Analyzing patterns and interpreting data for a wide range of audiences is a critical aim for research professionals. Data visualization methods range from data tables, graphics, maps, to fully-interactive web applications. Kyli Gallington, Research Associate for Survey Research Group (SRG), and I (Research Scientist for CRGC) have a strong interest in diversifying data visualization techniques.

For those who do not know Kyli, she joined PHI's SRG as a Research Associate in 2016. She has several years of experience conducting research and evaluation studies for community-based interventions. Ms. Gallington has expertise in both qualitative and quantitative data collection and analysis methodologies. She has experience conducting statistical analyses using SAS and STATA, and is skilled in geospatial analysis and mapping using Esri ArcGIS. In her current work, Ms. Gallington programs mapping applications and conducts geospatial analysis to select target populations and study samples, prioritize community-based intervention locations, and analyze geographic and spatial patterns in data.

Due to our shared interest, we decided to team up and explore a new data visualization technique using cancer registry data. We will be using Esri's ArcGIS platform. ArcGIS has many different methods to provide contextual tools for mapping and spatial reasoning so you can explore data and share location-based insights. When using GIS platforms, you can create a deeper understanding of where things are happening and how different sources of information may be connected. The tool we will be using to present our research findings is called Predominance Mapping. Predominance mapping is a smart mapping tool within ArcGIs that analyzes multiple variables of comparable data and displays predominant values, enabling patterns to easily be viewed and analyzed. For our purpose, we will be using predominance mapping to explore spatio-temporal patterns in cancer registry data. 

Click here for an example of interactive predominance mapping: 2008 Election Voting.  We will be using 2010-2014 incidence and mortality data for the top 5 cancers for both males and females. Incidence rate ratios and mortality rate ratios were calculated by county relative to California rates. Data will be displayed by county and sex. We will also be mapping the location of each NCI-designated cancer center to compare spatio-temporal patterns of cancer in counties with and without NCI-designated cancer centers. Although not all analyses have been conducted yet, we did discover some spatio-temporal patterns that suggest disparities in the predominant cancer site. This suggests more investigation to be done looking at stage at diagnosis, access to care issues, and potentially differences by race/ethnicity.

We are very excited to continue creating this interactive mapping tool to share with the scientific community. Our intent is to create a tool that can be used on the CRGC website for public use. Once it is created, we will write up another article for CHATS to announce its availability.