compile_trial_by_trial_results: Compile trial-by-trial results

View source: R/compile-trial-by-trial-results.R

compile_trial_by_trial_resultsR Documentation

Compile trial-by-trial results


Compiles a data frame of trial-by-trial results for each participant. This function is intended to be called once data have been collected for several participants; it allows the researcher to inspect MDT results on a trial-by-trial level.


  in_dir = "output/results",
  label = "MDT",
  combine = TRUE



Results directory to process.


Label that the MDT's results were saved under (typically "MDT").


Whether to combine results into one big data frame, or instead to return a list of data frames, one for each participant.


Suppose that a participant completes 5 items in the test. They will have 5 rows in the table, corresponding to their 5 items. You can find the participant's rows by filtering by the "p_id" column. The column "num" indexes these rows: 1 corresponds to the first item they took, 2 corresponds to the second item, and so on. The ability scores are given in the "ability_" columns; they tell you the participant's estimated ability after having answered that particular item. To get the participant's final estimated ability score, look for the highest value of "num", in this case 5. Then look up the ability score within that row.

There are different columns for the different ability estimation methods, make sure you are looking at the right one. The correct one to look at corresponds to the final_ability.estimator parameter in your adaptive test. We most commonly use weighted likelihood, i.e. "WL". This is stored in the column "ability_WL". If you are unsure that you are looking at the right column, you can compare the results to the CSV results that you can download from the psychTestR admin interface, which provide solely the participant's final ability score.


A data frame, or list of data frames, of trial-by-trial results.

pmcharrison/mdt documentation built on June 30, 2024, 1:21 a.m.