Is Regional Income Linked to CT Safe Connect Calls?

by Ryan Wiley

Last updated on January 19, 2022

for Data Visualization for All
with Prof. Jack Dougherty
Trinity College, Hartford CT, USA


Tables, maps, and charts play an important role in interpreting the data to tell a story. Our class collaborated with CTSafeConnect with the intention to create meaningful visualizations of the data and highlight important issues about intimate partner violence in Connecticut. CTSafeConnect is a project by the Connecticut Coalition Against Domestic Violence (CCADV) where survivors of domestic abuse have access to emotional support, safety planning and options

I thought it would be interesting to compare the number of individual callers to the hotline to the income of the area with the hope to highlight driving factors of domestic abuse. Is income a factor in domestic abuse? In other words, how does the income of a region correlate with calls about domestic violence in that region?

Key Findings

The map for the percentage of callers in the region is shaded based on the percentage of the population that made calls to the CTSafeConnect hotline. The darker regions represent areas with a higher percentage and areas with lighter colors show a lower percentage.

Figure 1: callers in a region to CTSafeConnect as a percentage of the total population of the region

Median household Income is a good variable to measure how much money a region makes in a year. The Stamford region has the highest median household income of over $172,000. The regions with the highest median household income are shaded darker and the lowest regions in the lighter colors. The areas with the highest median household income are concentrated in the southwestern area of Connecticut And the lowest being Kingly and Mansfield regions in the northeastern part.

Figure 2: Median household income measured in 2019

When we compare the two variables using a scatter plot, we see a moderate negative relationship between income and domestic abuse. The correlation coefficient of these two datasets is -0.50. A -1 correlation coefficient is a perfect inverse correlation and a coefficient of 0 is no correlation. The line is the line of best fit and it is sloping downward. As the median household income of a region increases, it correlates to a lesser amount of callers in the region. The reason I am reluctant to say there is a correlation is because the points are all pretty far from the line of best fit.. Stamford, for example, is by far the wealthiest region with a median household income of over $172,000, but it is not the region with the lowest proportion of callers to the hotline. Danbury, which has the second highest median household income, has the lowest percentage of callers in the region.

Data Sources and Methods

I needed two variables to compare domestic violence to wealth in Connecticut. I initially wanted to examine the poverty rates in each region to see whether an increase of the poverty rate led to an increase of domestic abuse. Unfortunately, the data about poverty rates from the American Community Survey (ACS) are lacking and could not be used because of too little data in smaller towns. I decided to use median household income data from 2019 as the variable because it was the most complete and useful for determining how wealthy a town is. I decided to use the percentage of first instance callers to the hotline to measure the domestic violence in a region. This percentage represents the callers in the region. For instance, the percentage for the Ansonia region is 0.42%; this means that 0.42% of the Ansonia population made a call to CTSafeConnect Hotline. The data I used was from the American Community Survey and the Masked CCADV SafeConnect Data.


Like I said in the previous parts, more research must be done on the relationship between income and intimate partner violence to determine a relationship. Household income can be used to weakly correlate the number of Callers in a region. The wealth of a region might be a better predictor of domestic abuse than income because wealth takes into account the total number of assets a household has.

REGION Median Household Income (2019) First Instance Callers of the Total Population
Ansonia region 85,861.00 0.42%
Bridgeport region 110,630.33 0.47%
Danbury region 113,520.91 0.26%
Enfield region 87,295.83 0.42%
Hartford region 95,897.17 0.50%
Killingly region 79,997.73 0.48%
Mansfield region 76,769.56 0.45%
Meriden region 86,727.33 0.59%
Middletown region 92,625.67 0.46%
New Britain region 84,455.29 0.58%
New Haven region 93,231.67 0.57%
New London region 83,829.10 0.38%
Stamford region 172,056.00 0.32%
Torrington region 87,562.05 0.41%
Waterbury region 88,110.56 0.57%


The Median household income data is from the American Community Survey and the data about domestic abuse in Connecticut was provided by CTSafeConnect