Happy Planet Index. Introductory Statistics. Regression analysis: A quick primer for media on a fundamental form of data crunching Journalist's Resource: Research for Reporting, from Harvard Shorenstein Center. Regression line (MIT) When journalists read academic studies, whatever the subject — the relationship between wages and well-being, the effectiveness of flu vaccines or the benefits of political incumbency — they frequently encounter the term “regression,” a mathematical tool for establishing the relationship between two or more variables. The technique is well known to data journalists, but even savvy reporters may feel a measure of discomfort when they come across it — they seldom have the expertise or time needed to understand advanced mathematics or dig into a study’s original methods and data. Still, there’s a lot to be said for getting more familiar with this somewhat obscure yet ubiquitous form of data analysis.
Simply put, regression analysis is a way to determine if there is or isn’t a correlation between two (or more) variables and how strong any correlation may be. A basic example to use for creating a regression is mortgage rates and housing prices. Gapminder: Unveiling the beauty of statistics for a fact based world view. Free Data Visualization Software. StatWorld - Interactive Maps of Open Data. Spurious Correlations.