Although KidStats can run with R, we recommend using RStudio. RStudio is an integrated development environment for R that works with Shiny, includes Pandoc (a program used to generate reports), and features used to develop the software. RStudio or R needs to be open for KidStats to operate. Thus, if you close R or RStudio the web interface for KidStats will be disabled.
Upon evaluation, some output is provided directly beneath the measurement input boxes. Specifically, the point estimate, upper and lower 95% prediction intervals, the R-squared value for the model, and the sample size used in the evaluation are provided. A more detailed output is available from the Output page under the Age Estimation tab.
Some output (male and female posterior probabilities and sample size) is provided directly beneath the measurement input boxes and age estimation results. A more detailed output is available on the Output page under the Sex Estimation tab and includes the posterior probabilities when the unknown is used in the created model, the FDA coefficients, variable importance, a confusion matrix that illustrates the trends in classification/misclassification for the created model, and the bootstrapped classification accuracy.
Comment: It is important, and almost imperative for the sex estimation analyses, to evaluate which variables were incorporated into the model in the extended reports. If variables were not included and have no variable importance (provided by a small evimp number [i.e., 0]) then they can be 'excluded' so they are removed from the next iteration of model creation. The removal of superfluous variables can increase the sample size, which will yield a more realistic prediction interval/classification accuracy. However, remember more variables incorporated into the sex estimation models generally yields higher classification accuracies. The overly optimistic effects of a small sample size can be mitigated by bootstrapping.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.