Enhancing Marital Enrichment Through Spirituality:Efficacy Data for Prayer Focused Relationship Enhancement

Steven R. H. Beach and Tera R. HurtUniversity of Georgia

Frank D. FinchamFlorida State University

Kameron J. FranklinUniversity of Georgia

Lily M. McNairSpelman College

Scott M. StanleyUniversity of Denver

We examined 393 African American married couples assigned to (a) a culturallysensitive version of a widely disseminated relationship enhancement program (CS-PREP); (b) a similar version of the same program that also included a focus on prayer(PFP condition); or (c) an information-only control condition in which couples receiveda self-help version of the same program. Husbands averaged 40.5 years of age andwives averaged 38.9 years. We found a significant interaction between intervention andtime of assessment, reflecting group differences in linear trends for the three conditions,with the two intervention conditions performing better than the control condition, andPFP producing superior outcomes to CS-PREP only for wives at post and marginallybetter results at 12-month follow-up. Results support continued exploration of theadjunctive use of prayer in the context of relationship enhancement programs whereappropriate to make them more culturally sensitive.

Keywords: prayer, marital, intervention, spirituality, African American

As efforts to strengthen marriage attractwidespread attention, identification of barriersto dissemination and ways to overcome thosebarriers are timely (Stanley, Amato, Johnson, &Markman, 2006). Such efforts are particularlyimportant for African American couples, who

are underserved by typical means of health-related program delivery and who may perceivefamily based services to be culturally irrelevant(Murry et al., 2004). At the same time, whereas33% of marriages among Whites end within 10years, 47% of African American marriages endduring this time frame. In addition, among thosewho are currently married, African Americancouples consistently report lower satisfactionthan do White couples (Acitelli, Douvan, &Veroff, 1997; Adelmann, Chadwick, & Bae-rger, 1996; Roebuck, & Brown, 2007), or thanBlack Caribbean couples (Bryant, Taylor, Lin-coln, Chatters, & Jackson, 2008). Accordingly,the need for relationship enhancement tar-geted to African American couples is greatand the need for culturally sensitive programsdesigned to enhance engagement is acute(Johnson et al., 2002; Karney, Garvan, &Thomas, 2003).

These considerations underscore the need forculturally sensitive programming that may en-hance both engagement and retention of AfricanAmerican couples in community-based, marital

This article was published Online First May 2, 2011.Steven R. H. Beach, Department of Psychology, Univer-

sity of Georgia; Tera R. Hurt and Kameron J. Franklin,Institute for Behavioral Research, University of Georgia;Frank D. Fincham, Family Institute, Florida State Univer-sity; Lily M. McNair, Department of Psychology, SpelmanCollege; and Scott M. Stanley, Department of Psychology,University of Denver.

This paper was supported in part by a grant from theJohn Templeton Foundation awarded to Steven R. H.Beach, Frank D. Fincham, and Lily M. McNair; and wasalso supported in part by a grant from the Administrationfor Children and Families to Frank D. Fincham. The lastauthor of this report co-owns a business that develops,refines, and sells the PREP curriculum upon which partsof the intervention tested here are based.

Correspondence concerning this article should be ad-dressed to Steven R. H. Beach, Institute for BehavioralResearch, 510 Boyd GSRC, University of Georgia, Athens,GA 30602. E-mail: srhbeach@uga.edu

Psychology of Religion and Spirituality © 2011 American Psychological Association2011, Vol. 3, No. 3, 201–216 1941-1022/11/$12.00 DOI: 10.1037/a0022207

201

enhancement efforts. Development of culturallysensitive variations of efficacious marital en-richment programs, taking into considerationracial, socioeconomic, and regional characteris-tics of the populations to be served, is timelyand may prove essential for successful programdelivery (Ooms & Wilson, 2004). For historicaland cultural reasons, the spiritual and religiouscontext is particularly pertinent for AfricanAmerican couples. A recent survey indicatesthat African Americans are markedly more re-ligious than the general population on a varietyof measures (Pew Charitable Trust [Pew], 2009;see also Chatters, Taylor, & Lincoln, 1999; Hunt& Hunt, 2001). A majority of African Americans(53%) report attending religious services once aweek (or more often) and 76% report praying atleast daily (or more often). On both of theseindices, African Americans are more religiouslyactive than persons from other racial and ethnicgroups. African Americans who are not cur-rently affiliated with a religious group reportpraying as frequently as mainline Protestants ofother ethnic groups (Pew, 2009).

