The reporting module
Last updated
Last updated
When BITTSy runs an experiment, it generates a detailed log file of all events and their timing. Afterwards, these log files can be used by a separate reporting program to generate summaries of what happened in a single study session, or in a set of study sessions (i.e. logs from all participants in a study).
Reports are output into CSV files, which can be imported into other programs for statistical analysis.
The BITTSy reporting module is open-source. Get the source code here, or follow instructions below to download as an executable.
Download the reporting module from the Google Drive folder. Like the main BITTSy program, it does not require any installation. Simply unzip the folder, navigate into the subfolders, and locate the .exe file. (Like when setting up BITTSy, you may wish to create a shortcut to move to the desktop, and may need to take some additional steps to allow Windows to run the program the first time you try to launch it.)
The user interface of the reporting program looks like this.
The reporting module can be used to generate a summary of events or looking behavior on a single study session by examining one log file document, or to average across study session logs by examining an entire folder. With the folder option, individual summaries are also available below the averages section. Note that on some report types, averaging across participants does not make sense; on these, analyzing a whole folder is simply a faster way of obtaining all of the individual summaries.
To analyze a log file individually, select Document and click the Load Document button. Navigate through the file browser pop-up to pick the desired log file.
To analyze a whole folder of log files, select Folder and click the button underneath (which will switch to saying Load Folder) to select the desired folder. This is especially useful for analyzing a whole study, and will generate averages across participants on measures of interest in addition to summaries of individual sessions. Note that prior to analyzing the folder, you should ensure that it contains:
all the log files you wish to analyze,
which were all from a single study,
and no other text files, including the protocol file or logs that were generated during test runs of the protocol.
All reports that are generated by the reporting module begin with specifying which files were included in the report. When looking at data files that result from running a folder, it is always easy to see whether the folder contained all the intended participants, and whose data is included in the reported averages.
In the right-hand column, there is a list of available report types. See this page on standard reports for descriptions of these, to help you determine which reports are appropriate and useful for your study.
You can check off multiple report types at once to get different types of information about your log file/folder of log files.
Sometimes certain stimulus types are presented in an experiment, but aren't important to look at in a report. These are often visual stimuli that serve as a fixation point while separate audio files are played - flashing lights during a HPP study, or a checkerboard pattern on the screen during an auditory-focused central fixation or habituation study. In these cases, we will end up analyzing attention toward to audio files, and attention toward the visual fixation stimulus is completely redundant. Therefore, reports that analyze looking time towards particular stimuli allow you to exclude certain stimulus types from consideration. This helps remove redundant information and sometimes makes it easier to work with the resulting CSV files.
When you check off a report type that allows you to restrict stimulus types, you will see checkboxes appear in the stimulus types section. Any stimulus type that is not present at all in your protocol will obviously not be present in the reports - so there is no need to uncheck anything that doesn't occur. But if there is a stimulus type present that isn't important to include in reports, you can uncheck that box so that it will not appear.
Whenever restriction of reported stimulus types is available for some selected reports and not others, a note will appear underneath explaining which reports your stimulus type selections will affect.
Analyzing a single log with single or multiple report types, or a folder of logs with a single report type will be output into a single CSV file. When you click the Save Location button, you can specify where this file should be saved and what to name it.
When analyzing a folder with multiple report types, the different reports will be saved as separate CSV files, as this is generally more readable. In this case, the following window will appear when you click the Save Location button. Instead of naming a single CSV file, you will choose a folder to save into, and make a "filename stub". The CSV files for each report will be named with the stub and the name of the report.
Once your source document/folder, report types, and save location are specified, you can click Generate Reports, and the CSV file(s) for your data reports will be created.
Additional reports can be generated immediately afterwards. A save location/file name must be specified every time, but other selections persist until they are changed.
The reporting module was designed for use with logs from BITTSy version 1.12 and later, and will not parse earlier logs correctly. If you were an early beta user and have older logs to analyze, contact us for a simple script to pull out trial and looking information. Please include the version number of BITTSy used to run your set of logs.
