Picture shows Shiny app where a user has selected a cut-off value for person-total correlation do divide a distribution into cases that are flagged as possible random responders, or spared. Classification metrics (accuracy, sensitivity, specificity) at the given cut-off are shown after the user clicks "Classify!"

If you do survey research with Likert-type items and worry about random responders, Michael John Ilagan & I made a Shiny app we think some of you may like.
falkcarl.shinyapps.io/BotApp1/

#SurveyResearch #DataQuality #Psychometrics #OnlineSurveys #SurveyBots #RandomResponders #rstats #Shiny

So, what are you all using? Michael and I think we have some solutions to random responders (a post for another day), but LLM-enabled bots or responders pose a different challenge.

#SurveyResearch #DataQuality #Psychometrics #OnlineSurveys #SurveyBots #RandomResponders