This wiki serves as an overview of the files and scoring procedures for some of the NIH Baby Toolbox measures.

• For scoring individual files, please only change information in the “score_data.R” file.
• For scoring all files in a folder, use “score_data_batch.R”.

Scores are available for the following measures:

• Sum looking = sum of looking to the center of the screen during Habituation/Familiarization portion
• Peak looking = maximum duration of center looking across items in seconds during Habituation/Familiarization portion
• Mean looking = the average looking duration to the center of the screen for that set of items in seconds during Habituation/Familiarization portion
• Scores are available for Round 1, Round 2, and Both Rounds (if the child did not look away during the entirety of Rounds 1 or 2 - otherwise, no Both Rounds score is provided).

• For English, CSS = Change Sensitive Score based on Item Response Theory (IRT) score, which you can use this as an index of change within the individual (e.g., test-retest) or comparing with your sample
• For English, note that there are 10 experimental items. At least 6 responses for any of the non-experimental items must be completed for the individual to receive a score.
• For Spanish, percentage correct = (# of items correct/# items completed) * 100

• CSS based on IRT
• Scores are available for MR in English and Spanish

• CSS based on IRT
• Scores are available for ME in English and Spanish

• CSS based on IRT
• Scores are available for ME in English and Spanish

• CSS based on IRT
• Scores are available for MTL in English and Spanish

• CSS based on IRT
• Scores are available for MTT in English and Spanish

• CSS based on IRT
• Scores are available for VDR in English and Spanish

• CSS based on IRT
• Scores are available for Subitizing in English and Spanish

• CSS based on IRT
• Scores are available for VA in English and Spanish

• CSS based on IRT
• Scores are available for WHM in English and Spanish

• CSS based on IRT
• Scores are available for Counting in English and Spanish

• Mullen Receptive (1-42 months)
• Mullen Expressive (1-42 months)
• Language composite score (1-42 months), provided that MR and ME are completed
• [IN PROGRESS] EF/Memory composite score (22-42 months)
• [IN PROGRESS] Math composite score (25-42 months)
• [IN PROGRESS] Cognition composite score (25-42 months)

There are three types of files most relevant for scoring: the narrow item exports (CSV), narrow registration exports (CSV), and gaze based data (JSON)

• The narrow item exports are necessary to transform item-level data to measure-level scores
• Narrow registration exports are necessary to obtain the participant's age for scoring measures that have different scoring rules by age (e.g., LWL English) as well as to calculate the norm-referenced age adjusted score
• JSON data is necessary to calculate the EF scores based on the Familiarization portion of this task