Further underscoring the potential impor-tance of addressing spirituality in marital pro-grams designed for African American couples,prayer is often a preferred way of dealing withadversity among African Americans (Chatters,Taylor, Jackson, & Lincoln, 2008; Ellison &Taylor, 1996; Fowler & Hill, 2004), and spiri-tual activities, such as prayer, are a more per-vasive practice in daily life (Taylor, Chatters, &Leven, 2004). Likewise, among African Amer-icans, spirituality has been associated with feel-ings of peace, guidance, and efforts to manageadversity (Newlin, Knafl, & Melkus, 2002).Spirituality is also an important means of findinghope in the face of social, political, and economicoppression (Mattis, 2002; Newlin et al., 2002),contributing to its association with optimism(Mattis, Fontenot, & Hatcher-Kay, 2003). Thishas led several authors to call for greater attentionto spirituality in marital enhancement programsaimed at African American couples (Ooms &Wilson, 2004; Wolfinger & Wilcox, 2008).

Spirituality has been defined as a relationshipwith God or a higher power that extends beyondreligious participation and is manifested throughprivate prayer and meditation, or through seekingspiritual guidance in daily decision making (Up-church & Mueller, 2005). Because spiritualitydefined in this way has substantial cultural res-

onance for African American married couples,programs that are embedded in a culturally sen-sitive framework compatible with prayer, mayhave enhanced effectiveness due to greater en-gagement. In addition, the church is often seenas the rightful home for efforts designed toprovide marital enrichment and to strengthenfamilies (Brown, Orbuch, & Bauermeister,2008), and among African American couples,religious participation, defined as greater churchinvolvement, predicts greater relationship qual-ity and other positive family outcomes (Brody,Stoneman, Flor, & McCrary, 1994; Taylor,Mattis, & Chatters, 1999). These considerationssuggest potential advantages for marital enrich-ment programs that are disseminated within aspiritual and religious context (cf. Hurt et al.,2006; Markman et al., 2004; Stanley, Markman,St. Peters, & Leber, 1995), including the possi-bility that embedding enrichment programs in aspiritual and/or religious context may enhanceoutcomes.

Many churches are supportive of premaritalpreparation, making churches an excellentvenue for the introduction of empirically sup-ported marital preparation or marriage enhance-ment efforts that have a spiritual component(Stanley et al., 2001). One of the currentlyavailable curricula that may facilitate workingwith churches is Christian PREP (CPREP:Christian Prevention and Relationship Enhance-ment Program), developed by Stanley and col-leagues (Stanley, Trathen, McCain, & Bryan,1998). This approach influenced the develop-ment of the brief, culturally sensitive interven-tions used in the current investigation. CPREPuses the typical PREP model (Markman, Stan-ley, & Blumberg, 2001) to teach couples thePREP approach with a strong focus on com-munication, listening, and problem solving,but it also integrates relevant religious teach-ings, providing a religious context for theskills being taught. In this way, CPREP at-tempts to provide a broader meaning contextfor the couples participating in the maritalenhancement program.

To address the need for a culturally sensitiveintervention for African American couples thatdid not include prayer, that is, one that would besuitable for secular settings and for communityprograms supported by funding that precludesthe inclusion of prayer, we developed CulturallySensitive PREP (CS-PREP). To create this cur-

202 BEACH ET AL.

riculum, we used existing materials from thePREP program (Markman et al., 2001; Stanleyet al., 1998), but the curriculum was modified inall cases to include culturally sensitive compo-nents, including a focus on language, use ofAfrican American presenters, and a discussionof the impact of racism on marriage. Becausefor many African American couples, inclusionof a spiritual component should further enhancethe cultural relevance of the program, we nextdeveloped Prayer Focused PREP (PFP).

