Basic Statistics Using R for Crime Analysis

In the previous chapters, I have discussed various data collection methods (e.g., Uniform Crime Report or National Crime Victimization Survey). Police departments and residents in the community can be great sources of data related to crime, but one source of the data we have not covered yet is inmates. Many inmates are in jails or prisons because they are arrested, prosecuted, and convicted for their accused crimes. If we survey inmates, they may provide useful information that can help us understand crime from offenders’ perspectives. This is part of the reason why inmates have been used for various academic articles. For this chapter, I will use the data from an inmate self-reported survey conducted in Korea (Choi & Dulisse, 2021). Specifically, we will first perform a simple linear regression to investigate the relationship between low self-control and risky lifestyles among inmates. Following that, we will conduct a multiple linear regression analysis, considering both low self-control and age as predictors, while evaluating their impact on risky lifestyle, which serves as the dependent variable.

Let’s first load the data. You will download the data from the shared Google Drive folder containing the Inmate Survey.sav data. The next steps should be familiar to you at this stage.

library(haven)
Inmate_Survey <- read_sav("Inmate Survey.sav")
View(Inmate_Survey)

A total of 986 inmates from 20 geographically distinct prisons participated in this survey. Risky lifestyles (RL) were assessed using four items that gauge involvement in unstructured criminogenic activities within the prison environment: (a) possession of prohibited items, (b) breaking away from designated areas, (c) participation in gambling, and (d) involvement in illegal transactions of prohibited products. Participants rated each item on a scale ranging from 0 (never) to 4 (more than 10 times). The scores for these items were summed to obtain a composite measure of risky lifestyles, with higher scores indicating greater involvement. This set of items demonstrates strong internal consistency, prompting students to recall their understanding of reliability testing. Age (AGE) is a continuous variable representing the participants’ age. Low self-control (LSC) was assessed based on six items: “I prefer to do things physically rather than verbally,” “When encountering difficult or complicated tasks, I usually give up,” “I lose my temper easily,” “I enjoy doing things that are a little exciting,” “I often tease others,” and “I prioritize immediate pleasure.” A composite measure of low self-control was created by summing the scores on these six items, with higher scores indicating lower levels of self-control.