There are several ways to obtain results for statistical significance on the aytm platform: right inside your charts and visualizations, with our Correlation Finder, and by exporting from the Banner Tables menu.
In This Article
- Real-time Significance Testing
- Correlation Finder
- Banner Tables
- Tips and Tricks for Analysis
- Related Articles
Real-time Significance Testing
When you hover over any bar in a chart, the confidence interval band and hatching update in real time to use that bar as the reference point. The green shaded band shows the confidence interval for the hovered answer.
- Bars that fall within the band are hatched—they're not statistically significantly different from the hovered answer.
- Bars that fall outside the band are solid—the difference is statistically significant.
⚠️ Note: In-chart significance testing uses a 95% confidence level by default—you can change this in View Options—and is always calculated relative to the bar you're hovering over. Hover a different bar to change the reference point and the results will recalculate.
Correlation Finder
To take your analysis a step further, or to run correlation scans (banner tables) on a single banner, try Correlation Finder from the Analysis menu. Correlation Finder performs an analysis on the banner and attribute of your choice right in your results, including a chart.
Learn how to use Correlation Finder.
Banner Tables
Banner Tables let you detect significant differences in your data by downloading data files to work with offline. Select outputs with counts and/or percentages, Top/Bottom Boxes, and Means for the banner groups you specify.
Learn more about Banner Tables.
Tips and Tricks for Analysis
Virtual Questions are a great way to roll up, organize, and more precisely slice and dice your data for use as banners. The resources below show you how to do that with real-world examples.
💡 Tip: Build your Virtual Questions before you export. Cleaning up and grouping your banners up front means your Banner Tables come out ready to read, with no post-processing in Excel.