Tables and graphs for monitoring temporal crime trends

Abstract

This article is a practical review on how to construct tables and graphs to monitor temporal crime trends. Such advice is mostly applicable to crime analysts to improve the readability of their products, but is also useful to general consumers of crime statistics in trying to identify crime trends in reported data. First, the use of percent change to identify significant changes in crime trends is critiqued, and an alternative metric based on the Poisson distribution is provided. Second, visualization principles for constructing tables are provided, and a practical example of remaking a poor table using these guidelines is shown. Finally, the utility of using time series charts to easily identify short-and long-term increases, as well as outliers in seasonal data using examples with actual crime data is illustrated.

Is the recent increase in national homicide abnormal? Testing the application of fan charts in monitoring national homicide trends over time

“…There are three disadvantages that come from using the percent change method. First, the variance of the statistic is not defined (Wheeler, 2016). This makes it difficult to identify actual outlying values (Wheeler & Kovandzic, 2018).…”

“…Second, low baseline values tend to result in high variance. Since the numerator in the percent change is the historical crime rate, crimes with lower rates will more easily have larger percent changes (Wheeler, 2016). Rotolo and Tittle (2006) argue against using percent change in their examination of the effect of city population on crime rates, especially for measuring homicide, and opt for using absolute changes to avoid the distortion accompanying modest absolute increases in cities with low crime rates.…”

“…Recent studies offered an alternative approach for monitoring crime trends; the use of forecasts and fan charts (Wheeler, 2016;Wheeler & Kovandzic, 2018). The procedure to identify outlying patterns in crime trends can be described as follows: (a) specify the statistical model reflecting the series of crime data, (b) use a forecasting method to estimate prediction intervals around crime during the period of time of interest, (c) determine whether observations of interest fall inside or outside the forecasted prediction intervals.…”

“…The procedure to identify outlying patterns in crime trends can be described as follows: (a) specify the statistical model reflecting the series of crime data, (b) use a forecasting method to estimate prediction intervals around crime during the period of time of interest, (c) determine whether observations of interest fall inside or outside the forecasted prediction intervals. Some researchers have employed this method and showed its utility in identifying abnormality in recent changes in crime patterns (Wheeler, 2016;Wheeler & Kovandzic, 2018). These studies focused on temporal crime patterns at the city level, with little attention paid to national crime patterns.…”

San Francisco Crime Visualization

“…Wheeler (2016) evaluated visualization methods for the analysis of crime trends in New York [6]. The author argues that the non-Poisson distribution of crime occurrences leads to difficulties when attempting to visualize the data using tables with metrics such as percent change.…”