Using GraphPad Prism’s Heat Maps for Efficient, Fine-Grained Analyses of Single-Case Data

Mitteer et al. (2022) wrote a how-to paper. They show BCBAs how to turn raw single-case numbers into color heat maps inside GraphPad Prism.

The maps can shade each response by error type, time of day, or any other tag you choose. No new data were collected; this is a tool guide.

The authors found that Prism can auto-color cells in seconds. Patterns pop out that line graphs often hide.

For example, a red stripe at 2 p.m. tells you errors spike right after lunch.

Snodgrass et al. (2022) also want clearer single-case pictures. They give six graph styles that keep rate and fidelity visible at the same time. Mitteer adds heat maps as a seventh style.

Blair et al. (2022) teach free Google Sheets graphs. Their steps cost nothing, while Prism needs a license. Use Blair when the budget is tight; use Mitteer when you already own Prism.

Watts et al. (2021) show Excel tricks for multiple-baseline lines. Heat maps do not replace those lines; they supplement them. You can paste the same data into both Excel and Prism and get two views for the price of one.

You can color-code tomorrow’s session file during a five-minute break. The heat map will show if problem behavior clusters around transitions or specific staff. Share the picture with parents and teachers; no stats talk needed. One glance often does the job of a long verbal explanation.

Start by choosing the right data table. In Prism 8 and later, a Grouped table or a Multiple Variables table both support heat maps. Enter your data so that rows and columns represent the two dimensions you want to compare, for example days down the rows and times of day across the columns, with the value in each cell.

With the data entered, create the graph. Go to the Graphs section, make a new graph from that data table, and choose Heat Map as the graph type. Prism then colors each cell automatically based on its value, using either a continuous gradient or categorical colors.

Finish by setting the color scheme and range. In the Format Graph dialog you can pick the low, middle, and high colors, set the numeric range that maps to those colors, and reverse the scale if needed. Locking the range across several graphs lets you compare them fairly.

By default a Prism heat map shows color only. To print the actual number inside each colored cell, double-click the graph to open the Format Graph dialog and turn on the Show values in cells option. Prism then overlays each cell's value on top of its color.

Once values are shown, you can adjust how they look, including the number of decimal places and the font, so the labels stay readable against light and dark cells. Showing values is useful when the exact number matters as much as the pattern, such as in a table you plan to publish.

This article shows behavior analysts how to use Prism heat maps for fine-grained analyses of single-case data. One example is plotting when problem behavior occurs across days and times, which can reveal patterns that point toward the behavior's function. Another is mapping within-session error patterns to expose which variables may be driving faulty stimulus control.

The authors note that these analyses are tedious to do by hand, and that conditional formatting in a spreadsheet does not always produce publication-quality figures. Prism's heat map codes each data point automatically with color, so a spreadsheet of raw values becomes a readable figure in about a minute, ready for clinical decision-making or a manuscript.