Both active couple interventions used in thisstudy were based on the PREP program, as wasthe information only control group. PREP typi-cally includes a focus on themes such as com-munication, problem-solving, enhancement offriendship, clarification of expectations, andstrengthening of commitment. These generalelements are elaborated further below. To en-hance the cultural relevance for African Amer-ican couples, we added a component that ad-dressed the impact of racism on marriage. Inaddition, we used this module to encouragepotentially supportive reactions when spousesexperience racism. Specifically, couples weretold that African American marriages are animportant bastion against the various forms ofracism encountered in everyday life. They werethen shown a videotape of a couple dealing witha racist experience using active listening andsupport to overcome the incident in a positivemanner. Group exercises were constructed to beengaging and language used in the exercises,discussions, and supplementary material wasreviewed. The resulting DVD-driven, brief in-tervention format was labeled Culturally Sensi-tive PREP (CS-PREP).

To enhance the spirituality component of theprogram, a Prayer Focused PREP (PFP) formatwas also developed. PFP included all the com-ponents of CS-PREP, and also incorporated astrong focus on prayer, with particular emphasison learning how (and why) to pray for one’smate (described below). This provided a spiri-tual framework for marital enhancement andwas designed to reinforce key processes tar-geted by the intervention including listening,enhancing satisfaction, and developing specificintentions to show love and care for the partner.Because the vast majority of African Americansself-identify as “Protestant” (78%; Pew, 2009),we utilized prayers reflective of African Amer-ican Protestant traditions in our examples.

Nonetheless, participants were from a range ofChristian denominations with potentially differ-ing theological commitments as well as fromIslamic traditions, and so efforts were made towelcome all participants regardless of their par-ticular faiths and beliefs. To do so, the programwas offered as “nonsectarian” and group leaderswere trained to provide an open, acceptingstance toward all religious orientations. It wasexplicitly stated that the program was open tothose of all religious backgrounds and theprayers were offered as “examples” only. Noproselytizing was allowed by group leaders andall group leaders agreed to this stipulation be-fore beginning the program.

Why develop the prayer examples using lan-guage and components reflective of Christiancontent? We focused on incorporating prayercontent that related to themes of “agape,” aconstruct that has a long tradition within Chris-tian theological circles. We adopted this focusbecause the theme of “agape” maps directlyonto the goal of enhancing positive couple in-teractions (Beach, Fincham, Hurt, McNair, &Stanley, 2008a; 2008b). In preliminary stages ofprogram development, we thought it might beuseful to develop generic, abstract instructionsto guide prayer activities, thereby allowing par-ticipants to craft their own specific language.However, early focus groups indicated that ge-neric instructions were off-putting for our targetaudience. Subsequently, similar observationshave been offered by experts who note that anoverly abstract presentation of prayer instruc-tions could be problematic for participants froma range of conservative religious traditions(Worthington, 2008). Because our target audi-ence was African American couples in northeastGeorgia and metropolitan Atlanta, it wasdeemed appropriate to utilize language fromlocal traditions (see Beach et al., 2008b for anexample prayer), and prayers were reviewed bylocal religious leaders and by focus groups be-fore being utilized in the program.

Accordingly, although Christian imagery andlanguage was incorporated into all prayers tocreate concrete examples of prayers that fo-cused on benefiting the partner, in all cases,couples were invited and encouraged to createtheir own prayers using language that was mostfamiliar and comfortable for them. It should benoted that any religious group with a strongtradition of prayer could replicate the current

203ENHANCING MARTIAL ENRICHMENT

approach using familiar, concrete language ap-propriate to their traditions.

