Just getting starting in evaluating risk communication? Here are five basic but vital tips:
1. Measure before and after.
The key to being able to attribute change in attitude or behavior is knowing levels of awareness, understanding, and behavioral intent before you start your campaign. You need a benchmark against which to compare the same measures after your campaign, and thereby judge your performance.
2. Be inclusive and expansive in targeting and evaluating.
No matter who or where you are, or what your risk is, your audience isn’t homogeneous. To be effective in today’s world, risk communication must transcend barriers of literacy, language, and ethnicity to ensure acceptance and understanding.
Do not lump your stakeholders into one big bucket. Take into account the diversity of your communities. And we’re not just talking non-English speaking. Have you considered the homeless and homebound? Faith leaders, veterinarians, tourists, local residents who might be out of town, institutionalized populations? The NRC Guidelines has an amazingly robust list of possible target audiences.
Even better, by segmenting your audience you can compare your results between groups. That way you’ll more readily discover which might be dragging your numbers down, and with which audience your messages are most effective.
3. Balance multiple choice with open ended questions.
If you survey your audience to test its knowledge of what to do in a crisis, then you can’t give them the answers ahead of time. Which is just what most people do when they ask multiple choice questions. What you really want to know is if they can recall your instructions without the aid of a prompt.
4. Crosstabs, crosstabs, and more crosstabs.
When you analyze your data to identify the strengths and weaknesses in your campaign, it is crucial to have a full data set in front of you. If your numbers aren’t moving in the right direction, you will need to be able to identify whether specific groups are responding better than others, and where they’re not responding at all.
5. If your survey vendor won’t provide you with the full data set, fire them.
If you paid for your data then you own it. If they refuse to turn over the actual data, it’s probably because it’s crap. ∞