/ Home


Car Preferences

Keywords: survey, contingency table, ordinal response


These data were collected as part of a project for the Federal Office for Road Safety conducted by the Research Institute of Gender and Health at the University of Newcastle. There is evidence that women drivers who are involved in motor vehicle accidents are more likely than men to be injured. A possible reason is that women often drive smaller cars that provide less protection in a collision. One of the aims of the project was to examine preferences for cars among men and women and investigate the extent to which safety was a factor in determining preferences.

The survey was conducted by research assistants who asked people in car parks to participate and administered a structured questionnaire. They were instructed to obtain data from men and women with small, medium and large cars, with 50 people per group for a total of 300 respondents. (The sample size was based on power requirements for another part of the survey that involved anthropometric measurements.) The research assistants approached people in car parks of the University of Newcastle and nearby shopping centres during December 1997 and January 1998. 

The data consist of 300 records each with 22 variables. The variables are:

Variable Description

ID Identification number of respondent
AgeAge of respondent (years)
Sex 1=female, 2=male
LicYrTime they have held a full driving licence, in years and months (years)
LicMthTime they have held a full driving licence, in years and months (months)
ActCarMake, model and year of car most often driven, coded to size of car 1=small, 2=medium, 3=large
Kids5Children under five, 1=yes, 2=no
Kids6Children 6 to 16, 1=yes, 2=no
PrefCarPreferred car, coded to size of car 1=small, 2=medium, 3=large
Car15k Preferred type of car if cost $15000, 1=small new car; 2=large second-hand car
Reason 1=safety, 2=reliability, 3=cost, 4=performance, 5=comfort, 6=looks
CostHow important is cost when buying a car? 1=not important, 2=little importance, 3=important, 4=very important
ReliableHow important is reliability ...?
PerformHow important is performance ...?
FuelHow important is fuel consumption ...?
SafetyHow important is safety ...?
AC/PSHow important is air conditioning/power steering ...?
ParkHow important is ease of parking ...?
 Room  How important is space/roominess ...?
 Doors  How important is the number of doors ...?
 Prestige  How important is prestige/style ...?
 Colour  How important is colour ...?

For further details, see the wording of the questionnaire.


Data file (tab-delimited text)


The data was contributed to OzDASL by Professor Annette Dobson, University of Queensland. Information on the data set was originally provided by Jenny Powers.



Home - About Us - Contact Us
Copyright © Gordon Smyth