0.00906 |
| (0.00591) | (0.00505) | (0.0113) | |
| 1975–1984 | −0.00459 | −0.00157 | −0.00906 |
| (0.00355) | (0.00181) | (0.0113) | |
| 1960–1974 | Reference | Reference | Reference |
| Observations | 7031 | 5002 | 2029 |
| R2 | 0.007 | 0.012 | 0.014 |
| Cohort FEs | Yes | Yes | Yes |
| Control mean | 0.00252 | 0.00228 | 0.00321 |
The cohort of men born in 1985 within 100 km of Bhopal has a higher likelihood of cancer compared with other cohorts and to those born in Madhya Pradesh beyond 100 km from Bhopal (1), with effect most pronounced among men who have never moved (2) instead of those who have moved (3). Estimates are from ordinary least squares regressions as in equation (1) with controls consisting of FEs for year of birth cohort and whether the DHS cluster is within 100 km of Bhopal. SEs are two-way clustered at the district and year of birth levels.
Robust SEs in parentheses.
***P<0.01, **p<0.05, *p<0.1.
DHS, Demographic and Health Surveys; FEs, fixed effects.