CACC: Conjunctive Analysis of Case Configurations

Description

This package contains a set of functions to conduct a Conjunctive Analysis of Case Configurations (CACC; Miethe, Hart & Regoeczi, 2008). CACC is an exploratory technique for multivariate analysis of categorical data, posing an alternative to other traditional methods. Although initially conceived as a technique for analysing criminological data, CACC can be applied in a wide variety of contexts, permitting composite profiles of particular units of analysis to be examined.

• “cacc” is the first and foremost function. This function collects a data frame to execute a CACC. As a result, it returns another data frame that contains only the dominant profiles observed in the data, the number of times each profile is observed, as well as the probability the outcome for each profile is observed.
• “importance_variable” is used to calculate the weight of the variables in the result. This function can be useful for the selection of variables to be included in the CACC from a data-driven approach.
• “cacc_xsq” applies a chi-square test on the “cacc” output to determine if there is a significant concentration of cases in the observed dominant profiles, as described in Hart (2019).
• “sci” applies a modified version for configural analysis of the Gini coefficient on the “cacc” output to quantify the relative magnitude of CACC results, as described in Hart (2019).
• “gg_lorenz_curve” plots a visualization of the “sci” output using the Lorenz curve.
• “main_effect” is used to calculate the specific weight of each of the variables included in the CACC on the result, while controlling for the effect of the others. This function is applied on the data frame to identify all pairs of identical profiles except for one variable, as described in Hart and Moneva (2018). This function then calculates the differential probability between each pair of profiles. As a result, the descriptive statistics of each of the variables analysed are presented, together with a boxplot that allows the effect of each variable on the result to be visualised.

References

Hart, T. C. (2019). Identifying Situational Clustering and Quantifying Its Magnitude in Dominant Case Configurations: New Methods for Conjunctive Analysis. Crime & Delinquency, 0011128719866123.

Hart, T. C., & Moneva, A. (2018). Análisis Conjunto de Configuraciones de Caso: una introducción al Pensamiento Configural. Revista Española de Investigación Criminológica, 16, 1-19.

Miethe, T. D., Hart, T. C., & Regoeczi, W. C. (2008). The conjunctive analysis of case configurations: An exploratory method for discrete multivariate analyses of crime data. Journal of Quantitative Criminology, 24(2), 227-241.