Putting Behavioral Science Insights into Practice
One of the goals of the Behavioral Interventions to Advance Self-Sufficiency (BIAS), BIAS-Next Generation, and Behavioral Interventions for Child Support Services (BICS) projects is to help practitioners apply insights from behavioral science to their own programs. In recent months, the BIAS project has shared a number of resources in support of this goal. BIAS released the SIMPLER framework of frequently used behavioral science insights, described how behavioral interventions can range from small changes to more comprehensive system reforms, and hosted a webinar on applying behavioral science to human services programs. This Behavioral Buzz issue builds on those resources with three concrete tips for designing interventions informed by behavioral science, based on the experience of BIAS and BICS.
1. Customize interventions to participants’ characteristics and contexts
Different types of interventions may be more or less effective for different types of participants. The BIAS project observed this during its partnership with the Paycheck Plus demonstration (which offers an Earned Income Tax Credit-like earnings supplement for low-income single adults). Participants who were employed part-time or who were unemployed at the start of the demonstration were significantly more likely to attend a meeting in response to the team’s behavioral science-informed messaging than they were to standard messaging, while those who were employed full-time were not more likely to attend in response to the behavioral messaging than they were to standard messaging. A next step could be to develop and test revised messages or interventions specifically targeting participants employed full-time, in an attempt to boost their meeting-attendance rates. As many for-profit businesses do, government agencies and nonprofit organizations could tailor interventions to individuals based on factors available in existing administrative data.
When developing an intervention, it can also help to consider whose decisions or behavior you hope to influence and the context in which those people operate. Are participants strapped for time and attention? An implementation prompt could help them make a decision about taking action and following through. Are they students taking classes together and therefore able to observe the behavior of their fellow students, or likely to care about their peers’ opinions of them? Consider making use of social influence, particularly if you suspect they may have inaccurate, counterproductive perceptions of what their peers are doing. Customizing interventions based on individuals’ characteristics or contexts can increase the effectiveness of those interventions. As always, testing the interventions is important to determine whether they are working as intended.
2. Time interventions carefully
Small factors can have large impacts on an intervention’s effectiveness. For example, interventions are more likely to succeed if they are delivered in the moments when people are already thinking about the subject of the intervention or are able to pay attention to it-- and are able to respond immediately. No one can pay attention to all of the stimuli, ideas, and behaviors that surround us on a daily basis. People have limited attention and typically focus on just one or two issues at a time. Instead of delivering an intervention when someone is thinking about another topic — in which case the message first has to catch a person’s attention before it can potentially affect that person’s behavior — program designers can time interventions for moments when staff members or participants are likely to be receptive. For example, researchers have identified a “fresh start effect,” finding that interest in dieting, exercise, and pursuing new goals peaks during certain time periods, such as the beginning of a new year, week, or school semester, or even following a birthday. The BIAS project observed the impact of timing in the Oklahoma child care subsidy study. Families in this study were more likely to renew their child care subsidy on-time when their child care providers received behaviorally informed outreach, but not when the families themselves received outreach. Providers may have been effective at encouraging clients to renew on time because they interacted with clients at key times when benefits of the subsidy were salient, when clients were at child care homes or centers with their children. Look for these sorts of distinctive times to encourage people to take up healthy living habits, look for work, or increase their saving.
3. Start from scratch
During the dozens of ACF-sponsored behavioral science trials that have been launched, the research teams have frequently encountered existing communication materials that do not reflect the insights of behavioral science. These materials may be confusingly worded, poorly organized, and visually unappealing. Instead of trying to edit existing materials, the teams have found that it is more effective to start from a blank page. Doing so prompts you to start by specifying the behavior you are trying to produce. Do you want participants to attend a meeting? Do you want staff members to remember to offer a specific service to participants? Answering these questions may help you realize that the existing materials provided unnecessary information or could be communicated more concisely if they were redesigned. For example, staff members from the Office of Child Support Services in Cuyahoga County, Ohio (Cleveland) and researchers working on the Behavioral Interventions for Child Support Services (BICS) project were able to replace a nine-page packet with a one-page form and still collect all the information necessary for an order-modification review.
Other resources
In addition to these tips and the BIAS resources noted at the beginning of this issue, several other researchers and organizations have also produced helpful resources for practitioners. Dilip Soman, Professor at the University of Toronto’s Rotman School of Management and a BIAS and BICS project adviser, has developed Behavioral Economics in Action, a self-paced online course introducing behavioral science and some essential concepts related to intervention design and evaluation. In 2016, the federal government’s Social and Behavioral Sciences Team released a guidance document to help agencies apply insights from behavioral science. This memo provides a review of common behavioral science-informed intervention strategies.
The BIAS Next-Generation project will be developing an extensive “toolkit” for practitioners over the next few years to provide further suggestions for how to incorporate insights from behavioral science. If you have ideas about what resources would help you and other practitioners apply behavioral science to your programs, send an e-mail to the project team at bias.info@mdrc.org.
|
