The lrd package provides a set of tools for quickly and accurately processing large amounts of lexical response data that are typically generated from cued-recall tests, while also being able to control for minor errors in participant responses. To simplify use, we have created an R Shiny application that can be used to run this package without R. A general user guide for this application is provided below.
Loading a Dataset
The input data will need to be saved as a .csv file. This file needs to be arranged in long format (i.e., each participant observation constitutes one row in the dataset) and must contain at least three columns arranged in the following order: A unique participant identifier, participant responses, and a scoring key. This file can also contain other columns (e.g., those denoting experimental conditions, demographics, etc.), however, these must be placed after the third column (Click here to download a sample .csv demonstrating the correct file format).
Scoring the Data
To begin the scoring process, select the appropriate settings (e.g., type of separator). Next, you will need to select the cutoff percentage used for scoring. lrd works by computing the percentage of characters that match between two strings (i.e., the percent match). Only string pairs with a percent match above a certain user specified threshold are scored as correct matches. The slider can be used to adjust this threshold (ranges from a 50% match to a 100% match). The example below is set to load a .csv file and score it using a 75% match criteria.
The data is now ready to be uploaded. Click browse and select your file to begin the upload process. The data will then be automatically scored based on the selected cutoff percentage. A preview of the output file showing both the percent match and score (1 = match, 0 = no match) will then be displayed in the "Scored Output" tab. The scored data can be saved as a .csv file using the download button at the top of this tab.
Each participant's mean proportion of correct responses and corresponding z-score can be viewed using the "Proportion Correct" tab. This output can be customized based on any of the optional condition columns that are attached to the upload .csv file (data can be split on up to two conditions at a time). These values can be downloaded using the download button at the top of the tab. Please note that z-scores are only generated when splitting the data on one condition.
The "Plots" tab can be used to visualize the dataset. Plots can be customized based on the optional condition columns in the dataset. If no condition columns are included, this tab will display the distribution of participant responses (This can also be viewed by selecting "id" as the grouping condition.