OBJECTIVE:To establish an appropriate scoring system using Rasch scores for the strategic, tactical, and operational levels of the Neck Pain Driving Index (NPDI) and to refine the NPDI by testing the unidimensionality of each driving performance level using Rasch analysis. DESIGN:Cross-sectional. SETTING:Tertiary institution. PARTICIPANTS:Individuals with chronic whiplash-associated disorders (WAD) (N=123). INTERVENTIONS:Not applicable. MAIN OUTCOME MEASURE:The NPDI, which was developed to evaluate self-reported driving difficulty in people with chronic WAD. RESULTS:On the basis of Rasch analyses, modifications were made to the response format, changing it from a 0 to 4 scale (0=no difficulty, 1=slight difficulty, 2=moderate difficulty, 3=great difficulty, 4=unable to drive because of the injury) to a 0 to 3 scale (0=no difficulty, 1=slight difficulty, 2=moderate difficulty, 3=great difficulty and unable to drive because of the injury). Unidimensionality of the strategic and operational levels was confirmed. Modification of the tactical level was necessary to ensure its unidimensionality. After removing the tasks "driving near (your) collision site," "driving in a bad weather condition," and "driving at night," unidimensionality of the 4-item tactical level was confirmed. A 9-item NPDI was established. A table converting raw total scores into Rasch scores was created for each level. CONCLUSIONS:This study established the 9-item NPDI, and its 3 subsections (strategic, tactical, and operational levels) are unidimensional. The magnitude of self-reported driving difficulty at each of the 3 subsections can be assessed with the use of a 0 to 3 scale and Rasch scores.