Skip to content

jonathan-g/analyzeBehaviorspace

Repository files navigation

analyzeBehaviorspace

License: MIT Lifecycle: experimental R-CMD-check

This is a shiny app to analyze table output from NetLogo BehaviorSpace experiments.

NetLogo outputs .csv files from BehaviorSpace runs that are very annoying to analyze in Excel because of the limited functionality of Excel pivot tables. This application reads in a .csv file from a BehaviorSpace run (use the table option for output, not spreadsheet) and allows the user to summarize the output by up to two variables, and generate plots and summary tables, both of which can be exported.

To use the app, you need R version 3.3 or later.

To install analyzeBehaviorspace, start R and type the following lines:

if (!require(devtools)) install.packages(devtools)
devtools::install_github('jonathan-g/analyzeBehaviorspace')

after you have done that once, you can run analyzeBehaviorspace by starting R and typing the following:

library(analyzeBehaviorspace)
launch_abs()

The app should start running, and open the analyzeBehaviorspace app in your web browser. When the app is running, load a .csv file containing the table output from a BehaviorSpace run. Then you will see a table of values for all the control and output variables at each tick of each BehaviorSpace run. You can generate summary plots and tables by choosing variables for the x and y axes and a grouping variable (if you choose a grouping variable, each different value of that variable will be represented by a different color in the plots).

The graph will plot the mean value of the y variable for each value of the x variable (grouped by each value of the grouping variable if one is chosen). You can select whether to show points, lines, or both, and whether to just show the mean values of the y variable or to show error bars as well (error bars show plus and minus one standard deviation of the y variable).

If you check the Summary table box, the table will show the mean value and standard deviation of the y variable for each combination of the x and grouping variables.

You can download the plots (as .png images) and the summary table (as .csv files).