Earnscliffe Strategy Group (Earnscliffe) is pleased to present this report to Health Canada summarizing the results of qualitative research exploring the new Canada’s Food Guide visual, look-and-feel and supporting messages.
Health Canada is revising Canada’s Food Guide (CFG) to reflect the latest scientific evidence, and to be more modern and relevant to users. This includes a release of a suite of products using new communication tools to meet the needs of a variety of audiences. The new products include a CFG visual. In addition, a new CFG look-and-feel will be applied across the suite of products. The total cost to conduct this research was $208,218.47, including HST.
The research was conducted over two phases.
Phase 1
The objective of the first phase of research was to test a sample of messages to assess whether the various tones and styles resonated with the various audiences, with the goal to inform the development of final messages. Health Canada was interested in understanding whether the messages are: motivating; using the appropriate tone; providing the right amount of information to meet their desired needs; practical, relevant and helpful, given the task setting at hand.
The messages were targeted to five different audiences and focused on a variety of healthy eating topics. The five audiences included: adults experienced in food preparation; adults with minimal experience in food preparation; seniors responsible for food preparation; parents of children who are responsible for grocery shopping and food preparation; and, youth aged 16 to 18.
This phase included a series of ten focus groups across two cities – Ottawa (March and 20) and Quebec City (March and 21). The focus groups in Quebec City were conducted in French.
Phase 2
The objective of the second phase of research was to test the look-and-feel and visual elements for the new Canada’s Food Guide to assess: effective use of text and graphics/images; credibility, relevancy and perceived value to the audience; acceptance; appeal, usefulness and appropriateness; relevance and engagement; and memorability (eye-catching and general visual appeal).
Audiences for the second phase included: those at risk of marginal health literacy [as screened by the Newest Vital Sign (NVS) and scoring <4/6]; those with adequate health literacy (score 4+); primary level teachers; community level educators; registered dietitians working in public health or community nutrition; registered dietitians working in clinical/private practice/media/bloggers; and, registered nurses working in public or community health.
Ten focus groups were conducted with members of the general public in five Canadian cities: Toronto, ON (June 5, 2018); Quebec City, QC (June 6, 2018, in French); Calgary, AB (June 7, 2018); Whitehorse, YK (June 11, 2018); and, St. John’s, NL (June 14, 2018). Fifteen (15) mini-groups were conducted with health professionals and educators in three Canadian cities: Toronto, ON (June 4, 2018); Calgary, AB (June 6, 2018); and, Quebec City, QC (June 18, 2018, in French).
Please refer to the Recruitment Screener in the Appendix of this report for all relevant screening and qualifications criteria.
It is important to note that qualitative research is a form of scientific, social, policy and public opinion research. Focus group research is not designed to help a group reach a consensus or to make decisions, but rather to elicit the full range of ideas, attitudes, experiences and opinions of a selected sample of participants on a defined topic. Because of the small numbers involved the participants cannot be expected to be thoroughly representative in a statistical sense of the larger population from which they are drawn and findings cannot reliably be generalized beyond their number.
Phase 1
REACTIONS:
- While the information presented was not necessarily new to respondents, their overall reaction to the messages was quite positive. They felt the messages included important reminders and tips to help them build better health eating habits.
- Participants felt the tone of the messages was appropriate, however, they preferredpositive and friendly tips that presented advice as a suggestion, rather than messages with a negative, condescending, and imperative tone.
- The language was clear and easy to understand, albeit wordy and redundant at times. Participants preferred short, concise messages, with the exception of longer messages that provided specific advice, particularly for those with minimal experience in food preparation.
- Most participants tended to prefer messages with a rationale, arguing that it gave them the “why”. Their preference was similar to the earlier point about a preference for suggestive over imperative messages.
- Reaction towards messages about what and how to eat was mixed, though the findings suggest there is room for both in the communications. Those experienced in food preparation felt information about how to eat was new. Those with minimal experience in food preparation tended to prefer advice about what to eat that included specific instructions and a rationale.
- The advice was deemed practical. On many occasions, participants remarked (unaided) that they would take action as a result of something in the messages.
- There was some confusion around the title of a couple of topics: Beginner Cook and Celebrations. In English, Beginner Cook seemed to refer to young people learning to cook; while, Celebrations seemed to refer to family holiday traditions. Participants generally appreciated the advice in both topics although some participants were frustrated by the advice under Celebrations to forego or modify family traditions and recipes.
- Youth and parents had a similar reaction to the topic of Eating Out – both felt that eating out in restaurants was an infrequent occurrence and a time for indulgence, not sacrifice.
- Youth appreciated the topic of Eating on the Go for its specific snack ideas to eat during points in the day when they are both hungry and rushed. That said, they would not use the topic heading, Eating on the Go. Something to do with “snacks” seemed more relevant to them.