Both PREP intervention conditions consistedof three, half-day sessions using a marital skills-training format. Consistent with the PREPframework, marital skills-training content(PREP; Markman et al., 2001) was a prominentpart of both programs. A typical session in-volved the couples coming together in a groupformat beginning at 8:30 a.m. After somewarm-up and icebreaking exercises, coupleswere presented with information on basic PREPconcepts via DVD. In Day 1, the focus was oncommunication training and listening skills. InDay 2, the focus was on problem solving, hid-den issues and expectations, and provision ofsupport, including support in the context ofdealing with experiences of racism. Day 3 fo-cused on fun and physical oneness and a con-tinuing review of listening skills. Basic infor-mation related to PREP was presented usingDVDs drawn from three established PREPvideo recordings: 1) Fighting for Your Mar-riage video series; 2) a program adapted fromPREP called From This Day Forward; and 3)Sweetheart’s Weekend, a weekend version ofPREP delivered by its founders to a group ofcouples in Oklahoma. Interspersed with theDVDs were opportunities to practice skills anddiscuss the information provided on DVDs.Trained facilitators were available at all ses-sions to provide feedback to the couples, helpthem with their practice, and guide discussion.Couples broke for lunch at which time men andwomen separated for gender-specific fellowshipand discussion of the morning’s lessons.

In the Prayer Focused PREP condition (PFP),spouses also received a focus on prayer; thiswas integrated with the presentation of othermaterial. In particular, in PFP, all sessions be-gan with prayer, providing an important oppor-tunity to model the kind of prayer that wasbeing taught (i.e., prayer that emphasized be-neficence toward the partner and agape). Partic-ipants were also given examples of prayers andencouraged to generate their own prayers. Inaddition to the DVD footage about communi-cation, problem solving, and couples’ activitiesmentioned above, participants were also given aconceptual framework for prayer for their part-ner and given specific encouragement to prayfor good things to happen to their partners. Allprayers were introduced as being in keeping

with the higher order goal of “helping you to bea vehicle of God’s love in your relationship.” Inaddition, participants were encouraged to prayon their partner’s behalf regarding their part-ner’s needs and aspirations (Beach et al., 2008a)and not to focus on nonconstructive themes,including retribution or “praying for God tochange my mate.” Discussion of prayer oc-curred throughout the program and prayer wasused to introduce and conclude all segments ofthe PREP instructional materials.

The control group was assigned to conditionon the basis of a block randomization schedule.Couples in the control group were assessed onthe same schedule as those in PFP group, therebycontrolling for effects of repeated measurement,maturation, individual differences, and externalsocial changes. In addition, couples in thecontrol group were provided the book, “12Hours To A Great Marriage” (Markman et al.,2004) at the conclusion of their baseline assess-ment. This popular guidebook provides positivereasons for enhancing marriage, guidelines, andexamples of communication and problem-solving strategies, exercises that could be im-plemented by individuals and couples to en-hance their relationship, and suggestions forenhancing positive aspects of the marital rela-tionship. In brief, the book provided all theskills-oriented content of PFP using the sameapproach, and couples were given written sug-gestions for reading and working through thebook, making this an information-only controlgroup. At the same time, the control groupcannot be considered an “attention control” be-cause time spent with project personnel was notequivalent between control and treatmentcouples.

Method

Design and Plan of Analyses

To examine whether change attributable to thetreatment conditions differed from change in thecontrol condition we utilized a 3 (Treatment Type:Culturally Sensitive PREP (CS-PREP) vs.Prayer-Focused PREP (PFP) versus Informa-tion-Only Control) � 2 (Sex: Husband vs.Wife) � 4 (Time of Measurement: Baseline,Post-Intervention, 6-Month Follow-up, and12-Month Follow-up) mixed-model repeated-factors analysis of variance (ANOVA), exam-

204 BEACH ET AL.

ining change across the three measurementdomains: communication, satisfaction, andpositive intentions. The three measurement do-mains were treated as correlated dimensions ofoverall marital outcome. Accordingly, eachmeasure was converted to a z-score and thensummed within assessment time and spouse tocreate a single overall index of each spouses’marital outcome at each point in time.1 Treat-ment type was analyzed as a between-subjectsfactor. The couple was the primary unit ofanalysis, and so degrees of freedom are basedon the number of couples in the analysisrather than the number of individuals. Time ofmeasurement was treated as repeated withinindividual.

