Different modes of gathering data on self-reported health measures and self-reported risk factors are used frequently in research. However, data on the influence of the mode of collection of data on self-reporting are limited. The aim of the study was to identify associations between the mode of data collection and self-reported smoking in two distinct ethnic groups, Jews and Arabs in Israel. During the last 2 years, data were collected in two national surveys regarding the smoking behaviors of Jews and Arabs in Israel. In the telephone surveys 4713 Israeli residents were interviewed and in the face-to-face interviewees 3239 people were interviewed. The interviewees were between the ages 25 and 64. There was no significant difference in smoking rates between face-to-face interviews and telephone interviews among Jewish men or women after adjusting for other variables associated with smoking. However, there was a difference between the two methods of data collection in the Arab population also after the adjustment. In this group, respondents tended to report more often being a smoker in the face-to-face interviews. This was especially apparent in Arab women. There was no significant difference in the reported number of cigarettes smoked in the two modes of data collection. In Arabs compared to Jews there is a significant difference between reporting smoking during a telephone interview and a face-to-face interview. The mode of data collection can affect comparisons between different groups.