splithalfr

splithalfr estimates split-half reliabilities for scoring algorithms of cognitive tasks and questionnaires.

Key Features:

• First-Second Half Split: Splits trials into first and second halves for reliability estimation.
• Odd-Even Trial Split: Splits trials into odd and even numbered trials for reliability estimation.
• Permutated Split: Uses permutation-based splitting of trials for reliability estimation.
• Monte Carlo Split: Implements Monte Carlo-based splitting, recommended as the most robust method against confounding effects of time, task design, trial sampling, and non-linear scoring.
• Stratification by Task Design: Allows each splitting method to be combined with stratification by task design.

Scientific Applications:

• Cognitive Task Reliability Estimation: Estimates reliability of main outcome variables from cognitive task datasets scored with non-linear algorithms.
• Confounding Effect Assessment: Evaluates how time effects, task design effects, trial sampling effects, and non-linear scoring effects influence split-half reliability estimates.

Methodology:

The package implements four splitting methods (first-second half split, odd-even trial split, permutated split, and Monte Carlo-based split), each combinable with stratification by task design, applied to cognitive task datasets typically scored with different non-linear algorithms.