Random assignment to CS-PREP, PFP, andControl was designed to oversample the inter-vention conditions on a 2:1 ratio, allowing moresensitive comparisons of response to the differ-ent intervention formats while maintaining ad-equate power to detect medium or greater effectsizes between the control group and the activetreatment conditions. According to Kazdin andBass (1989), an average sample size of N � 71participants per group is required to retainpower of .80 to detect differences between ac-tive treatment conditions assuming a mediumeffect size of the sort characteristic of publishedoutcome reports. Since our smallest group (thecontrol group) exceeded this number (N � 79)we were adequately powered to detect mediumeffect sizes or greater involving the controlgroup, and small to medium effect sizes be-tween the two active conditions.

For all significant effects, we reported partialEta-squared (�p

2), that is, the proportion of vari-ance in outcomes accounted for by the effectrelative to the error variance. Eta-squared pro-vides an index of effect size comparable to R2 inregression analyses. However, because unequalgroup sizes decreases Eta-squared, and becauseit does not provide an index of pairwise com-parisons between cells, we also reported meandifference effect sizes for key comparisons us-ing Cohen’s d.

To explicate the overall Group � Time ef-fect, we plotted marital quality over time foreach treatment group, and we examined linear,quadratic, and cubic trends in the data. We alsoexamined within-group contrasts between base-line and means at every other assessment pointto explicate significant change over time in each

group. To explicate moderation of the Group �Time effect by Gender, we examined and re-ported different patterns of Group � Time in-teraction within gender, and described in greaterdetail the significant differences between activetreatments for wives.

Plan of Assessment and Treatment ofMissing Values

The present study reports results from 393married couples randomly assigned to eitherCulturally Sensitive PREP (CS-PREP), Prayer-Focused PREP (PFP), or Control (C). Seventy-nine couples were assigned to the control con-dition and 157 couples were assigned to boththe CS-PREP and the PFP conditions. Datawere collected from husbands and wives atbaseline assessment prior to the start of inter-vention, again following the completion of theintervention program, at 6-month follow-up,and at 12- month follow-up. At all assessmentsafter baseline, there were significant amounts ofmissing data due to attrition, data from only onespouse, or missing data for that particular as-sessment. Fifteen percent of couples provideddata only at baseline, and for the remainingcouples there was an 11.5% rate of missingness.Rates of missing data did not differ across con-ditions (all chi-squared comparisons nonsignif-icant) and missingness was not correlated withbaseline scores (average correlation �.055). Inpart the missing data reflect the community-based focus of the study and difficulty in track-ing some couples after initiation of the study.Accordingly, missing data values were imputedusing the Markov Chain Monte Carlo (MCMC)framework for all couples to maintain an intentto treat framework for the analyses. The averagetime between baseline assessment and posttreat-ment assessment was approximately 10 weeksfor both intervention and control couples. Fol-low-up assessment data were collected at 6months and 12 months following baseline.

1 Results using raw scale values in a MANOVA frame-work yielded an identical pattern of significant effects.However, the results using the summed z-scores lent them-selves to easier characterization. Accordingly, we usedsummed z-scores throughout.

205ENHANCING MARTIAL ENRICHMENT

Recruitment

Based on preliminary focus group results(Hurt et al., 2006), we recruited couples throughdirect advertising, participating in community-based recruitment events, distributing brochuresto businesses, appearing on local radio shows,obtaining local press coverage, and involvinglocal pastors in attracting couples to the pro-gram. Pastors were engaged by allowing themto evaluate the program and then endorse itfrom the pulpit. Couples were told by theirpastors that there was a study of ways to en-hance the marital relationships of AfricanAmerican couples and that recruiters for thestudy were available. The recruiters were thenidentified by the pastor and couples were en-couraged to talk to the recruiters and to considervolunteering for the study. This approach led toopportunities to recruit in a variety of settingsincluding church meetings, Bible studies, wor-ship services, and other church events.

