• Ensure that the evaluation design is appropriate for the particular Fish Consumption Advisory (FCA) communication activity. Even if people are not exposed to the FCA program’s communication, they are likely to be exposed to some communication on the same topic. In these situations, appropriate designs include comparisons between cross-sectional studies (such as independent surveys taken at different points in time); panel studies (the same people are interviewed or observed multiple times); and time series analyses (comparisons between projections of what would have happened without the intervention versus what did happen). However, each is appropriate in different situations; seek the advice of an evaluation expert before selecting a design.
• Consider how the FCA messages and activities are expected to work and the time period in which they are expected to work. Then make sure evaluation is in accordance with those expectations. For example, if there is an expectation that members of the target audience immediately change their behaviors and eat less fish, then the outcome measurement should take place soon after the target audience receives the FCA messages. Conversely, if the FCA program does not expect to see effects for a certain duration of time, e.g. mercury blood levels dropping, then outcomes should not be measured until blood levels are expected to drop assuming the FCA is followed.
• Consider what level of evidence is acceptable for the outcome evaluation's purpose.
• Consider what baseline measures are available or can be collected and how to track changes related to desired outcomes (e.g., how, and how often, data will be collected). For example, a baseline measure can include survey results of target audience focus groups prior to issuance of a new or revised FCA or it could include collection of blood mercury levels prior to implementation of a new or revised FCA.
• Ensure that progress toward outcomes is captured. For example, if it is expected that members of the target audience think about changing a behavior, and perhaps adopt new fish consumption habits before making and sustaining the change, make sure the evaluation can capture these intermediate outcomes. If the objective is to increase the percent of the target audience who modify their fish consumption, it would be important to determine the target audience’s current behavior, and whether its members have thought about changing their behavior and whether they have actually changed it.

For more information, refer to What We Know about Fish Consumption Advisories: Insights from Experts and the Literature Exitor National Risk Communication Conference Proceedings Document (PDF) (196 pp, 708 K).