Home

Menu

Loading wiki pages...

View
Wiki Version:
<p>This components contains the analysis scripts and all needed files for to recreate the results reported in Kellen and Singmann "ROC Residuals in Signal-Detection Models of Recognition Memory".</p> <p>The following gives a description of the files:</p> <ul> <li> <p><code>analysis-residuals.R</code>: R script file that performs the analysis reported in the manuscript (this is the main file). Produces the values reported in the text as well as Figures 3 to 6. Needs the following files to run succesfully:</p> <ul> <li><code>6-point-ROCs.rda</code> and <code>8-point-ROCs.rda</code>: binary R data files containing the ROC data (the data in txt format is available in the <code>data (txt)</code> component which also contains the references to all data used)</li> <li><code>uvsdt.6point.model</code>, <code>dpsdt.6point.model</code>, <code>msdt.6point.model</code>, <code>uvsdt.8point.model</code>, <code>dpsdt.8point.model</code>, <code>msdt.8point.model</code>: <code>MPTinR</code> model files containing the model equations.</li> </ul> <p>The analysis script tries to read the following binary R files containing model fits. If these files are not present, they will be created by performing the fits (SDT models and LMMs, this might take quite some time). Delete those files to redo the fitting: - <code>fits.rda</code>: fits to original data for UVSD, DPSD, MSD0 and MSD. - <code>lmms.rda</code>: LMMs of residuals (fitted to the original data). - <code>fits_pred.rda</code>: fits to model predictions (i.e., "Residual Analysis of Model-Generated Data"). - <code>lmms_predicted.rda</code>: LMMs of residual from models fitted to other models' prediction.</p> </li> <li> <p><code>6_resid.pdf</code>, <code>8_resid.pdf</code>, <code>fit_predicted.pdf</code>, and <code>fit_predicted_8.pdf</code>: Figures 3 to 6 in the manuscript (are produced by the analysis script)</p> </li> </ul> <p>Note that when rerunning the analysis using the provided scripts and new model fits, the actual values obtained can vary from the values reported in the text. The reason for this is the numerical error brought about by using numerical estimation in the fitting of the models. The differences are however not critical to the conclusions drawn from the analysis.</p>
OSF does not support the use of Internet Explorer. For optimal performance, please switch to another browser.
Accept
This website relies on cookies to help provide a better user experience. By clicking Accept or continuing to use the site, you agree. For more information, see our Privacy Policy and information on cookie use.
Accept
×

Start managing your projects on the OSF today.

Free and easy to use, the Open Science Framework supports the entire research lifecycle: planning, execution, reporting, archiving, and discovery.