Participants

Participation in the study was open to allmarried couples in which both partners wereover 21, one or both partners were 60 years ofage or younger, and in which both partners(89.3%) or one of the partners (10.7%) self-identified as African American. The number ofcouples with only one partner who self-identified as African American is consistentwith national norms indicating that AfricanAmerican couples are more likely than couplesof other ethnic groups to be comprised of indi-viduals with different racial identification (U.S.Census Bureau, 2005). Participants who indi-cated marital crisis or a need for immediateservices at the time of recruitment were givenreferrals for community services and were notrandomized. On average, husbands entering thestudy were 40.5 years of age and wives enteringthe study were 38.85 years of age. In all cases,the couples were comfortable describing them-selves as “African American” couples. Coupleshad been married an average of 9.7 years, andreported an average of 1.65 children at home. Itwas a first marriage for 293 of the husbands andwives. Educational attainment was somewhathigher in this sample than national census com-parison data, with 40.2% of husbands reportinga bachelor’s degree or more compared to 23.3%

of all married African American adults nation-ally and 29.1% of married adults in the 35–44year age range (U.S. Census, 2006). In terms ofreligious involvement, a large percentage ofcouples in the sample reported praying for theirpartner nearly every week or more often (hus-bands 86.7%; wives 91.1%). Comparison ofgroups on demographic variables producednonsignificant multivariate analyses of variance(MANOVAs) for age, education, and number ofchildren for both husbands and wives, and anonsignificant mean difference on baseline fre-quency of prayer for the spouse for both hus-bands and wives. Spouses also did not differ bygroup at baseline assessment on any of the studyoutcome variables described below.

Group Leaders

Sixteen group leaders, all of whom were ofAfrican ancestry, were involved in leadinggroup sessions and providing feedback to cou-ples. Group leaders were selected for their en-thusiasm for marriage education and their com-fort with leading an educational program forcouples. Race of group leaders may be impor-tant in the context of programs targeting a par-ticular ethnic group, and so this potential sourceof variability in response was controlled in thecurrent investigation. In addition to having Af-rican American group leaders, a majority of thematerial in the DVD portion of the program waspresented by African American presenters. Tofacilitate couples’ access, group meetings wereheld in a range of settings with most being heldin community centers like Boys and GirlsClubs. Churches were not used as meeting areasbecause of the potential to dissuade attendanceof couples uncomfortable with a particulardenomination.

Training for the group leaders was providedby Dr. Scott Stanley, one of the founders ofPREP and Christian PREP, and by the firstauthor. All group leaders received over 20 hoursof instruction and became proficient in the useof the speaker-listener technique, delivery of thestructured intervention format, and presentationof information about prayer. Treatment fidelitywas monitored on an ongoing basis using vid-eotaped recordings of all leader interactionswith groups. Training emphasized the impor-tance of focusing on enhancing marital interac-

206 BEACH ET AL.

tion and of being accepting of all participantdenominations and religious orientations.

Treatment Fidelity

Treatment plans were highly structured, withinformation presented using a prerecordedDVD format, ensuring high treatment fidelity.In addition, the major foci of discussion andpractice were outlined in advance, and all ses-sions were video-recorded for fidelity monitor-ing purposes. For each session, two independentraters listened to all sessions, noting majorpoints of discussion and activity. In all casesgroup leaders played the appropriate DVD con-tent that supplied the bulk of the interventioninformational content. In addition, fidelity cod-ers rated the group leaders as covering, on av-erage, more than 90% of the targeted supple-mentary discussion topics in each session, withagreement between raters of r � .81, p � .001.

Outcome Measures

Communication. An eight-item measurereflecting the specific communication targets oftraining was adapted from the CommunicationSkills Test (Jenkins & Saiz, 1995). Variations ofthis measure have been used in a number ofsimilar studies, with evidence of both reliabilityand validity (e.g., Stanley et al., 2005; Stanleyet al., 2001). The measure asked questionsabout four areas: 1) summarizing, 2) taking timeouts, 3) staying focused on the problem, and 4)feeling understood. A sample question is “Mypartner tries to understand my feelings and con-cerns.” Response options ranged from 1 � Al-most Never to 7 � Almost Always. Total scorespotentially ranged from 8 to 56, with higherscores reflecting greater use of communicationskills. The internal consistency was high (coef-ficient � � .91 for husbands and .90 for wives).Accordingly, this measure provided an assessmentof an important behavioral domain in marriage.

