Program development: at its best, it’s a mixture of science and art, of empirical evidence and informed, innovative guesswork. That’s the message at the heart of a new book on conflict between parents.

Using family relationships as an example, two psychologists set out three steps for tackling the biggest challenge in translational research: how to translate the findings of basic research into real, workable programs.

The first step is to be clear about what the literature says. Systematic reviews or meta-analyses can help to highlight similar findings and to show where there are significant relationships between factors.

Second, have a theoretical framework to guide your program development. Theory provides the conceptual basis that brings the program components together. (Within the example of marital conflict and family relationships, the authors advocate Emotional Security Theory.)

Third, look further afield – outside your own discipline – for evidence that might complement your work. Innovations on other topics may offer new insights and evidence to support a course of action.

However, after these steps are completed, program development is “as much art as science”, Mark Cummings and Patrick Davies, both leading researchers on family relationships, argue. After all, science provides answers only to those questions that have been asked, so the three scientific steps only go so far: a degree of art, innovation or “best guess” will be necessary to fill in the rest.

In the specific case of conflict between parents, Cummings and Davies show how evidence and expertise combine to create real-world interventions.

The process of designing a program begins with the target population and intended outcomes, which guide the approach and format of the intervention. There will be differences, for example, between prevention work targeting family functioning in the community and clinical interventions providing intensive treatment to families where breakdown has already occurred.

To reduce the impact of marital conflict on children, interventions should be aimed at parents rather than children, the authors say. Educating parents – teaching them about how their conflicts affect their children and about how to resolve their disagreements – has a larger impact on children than working with children directly. But there is a balance to be struck: teaching young children how to cope with their parents’ conflicts may pay dividends later, by preparing them to deal with their own adolescent relationships. On the other hand, there is a risk that interventions alter how children respond to conflict without paying attention to how this feeds back into the wider family dynamics.

The authors also looked at the balancing acts involved in deciding the duration of a program and the best way to engage participants. They found that other studies provided evidence about the right “amounts” of service and successful methods, such as games, group discussions and one-on-one interactions. In addition, staff must be well trained and supported by manuals that provide both the process of delivery and the content of the intervention. Fidelity measures ensure a consistent approach to delivering the program.

Finally, how will the success of the program be measured? The evaluation should fit with the scope of the intervention, and it should assess whether, and to what degree, the intended outcomes were achieved. This is arguably one area where science should win out over art.

See Cummings, E.M. and Davies, P. T., Marital Conflict and Children: an emotional security hypothesis, The Guilford Press, New York, 2010.