2. What explanatory factors are associated with the varying rates of participation, violence, and saving of other individuals (the Schindler-syndrome)?

Daily Victim Totals

Estimation

Victim totals were estimated with a negative binomial regression. This basically lets the researcher discover which variables were related to daily victim totals and how much movement in each variable was expected to increase or decrease daily victim totals. We only report the most robust results below.

Findings

All else equal, the commune with the lowest population had 1700 fewer deaths on average than the most populated commune. This trend was reinforced by the prefecture population which at its highest produced, on average, 334 more daily deaths than the prefecture with the lowest population. In other words, more populous prefectures tended to have greater numbers of victims.

The commune with the greatest surface area had on average 29 fewer deaths per day than did the commune with the lowest surface area. Thus, if there was a larger surface area that had to be searched, individuals were more likely to survive.

One of the most important variables in this analysis was the proportion of the prefecture that was Tutsi. Communes with the highest percentage of Tutsi inhabitants, on average, had 18,000 more daily deaths than did communes with a relatively lower percentage of Tutsi.

Last, but certainly not least, events that took place in a religious space had on average 62,000 more victims than events taking place elsewhere. All of these findings can be seen in the first and second columns of Table 1.