Marital satisfaction. The six-item, Qual-ity of Marriage Index (QMI) is a unidimen-sional index that measures global perceptions ofmarital satisfaction (Norton, 1983) and has beenwidely recommended for use with communitysamples. Consistent with our use of the QMI inthe current study, it has been noted that longermeasures of marital adjustment, such as theMAT and DAS have serious conceptual prob-

lems that limit their interpretation (Fincham,Beach, & Kemp-Fincham, 1997). These limita-tions may be particularly problematic for stud-ies of community couples for whom measuresof clinical dysfunction may be less appropriateand/or insensitive to change (Fincham et al.,1997). In addition, couple evaluations of theirrelationship have long been held to be the finalcommon pathway for marital intervention ef-fects (Jacobson, 1985), making direct assess-ment of evaluation of the relationship an impor-tant assessment domain. On the QMI, lowscores indicate a more negative evaluation. Asample item is, “Our relationship is strong.”Response options ranged from 1 � StronglyDisagree to 5 � Strongly Agree. Total scorespotentially ranged from 6 to 30, with higherscores reflecting a more positive evaluation of therelationship. Internal consistency was high (coef-ficient alpha for husbands � .95; for wives � .96).Accordingly, this measure assessed an importantevaluative domain in marriage.

Positive intentions. The Positive MaritalIntentions (PMI) scale was developed based onFincham and Beach’s (1999) conceptualizationof relationship conflict within a “goal frame-work.” The five-item measure was designed toassess intentions to engage in positive behaviortoward the spouse. The measure assessed bothgeneral intentions as well as implemental inten-tions to engage in specific positive behaviortoward the partner. A sample item is “I look foropportunities to do things that will show love tomy mate.” Response options ranged from 1 � Ialmost never do this to 5 � I do this every dayor more than once a day. Total scores poten-tially ranged from 5 to 25, with higher scoresreflecting greater time spent thinking about pos-itive activities and outcomes for one’s mate andplanning or looking for opportunities to imple-ment plans. Internal consistency was high in thecurrent sample (coefficient alpha for husbands �.87; for wives � .87). Accordingly, this measureprovided an assessment of an important cog-nitive/motivational domain in marriage.

For all analyses reported below, each of themeasures was converted to a z-score reflectingthe respondent’s relative position within theoverall distribution of scores. After renderingall scores commensurable, they were summedacross the three measures, within each point intime, to create an overall index of marital out-comes for that participant at that point in time.1

207ENHANCING MARTIAL ENRICHMENT

Results

Data were analyzed using a mixed-designANOVA with a between-subjects factor ofgroup and within-subjects factors for time andsex of spouse, yielding a 3(Group) � 4(Time ofassessment) � 2 (Sex) design. Missing data wascommon and so to avoid loss of couples withpartial data and maintain our intent to treatsample, all missing data was imputed usingMCMC (Schafer, 1997), a Markov Chain esti-mation procedure with high efficiency. Meansand standard deviations for overall outcome ateach time point are provided separately by genderand treatment group for each assessment time inTable 1. The correlations between the marital out-come measures at premarital assessment werer(393) � .61, p � .001 for QMI and Communi-cation; r(393) � .51, p � .001 for QMI andPositive Intentions; and r(393) � .46, p � .001 forCommunication and Positive Intentions.

Overall Intervention Effects

Mauchly‘s test indicated that the assumptionof sphericity had been violated, �2(5) � 21.2,p � .001, therefore degrees of freedom werecorrected using Greenhouse-Geisser estimatesof sphericity (ε � .96). There was a main effectof time F(2.89,1127) � 8.17, p � .001, partialEta squared (�p

2) � .021, indicating improve-ment in marital outcomes over time. In addition,there was a significant effect of sex F(1,390) � 6.58, p � .05, �p

2 � .017, indicating thathusbands scored higher on measures of maritaloutcome on average. The anticipated group bytime interaction was also significant F(5.78,2254) � 2.16, p � .05, �p

2 � .011, reflectingsignificantly greater change, for the interventiongroups than the control group. These patterns areexplicated in Figure 1, which provides a compar-ison of change across time for couples in eachgroup.

