Measuring differences-in-differences of crimes: a case of Dolphin Force in Punjab

Abstract

Learning outcomes The case has the following learning objectives: 1. understand the various types of comparisons that are possible between groups over time and across space; 2. evaluate a policy intervention using relevant data and different methods; and 3. understand the meaning of the phrase “controlling for other relevant factors” in regression and non-regression contexts. Case overview/synopsis The difference-in-differences (DID) approach is a useful tool for making meaningful comparisons. This case tries to provide a non-technical introduction to the approach using a basic comparison of crime rates among districts in Punjab (Pakistan's largest province). Being the most populous region of the country, Punjab faces many governance challenges, and street crime is one of them. (Exhibit 5 provides additional information about the geographical and administrative setting used in this case study.) In 2016, Chief Minister Shahbaz Sharif established the Dolphin (police) Force in different locations to improve urban patrolling and reduce street crime. There were debates about the effectiveness of the Dolphin Force (DF). Those who are skeptical of DF point to various situations that were handled incorrectly by DF personnel, as well as other administrative and operational problems in the initiative. Optimists believe it is beneficial and want it to be expanded to other districts and regions. The threat of street crime claims many lives and, according to optimists, necessitates the formation of a special force. Whether the huge resources invested in the DF worth their lot or not can be known through sound statistical analysis that can identify the difference in the rate of crime because of the DF. In this instance, the case provides information to answer the following question: Is there a significant difference in crime rates between areas where the DF is operating and districts where it has not yet been installed? Complexity academic level In quantitative/statistical analysis classes, the case can be used to teach the DID technique to MBA/MS Applied Statistics/Applied Data Analysis students. It can also be used in undergraduate Econometrics classes. Supplementary materials Teaching notes are available for educators only. Subject code CSS 7: Management Science.