Factors Impacting Program Retention

By Sarah Hokanson

Bottom-Line First: A recent study of a home-based couple relationship education program for lower income couples looked at factors that predicted early drop-out and session attendance. About a quarter of couples dropped out after the first session. This drop-out rate is not much different from what has been found in other studies, despite the more flexible delivery system. Younger couples and cohabiting couples were more likely to drop out. And interestingly, more committed couples were more likely to drop out early.

Low-income, unmarried couples tend to be more at risk for relationship problems and family dissolution than other couples. In addition to financial stress and less certainty about the future, being younger, less healthy, and less educated increase their relationship risk factors. Some past studies have found that couple relationship education (CRE) has no significant impact for low-income unmarried parents, but others have found positive effects. One factor that likely contributes to low or no impact may be high drop-out rates, which were seen in some studies. Also, higher participation rates have been shown to be associated with significantly stronger effects.[1]

The current study[2] looked at factors that were associated with participant retention in CRE programs.  This study used the 7-session Couple CARE for Parents (CCP) program, a home-based program using web-based content and workbooks along with personalized coaching (home visits in sessions 1-4 and phone calls after that).

Providing relationship education in the home or online lowers the barriers to enrollment and retention because it allows for more flexibility for participants and lowers logistical barriers to participation. Using the CCP program, this study looked at how various individual, relationship, and program factors were associated with early drop-out (after only the first session) and overall session attendance. The sample was made up of 467 couples (43% White, 21% Hispanic/Latino, 24% African American) from the New York City area who had recently had a baby.

Here is a brief summary of key findings:

  • Almost a quarter of participants (23%) dropped out after only one session; drop-out rates for subsequent sessions were much lower (between 3-11%).
  • 43% participated in all seven sessions. On average, participants completed 4.4 of 7 sessions.
  • For both men and women, younger age predicted participation in fewer sessions.
  • Lower ratings of female program engagement (rated by the coach) and lower ratings (by the coach) of the facilitator-participant alliance (in the first session) predicted early drop-out (for both women and men).
  • Participants with higher relationship commitment at the start of the program were more likely to drop out early. Similarly, cohabiting couples were more likely to drop out compared to married couples.

Implications of these findings highlight what relationship educators can do to get past these program-retention hurdles.

  • There is a need for CRE for low-income, unmarried couples. Over 700 couples enrolled in the program (though not all were included in this study), indicating that there is a significant interest in CRE from low-income, unmarried couples. This suggests that program administrators could spend more time focusing on this population through intentional recruitment and building programs that are relevant to their needs. We might consider providing relationship education programs at locations that offer other community resources in order to better reach low-income families.
  • Flexible delivery does not significantly improve program completion rates. The dropout rate for this study was at a similar level to dropout rates in other programs, which shows that eliminating the logistical barriers to participation by providing flexible delivery does not do much to improve couples’ program completion. Providing flexible delivery for our participants may be helpful in reaching those who could not come to a traditional relationship education program and may lead to increased enrollment, but it is no guarantee that enrolled participants will complete the program.
  • Active recruitment may contribute to high dropout rates. This study used active recruitment methods, which means that couples were not actively searching for relationship help from a CRE program. Couples may have initially enrolled in the program but decided not to continue after realizing that they did not have a lot to gain from the program. This could help to explain why couples with high commitment had high early-dropout rates. Focusing our recruitment towards those who need it (perhaps through internet searches and social media ads, as discussed in this previous blog post) can improve retention and allow us to focus our resources on those most in need.
  • Educators should pay attention to participant engagement and coach-participant alliance in the first session. Another predictor of early dropout was low engagement and alliance. Building a good relationship with coaches right from the start can help to counteract this problem. The study’s authors recommended telephone contact with coaches before the first session and discussions between coaches and participants to talk about expectations. These discussions may include talking about the couple’s expectations and goals in strengthening their relationship and making plans about what the coaches can do to help each couple with their specific needs.
  • Increase flexibility by allowing customized content. The study’s authors make the argument that since many couples only receive one or two sessions of a CRE program, we should fill those sessions with the most important and relevant content. Talking to participants about what they need and want to learn and then organizing content accordingly can help to maximize the benefit to participants. If participants are interested in doing things in a different order or only have time for a few lessons, this kind of tailoring will help them to make the most of their CRE experience.



[1] Dion, M. R., Avellar, S. A., & Clary, E. (2010). The building strong families project: Implementation of eight programs to strengthen unmarried parent families. U.S. Department of Health and Human Services, Administration for Children and Families, Office of Planning Research, and Evaluation; Wood, R. G., McConnell, S., Moore, Q., Clarkwest, A., & Hsueh, J. (2012). The effects of building strong families: A healthy marriage and relationship skills education program for unmarried parents. Journal of Policy Analysis and Management, 31, 228–252. https://doi.org/10.1002/pam.21608

[2] Bulling, L. J., Baucom, J. W., Heyman, R. E., Smith Slep A. M., Mitnick, D. M., & Lorber, M. F. (2020). Predicting program retention in a flexibly-delivered relationship education program for low-income, unmarried parents. Journal of Family Social Work, 23(3), 234–256. https://doi.org/10.1080/105221582019.1681337