a1 Department of Nutrition, University of Oslo, PO Box 1046 Blindern, 0316 Oslo, Norway
a2 Department of Public Health, Erasmus University Medical Centre, Rotterdam, The Netherlands
a3 EMGO Institute, VU University Medical Centre, Amsterdam, The Netherlands
Objective To explore why boys eat less fruit and vegetables (F&V) than girls, using longitudinal data following pupils from the age of 12.5 to 15.5 years, including perceived accessibility, modelling, intention, preferences, self-efficacy and knowledge of recommendations as potential mediators.
Design/setting/subjects A longitudinal study, based on data collected among the control schools within the intervention project ‘Fruits and Vegetables Make the Marks’. This sample contains 896 6th and 7th graders from 20 randomly selected elementary schools within two Norwegian counties. Questionnaires were administered in May 2002 and May 2005 (when the participants were in 9th and 10th grade in 18 secondary elementary schools).
Results In single mediation analyses all determinants mediated parts of the gender difference, but only preferences decreased the gender difference to a level below statistical significance. Preferences alone explained 81% of the gender difference. In the multiple mediation analyses, the six mediators together explained 91% of the gender difference, but only preferences and perceived accessibility contributed uniquely to the explanation, with 25% and 10% respectively.
Conclusions Preference appears as the strongest mediator of the difference in F&V intakes between boys and girls. Further research should explore why girls like F&V more than boys.
(Received December 05 2006)
(Accepted May 15 2007)