Intended for healthcare professionals


Determinants of car travel on daily journeys to school: cross sectional survey of primary school children

BMJ 1998; 316 doi: (Published 09 May 1998) Cite this as: BMJ 1998;316:1426
  1. Carolyn DiGuiseppi, senior research fellow (C.DiGuiseppi{at},
  2. Ian Roberts, director,
  3. Leah Li, medical statistician,
  4. Diane Allen, data manager
  1. Child Health Monitoring Unit, Department of Epidemiology and Public Health, Institute of Child Health, University College London Medical School, London WC1N 1EH
  1. Correspondence to: Dr DiGuiseppi
  • Accepted 5 February 1998

The annual distance walked by children has fallen 28% since 1972, partly because car travel has replaced walking on many school journeys.1 Increasing car use has been linked with obesity, adverse health effects in later life, limitations on children's independence, traffic congestion, and pollution. 2 3 To inform the development of strategies to reduce school related car travel, we surveyed the travel patterns of urban primary school children.

Methods and results

The survey was conducted in the inner London boroughs of Camden and Islington. The questionnaire—based partly on published surveys 4 5 and prepared in English, Bengali, Turkish, Greek, and Cantonese (first languages of 85% of eligible pupils)—asked about that day's school journey, children's independent travel, and parental concerns. From the sampling frame of all primary schools (excluding pilot, boarding, and special schools), 31 of the 100 eligible schools were randomly selected. We weighted sampling probability by combined class sizes in year 2 (ages 6-7 years) and year 5 (ages 9-10). Questionnaires, with a letter from the head teacher and a multilingual request form for translation, were distributed to pupils for completion at home. Questionnaires were left for absentees. One week later, we collected completed questionnaires, gave new questionnaires to non-respondents, and distributed requested translations. All pupils were given pencil cases.

We used logistic regression, including a random effect (school) to account for cluster sampling, to estimate odds ratios and 95% confidence intervals for determinants of car travel versus walking. We excluded pupils who used public transport.

Thirty schools (97%) agreed to participate. Of 2476 enrolled children, 2086 (84%) returned usable questionnaires: 96% English, 2% Bengali, 1% Turkish, and 1% English and Bengali (duplicate versions returned). Response rates were highest in independent schools (96%) and lowest in local authority schools (81%). Excluding independent schools, for which the information was unavailable, the respondents' ethnic distribution (54% white, 18% black, 14% Asian, and 15% other) was similar to that of the school population (50% white, 18% black, 15% Asian, 17% other).

Determinants of car travel versus walking on daily journeys to and from primary school

View this table:

Most children walked (69%) or travelled by car (26%). Four (0.2%) cycled, and the rest travelled by bus, underground, or train (5%). Proportions were similar for the journey home. Adults accompanied 84% of children to and from school. Most children (61%) were rarely or never allowed out without an adult for school or leisure. Only 3% of bicycle owners were allowed to cycle on main roads. Ninety per cent of parents were very or quite worried about abduction or molestation, and 89% were very or quite worried about traffic. The strongest predictors of car travel to school were car ownership, greater distance to school, attendance at an independent school, and parental worry about abduction (table). For the journey home, the strongest predictors were greater distance to school, car ownership, and attending an independent school.


Distance to school and car ownership were principal determinants of car travel. After adjustment for these factors, children at independent schools were still more likely to travel by car. Parental fear about “stranger danger” also influenced the decision to drive children to school.

Although few translated questionnaires were requested, the study population adequately represented the ethnic distribution of children attending school in the two boroughs. Our results might appropriately be generalisable to other urban primary school populations.

Increasing emphasis on school choice has been accompanied by a 20% increase in average distance travelled to school.1 Policies that encourage children to attend nearby schools are likely to reduce car travel and increase walking. Parents who currently drive their children might forgo the car for safe, convenient alternatives that address their fears. Unless such alternatives are developed, parents who do not currently drive to school are likely to do so when the option becomes available.


This study originated from joint work with the Camden and Islington Accident Prevention Alliances.

Contributors: CD designed the protocol and questionnaire, implemented the study, and participated in the study design, data analysis and interpretation, and writing the paper. IR initiated the research and participated in study and questionnaire design, data analysis and interpretation, and writing the paper. LL performed data analysis and edited the paper. DA managed and audited the data, performed preliminary analyses, and edited the paper. We received helpful advice on questionnaire and study development from Mayer Hillman, Belita Clahar, Michelle Walsh, and Suzanne Slater, and on statistical analysis from David Dunn. We acknowledge the assistance of Jackie Payne, Elaine Morrison, the local education authorities in Camden and Islington, and the participating schools, teachers, and pupils.

Funding: The London Boroughs of Camden and Islington funded the study. The Camden and Islington Health Authority funded CD and DA.

Conflict of interest: None.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
View Abstract