Assessing the reliability and validity of the Driving Practices Questionnaire in a sample of Hispanic south Florida drivers

Date of Award




Degree Name

Doctor of Philosophy (Ph.D.)



First Committee Member

Georgie C. Labadie - Committee Chair


The purpose of the study was to measure the reliability and validity of the Driving Practices Questionnaire (DPQ) and to test additional investigator developed driving behaviors in the Driving Practices Questionnaire-Test (DPQ-T). The DPQ (Kidd & Huddleston, 1994) has not been tested in Hispanics. Triangulated design using survey and content analysis was used. An urban 412-bed community medical center setting located in Miami-Dade County, Florida was selected and a sample of 1,333 English speaking Hispanics participants were targeted.Four instruments were used the: (a) DPQ (Kidd & Huddleston, 1994); (b) DPQ-T; (c) Driver Sociodemographic Questionnaire (DSQ); and the Heritage Assessment Tool (HAT) (Spector, 1996). The DPQ-T and DSQ were investigator developed. Surveys were distributed to 1,333 hospital employees and physicians to obtain the study sample. Crude response rate was 29.1% (389) and net response rate was 57.4%. Inclusion criteria in the analysis was ≥ 34 (80.9%) complete responses of 42 driving items. A Hispanic driver sample (N = 193) was used for analysis.Reliability and principal component factor analysis were run on 42 driving items on the DPQ/DPQ-T. Original ten-item and 15-item DPQ factor structure was replicated which represented the theoretical dimension of "at-risk" driving behaviors. The 26-item combined DPQ-T was the most parsimonious of all two factor scales reflecting the theoretical dimensions of "at-risk" driving in Factor I (eigenvalue = 9.2) accounting for 32.9% of the variance and the six items in Factor II (eigenvalue = 3.1) representing "seat belt use" accounted for 14.5% of the variance. The overall variance accounted for by both factors was 47.4%. Five scales were created. Four reflected "at-risk driving" behaviors and one reflected "seat belt use." Scale alpha reliabilities ranged from .7490--.9001. The DPQ 10-item scale was the most parsimonious with an alpha reliability of .9001. Cultural assessment revealed respondents had a low heritage affiliation and 50% or more lived in the U.S. for 30 years.Content analysis of US DOT, National Highway Traffic Safety Administration U.S. Hispanic focus group interviews demonstrated item reliability and validity. Of the 42 DPQ/DPQ-T items 35 (83%) were mentioned at least once by NHTSA participants. Four items not mentioned by the NHTSA participants had inter-item correlations ≥ .50 and factor loadings ≥ .57.


Health Sciences, Nursing; Health Sciences, Public Health; Sociology, Ethnic and Racial Studies; Psychology, Psychometrics

Link to Full Text


Link to Full Text