- Other topic areas upon which participants suggested included:
- Advice related to Fueling Your Body for Sport/Exercise;
- Eating Healthy on a Budget; and
- Eating Healthy While Travelling (e.g. road trips).
Phase 2
REACTIONS: LOOK-AND-FEEL ELEMENTS
- The overwhelming majority of participants preferred the light blue concept with the illustration of the tagline for look-and-feeling purposes (Concept B). It was the look-and-feel concept that most described as warm, friendly, relatable, positive, interesting, and attention-grabbing. Participants described the illustration as a representation of Canada, including people in urban and rural settings, and depicting food from farm to table. This concept was most closely linked to food, though many felt that food could be depicted more clearly, citing examples such as apples in the trees, or crops and/or animals in the fields, etc.
- The green concept (Concept A) was eye-catching, but unappealing. Most found the green to be a little jarring and had trouble reading the tagline. Participants described the overall look as amateur and juvenile. They also felt this illustration suggested that this was Canada’s Food Guide for urban/city dwellers (given the heavy emphasis on buildings in the illustration). The majority of participants did not think this look-and-feel naturally fit with food or Canada’s Food Guide, nor did it tie in closely with the campaign tagline: Where You Live, Learn, Work, and Play.
- Reactions to this concept worsened when participants were shown the web button formats. The placement of the buildings made the concept feel amateur.
- The plain blue concept (Concept C) was felt to be boring and “typical government”. Very few, if any, felt they would be interested or inclined to read the guide based on these look-and-feel elements. Participants also did not like that this look-and-feel concept did not have an obvious link to food or the tagline. However, of all three look-and-feel concepts, it was most often deemed modern because it was simple and minimalist.
- For this reason, some participants thought that this concept worked well as a web button (the smaller format). This simpler format would be eye-catching and something they might click on.
REACTIONS: VISUAL ELEMENTS
- Overall reaction to the digital and print versions of both concepts was generally positive although many participants, particularly health professionals (in Quebec City), were slightly underwhelmed. In large part, this was in reaction to the digital concept which seemed to have two perceived weaknesses: 1) the lack of interactivity; and 2) the use of icons rather than real photos/images. The compounded effect of the two suggested an amateur product, not that different than the current Food Guide. However, impressions tended to improve when participants were shown the printed version of the “at-a-glance” concept, because all the information was visible at a glance on one page, as opposed to the digital experience which required scrolling to get the full picture. Impressions of the digital experience improved once participants were told that the intention was to create an interactive experience.
- The majority of participants felt that both digital and print versions would be useful.
- Health professionals and educators would use the digital concept with their target audiences and then send them home with a printed version to put up on the fridge.
- Members of the general public felt they would use the digital version when they were shopping or preparing their grocery lists, and tack a printed version up on their fridges.
- Participants were surprised and encouraged to see new information that encourages healthy eating behaviours, featured so prominently, in addition to advice about healthy food choices.
- In terms of overall look and feel, participants generally seemed to coalesce around Concept B (blue background), particularly when they were shown the printed version, due to:
- The overall professional quality;
- How the illustration complemented the tagline and food;
- The number and size of food icons; and,
- The clarity of the black font against the blue background.
- In terms of the presentation/order of information, preferences were mixed but most tended to prefer the order of Concept B. They suggested that this order allowed them to very quickly get information about how to eat and what to eat, which they felt was a positive start, followed by information about what to limit. Those who preferred the order of Concept A preferred to understand how to eat and what to limit, and then equipped with that information, they could move to what to eat.
- Of note, almost all participants tended to prefer the order of Concept B in the printed version, with suggestions for adjustments to some of the icons. These adjustments included suggestions like using the food icons for “Limit foods high in sodium, sugars or saturated fat” instead of the warning label.
- Respondents were confused about what the icons represented and felt they could be improved. Participants felt the icons in the how to eat section (i.e., cook more often, eat with others, etc.) were sterile and impersonal. They suggested including people in the visuals. In terms of the food icons, participants were generally confused by the: squash, bagged lettuce, frozen berries, wild rice, oats, quinoa, brown rice, whole grain pita bread, tofu, peas, lentils, and the pork chop.
- The inclusion of canned and frozen goods was met with mixed reaction. Those in favour argued that it was important to include economical and time-saving food options. Those opposed argued that they could lead Canadians to making unhealthy choices (i.e., processed foods), but also that the depiction of packaged goods promotes the use of excessive packaging.
- Unprompted, most participants did not infer much in the way of proportionality. When asked, some could delineate that because vegetables/fruits occupy a larger space visually, or in the example of Concept B that vegetables/fruits were displayed at the top, that most of their food should come from this category.
- Most indicated that they would have preferred a more direct reference to either specific proportions or an image that illustrated proportions.