A picture is worth a lot of words and it can really help us explore the decisions people make. This is the main lesson from a recent online ethnography study conducted around food and nutrition. With plenty of quantitative data, the online approach was adopted to not only gives colour to the quantitative data but to explore the decision-making process for Canadian families when it comes to food.

When we can see the kitchen, the plate, the food and the other aspects of food consumption, we gain considerable insight into how people eat.

Many people say that they eat healthy and claim to be knowledgeable about nutrition. At the same time, it seems like we continue to have growing problems of obesity and numerous other health issues.

Self-reports have always been a challenge for survey research. We can expect two things to happen with self-reports. The first is that people accidentally (they forget; or view their eating through distorted glasses) or intentionally misrepresent their eating behaviour. The second is that they have individual understandings of what healthy eating and nutrition mean because, for example, they are not well informed about healthy eating.

Consider this picture of a shopping basket. The basket seems balanced with fresh fruits and vegetables along with milk, cheese and chicken. There is also prepared breakfast potatoes. Picture the person who is pushing this cart. How healthy does this family seem? Why are they buying prepared breakfast potatoes?

For five days, we (Pierre Belisle moderated the discussion) asked a group of Canadians, from all walks of life to undertake a series of daily-life tasks (eg. A solo meal; a shopping excursion) and had them upload pictures and tell us about what was happening. The project, an initiative of the Canadian Council of Food and Nutrition (CCFN), was designed to provide a unique perspective on the long-running Tracking Nutrition Trends (TNT) study. The results of the study will be available in the fall of 2010.

With field now closed, our assessment of the process is largely positive. In a modified form, we were able to observe participants (through their photos) in an efficient manner and still get the benefits of an active moderator. Additionally:

  • Very few of the participants who were recruited did not participate in the five day project. In addition, most who did participate were active in their participation from beginning to end.
  • Participants covered a diverse range of family compositions and the online tool appears to be useful for a range of study types.
  • The online survey software, Revelation, was quite intuitive and easy to use for participants and observers.

In quantitative research, self-reports are inevitable. We just can’t help but ask people about what they do or have done. When possible, getting a chance to observe is even more valuable.