OBJECTIVES: To test to what extent the presentation of choice tasks contrasting display in words or graphics, influences the attribute estimates, relative importance and participation probability and the conclusions drawn from a Discrete Choice Experiment (DCE).  METHODS: A DCE questionnaire was sent to the parents of 2,500 newborn babies aged 6 weeks at maximum. Each questionnaire contained two versions of the same 9 choice tasks, one in which the levels were presented in words, and one with graphic attribute levels. The DCE consisted of five attributes: vaccine effectiveness, frequency of severe side effects, protection duration, location of vaccine administration, out-of-pocket costs. Choice consistency was estimated, panel-mixed logit models were conducted to estimate the relative importance of the attributes and internal sample validation was calculated.  RESULTS: In total 13% answered consistent on all choice tasks, 19% answered inconsistently in one choice task, and 51% answered inconsistently in more than two choice tasks. Respondents who were presented with word choice tasks at first were significantly more consistent compared to respondents that were presented with graphic choice tasks at first. Although out-of-pocket costs was the most important and frequency of severe side effect was the least important attribute  in both datasets, the relative importance of the other attributes differed.  All results differed by educational level. Estimated and observed choices showed higher correlation in the word dataset. CONCLUSIONS: The presentation of the choice tasks by either using words or graphics influences study outcomes. The use of graphics to present attribute levels in choice tasks seems less preferred by respondents and showed a reduced validity. Besides extensive pilot testing, discussions about the presentation of the choice tasks should be included in the focus group stage of the designing process. Extensive research on respondents’ interpretation of choice tasks should be conducted to enable the formation of choice task presentation guidelines.