Does Relationship Education Work? What Named Studies Found

A plain-language tour of named meta-analyses and trials on marriage and relationship education: what they measured, what they found, and their limits.

kiss.events · Last reviewed October 24, 2026 · 6 min read

Relationship education programs, from premarital courses to weekend skills workshops, often describe themselves as "research-based." That phrase can mean anything from a carefully tested curriculum to a single satisfaction survey. If you are deciding whether a program is worth your time, it helps to know what the actual research has found.

The short version: relationship education has been studied a great deal, and the results are generally positive but modest, with important limits. The details matter, because which couples were studied, how outcomes were measured and how long researchers followed up all change the picture.

This guide walks through a handful of named, published studies in plain language. Every number below comes from the study's own published abstract. Relationship education is not therapy, and none of this replaces advice from a licensed professional about your own relationship.

First, how to read these studies

Two ideas make the findings easier to follow.

Control groups matter. The strongest studies randomly assign some couples to a program and others to a comparison group, then compare outcomes. Weaker designs simply measure couples before and after a program. Without a comparison group, it is hard to tell whether changes came from the program or from something else, so before-and-after studies can make a program look more effective than it is. Hawkins and Erickson's 2015 meta-analysis, described below, found exactly this: one-group before-and-after studies showed much larger effects than control-group studies.

Effect sizes. Many studies report a statistic called d, a standardized measure that lets researchers compare results across studies that used different questionnaires. Bigger numbers mean bigger differences between the program group and the comparison group. You do not need to memorize values; just compare them within the same study.

The 2008 and 2009 meta-analyses

A widely cited starting point is "Does marriage and relationship education work? A meta-analytic study" by Alan Hawkins, Victoria Blanchard, Scott Baldwin and Elizabeth Fawcett, published in 2008 in the Journal of Consulting and Clinical Psychology. The authors combined 86 reports covering 117 studies and more than 500 effect sizes.

For experimental studies, effect sizes for relationship quality ranged from d = .30 to .36, and for communication skills from d = .43 to .45. Moderate-dosage programs produced larger effects than low-dosage programs.

The authors were candid about the gaps. A lack of racial, ethnic and economic diversity in the samples prevented reliable conclusions about effectiveness for disadvantaged couples, and outcomes important to policymakers, such as relationship stability and aggression, had rarely been studied.

A 2009 follow-up by Blanchard, Hawkins, Baldwin and Fawcett in the Journal of Family Psychology looked specifically at communication skills, reviewing 97 reports covering 143 evaluation studies. They found modest evidence that relationship education works both as general prevention and for couples already showing problems: effects showed up at longer-term follow-ups for well-functioning couples, and at post-program and shorter-term follow-ups for more distressed couples.

They also found that observational assessments, where trained raters watch couples interact, produced larger effects than couples' own self-reports.

How much is enough?

In 2012, Hawkins, Stanley, Blanchard and Albright published a meta-analysis in Behavior Therapy that coded 148 evaluation reports for program features. Moderate-dosage programs, defined as 9 to 20 contact hours, were associated with stronger effects than low-dosage programs of 1 to 8 contact hours.

Some features that might seem important were not. There was no evidence that more institutionalized programs, with formal manuals, ongoing presence, formal instructor training and multiple evaluations, had stronger effects, and little evidence of differences between university or laboratory settings and religious settings.

Premarital education in a large survey

Scott Stanley, Paul Amato, Christine Johnson and Howard Markman analyzed a large random household survey across four middle American states, published in 2006 in the Journal of Family Psychology. Participation in premarital education was associated with higher marital satisfaction and commitment, lower conflict and reduced odds of divorce, and these associations held across race, income and education levels.

The authors note the key caveat themselves: survey methods allow only weak inferences about cause and effect. Couples who choose premarital education may differ in other ways from couples who do not.

Lower-income couples and public programs

Because federal grants have funded relationship education for lower-income couples, researchers have studied that group closely.

Hawkins and Sage Erickson's 2015 meta-analysis in the Journal of Family Psychology reviewed 38 studies with 47 independent samples of lower-income participants. Control-group studies showed small, positive effects (overall d = .061), especially on self-reports of relationship quality, communication and aggression. One-group before-and-after studies showed much larger effects (overall d = .352).
A 2022 meta-analysis led by Hawkins in Family Process examined programs funded through the U.S. Administration for Children and Families' Healthy Marriage and Relationship Education initiative, to which, the authors report, ACF has allocated $1.2 billion since 2006. Across 32 control-group studies, they found small but significant effects on couple relationship quality (d = .114), relationship skills (d = .132), mental health (d = .074) and coparenting (d = .033), but non-significant effects on relationship stability, parenting and child well-being.

A randomized trial with military couples

Stanley and colleagues tested a community-delivered version of PREP, the Prevention and Relationship Enhancement Program, with 662 married couples with a spouse in the U.S. Army, reported in Family Relations in 2014. The randomized trial found no evidence of overall enduring effects on relationship quality after two years. At the higher-risk site, however, couples assigned to the program were significantly less likely to be divorced at the two-year follow-up than controls (8.1% vs. 14.9%), and that effect was strongest for minority couples.

What this means for you

Taken together, these studies suggest a realistic set of expectations:

Relationship education can help couples build communication skills and modestly improve relationship quality, especially when programs offer a moderate number of hours.
Effects in the strongest studies are often small, and effects on staying together are mixed.
Before-and-after testimonials overstate what a program does. Ask whether a program has been tested against a comparison group.
Education is a prevention and skill-building tool. It is not designed to treat serious distress, mental health conditions or abuse.

If your relationship is in serious distress, a licensed therapist is a better starting point; see our guide Couples Counseling Explained: What It Is and How It Works. If you are afraid of your partner, contact the National Domestic Violence Hotline at 1-800-799-7233; the Hotline does not encourage anyone in an abusive relationship to seek counseling with their partner. If you or your partner are in crisis, call or text 988.

Key takeaways

A 2008 meta-analysis of 117 studies found positive effects of relationship education on relationship quality and communication, but limited diversity in the samples studied.
Programs of 9 to 20 contact hours were linked to stronger effects than shorter ones in a 2012 meta-analysis.
Before-and-after studies show bigger effects than controlled studies, so treat testimonials with caution.
For lower-income couples and federally funded programs, effects were small but positive for relationship quality, with non-significant effects on stability in the 2022 review.
Relationship education builds skills; it is not a substitute for therapy and is not appropriate when abuse is present.

Sources

Spotted something out of date? Let us know.