Many of the behaviors that we would like to promote have no outward visibility in our communities. When behaviors such as getting vaccinated are not visible to others, we weaken the likelihood that the action will become normative. Unlike vaccinations, wearing a mask or physically distancing are easily observed, and in many communities, they have become normative as a result. A key aspect, then, of designing programs involves making less visible behaviors visible so that they can become normative. To that end, consider how you can best let others know that you have either been vaccinated or intend to be vaccinated and encourage others to do the same. I feel very fortunate to have received my first shot this morning.
Selecting Behaviors
As noted in Parts 1 and 2 of this missive on selecting behaviors, choosing which action(s) to target involves several steps. First, ensure that each behavior under consideration is non-divisible and end-state. Remember that you want to create a list of specific behaviors, such as cycling to work, rather than a list of general behaviors, such as active living. Further, recall that end-state behaviors produce the desired environmental, health, or safety outcome (e.g., biking to work, rather than purchasing a bike with the intention of cycling to work).
Once you have created a list of end-state, non-divisible behaviors, these actions need to be scrutinized by analyzing three characteristics:
- The impact of each behavior;
- The probability of adoption of each behavior; and
- The level of penetration that already exists for each behavior.
In last week’s Minute, I described in detail impact, probability and penetration. In this week’s edition, I’ll cover how to measure impact, probability, and penetration and demonstrate how to utilize these characteristics to determine which behaviors are most worth promoting.
Determining Impact
Two methods exist for determining the impact of each of your listed behaviors. The first, and preferred method, involves consulting research that sets out the impact of each behavior. Imagine that you are delivering a program to reduce residential CO2 emissions. Research reports exist that detail the emissions associated with such diverse behaviors as installing and configuring a programmable thermostat, turning down the hot-water heater temperature, and washing clothes in cold water.
When reliable information regarding the impact of these behaviors does not exist, you can estimate impact using a second method.
To estimate the impact of any particular behavior, survey individuals who have expertise regarding the actions on your list and ask them to rate each behavior using a scale from “0 – no impact” to “10 – significant impact.” Your findings will be more reliable if you survey a minimum of 12-15 experts, and you obtain the ratings independent of one another (e.g., do not bring the experts together and have them discuss the relative importance of the behaviors). Once you receive each expert’s ratings, calculate the average impact for each behavior. It is common to be able to obtain impact ratings from experts in the space of several hours. My preferred way of conducting these expert surveys is by phone.
Determining Probability
Two methods also exist for determining probability. The favored method is to examine past programs to assess their effectiveness in encouraging specific behavioral changes. In reviewing these programs, collect information regarding what percentage of an audience adopted the behavioral change and how the program was delivered. You will also want to examine the context within which the program was delivered (e.g., does a community have high energy costs, which would provide additional motivation for community members to adopt energy efficiency behaviors?).
While investigating past programs provides the most reliable information regarding probabilities, what if your list of end-state, non-divisible behaviors is long? In this case, you should use the second method. The second method involves surveying your target audience and asking them to rate the probability of engaging in various behaviors. To obtain these ratings, use a scale where “0” equals “no likelihood” and “10” equals “high likelihood.” For these ratings to be meaningful, you will need to provide some context for the questions. For example, rather than simply asking, “How likely are you to install a high-efficiency showerhead?” ask, “How likely are you to install a high-efficiency showerhead if you had to purchase and install it yourself?” Strive to survey a representative sample of your target audience to ensure that the obtained probabilities are indicative of your community.
Determining Penetration
As with determining impact and probability, two methods also exist for determining penetration or adoption rates. If the behavior is observable, such as transit usage, curbside recycling, or bicycling, direct observations of the behavior can be used to determine the percentage of an audience that has already adopted the behavior. Unfortunately, many behaviors have little or no visibility. These include most behaviors related to residential energy efficiency or dietary choices made in the privacy of one’s home. In these cases, survey the target audience to determine participation in the behavior. To save time and money, you can combine the surveys investigating probabilities and penetration.
Calculating Impact x Probability x Penetration
The final step is to multiply the values you have obtained for impact, probability, and penetration. But, before you do, invert the penetration values. For instance, if 20% of households have installed water-efficient showerheads, subtract this number from 100 to obtain the percentage that has yet to participate in the behavior (80%). Create a table with five columns. The first column includes your list of behaviors. The second column is the impact rating for each behavior that you obtained from the research reports or experts. The third and fourth columns are the averages for probability and penetration (inverted), respectively. Finally, calculate the values for the fifth column by multiplying impact x probability x the inverse of penetration for each behavior. The values in the fifth column can then be rank-ordered to determine which behaviors are most worthwhile to promote (apologies that I haven’t included a table to demonstrate this, but the software that I am using for the Minute doesn’t permit the inclusion of tables – ugh!).
Final Thoughts on Behavior Selection
Too frequently, programs are delivered because someone in an agency felt it would be worthwhile to promote a specific behavior. If we are to make more significant progress in fostering health, safety and sustainability, we need to be more rigorous in selecting which behaviors are most worthwhile to target. By focusing on those behaviors with the best combination of impact, probability, and penetration, we can be more assured that our programs will have the desired outcome.
If you find the Minute helpful, please considering sharing it with your colleagues. If you would like to learn more about community-based social marketing, sign up for my in-depth monthly Fostering Behavior Change Reports here. I’m also delivering virtual Fostering Sustainable and Healthy Behavior workshops in June. Full details here.