As can be seen in Figure 1, there was nochange overall for those in the control condi-tion, but a comparable level of overall changefor those couples assigned to the two treatmentconditions. These observations were confirmedby within-Group ANOVAs which showed asignificant effect of time for those in the twoactive treatment conditions: F(2.85, 445) � 5.81,p � .001 and F(2.88, 448) � 4.55, p � .005 forhusbands in PFP and CS-PREP respectively,

and F(2.82, 439) � 11.39, p � .001 and F(2.85,445) � 3.46, p � .05 for wives in PFP andCS-PREP respectively. Conversely, there was anonsignificant effect of Time for both husbandsand wives in the control condition F(2.88,224.9) � .34, ns and F(2.5, 195) � .19, ns forhusbands and wives respectively.2

To better characterize the reliable compo-nents of change reflected in the overall signifi-cant Group � Time interaction, we also exam-ined linear, quadratic, and cubic trends. Only dif-ferences in the linear component of the Group �Time interaction were found to be reliable F(2,780) � 2.99, p � .05. Accordingly, we plottedonly the linear component of change in Figure 1.

Interaction With Sex

The significant group by time interaction wasqualified by a marginal interaction with sex,resulting in a marginally significant three-wayinteraction F(5.85, 1140) � 2.09, .05 � p �.06, �p

2 � .011, reflecting no significant im-provement for husbands or wives in the controlcondition, but somewhat different patterns ofimprovement over time for husbands and wivesin response to active intervention. Because therewas no reliable change as a function of time inthe control group, we removed the controlgroup to better examine sex differences in re-

2 For husbands in both treatment conditions only thelinear component of change was significant. For wives inPFP, linear, quadratic, and cubic components were all sig-nificant, but for wives in CS-PREP only the linear compo-nent of change was significant. Paired t tests between eachassessment and baseline assessment indicated no signif-icantly different means in the control group for eitherhusbands or wives for any assessment time. Conversely,for husbands and wives in PFP, all comparisons weresignificant, husbands pretreatment versus posttreatment,t(156) � 2.29, p � .05; pretreatment versus 6-monthfollow-up, t(156) � 3.18, p � .005; pretreatment versus12-month follow-up, t(156) � 3.43, p � .001; wives pretreat-ment versus posttreatment t(156) � 5.48, p � .001; pretreat-ment versus 6-month follow-up, t(156) � 3.48, p � .001;pretreatment versus 12-month follow-up, t(156) � 4.90,p � .001. For husbands and wives in CS-PREP all contrastswere significant except for pretreatment versus posttreat-ment for wives. Husbands pretreatment versus posttreat-ment, t(156) � 1.98, p � .05; pretreatment versus 6-monthfollow-up, t(156) � 2.09, p � .05; pretreatment versus12-month follow-up, t(156) � 3.67, p � .001; wives pre-treatment versus posttreatment t(156) � 1.03, NS, pretreat-ment versus 6-month follow-up, t(156) � 2.99, p � .01;pretreatment versus 12-month follow-up, t(156) � 1.97,p � .05.

208 BEACH ET AL.

sponse to the two active treatments. This re-sulted in a significant three-way interaction ofsex, condition, and time F(2.93, 915) � 3.59,p � .05, indicating significant gender differ-ences in response to the two treatments. How-ever, examination of linear, quadratic, and cubictrends indicated that only the cubic component

of this effect was significant F(1, 312) � 11.98,p � .001.

To further explicate the interaction ofGroup � Time � Sex, we compared responseto CS-PREP and PFP separately for husbandsand wives. We used two, 2(Group: CS-PREPvs. PFP) � 4 (Time) repeated-measures

Figure 1. Level of standardized outcome at baseline, postintervention, 6- month and 12month follow-up averaged across husbands and wives for PFP, CS PREP, and Control group.

Table 1Standardized Outcome Scores by Condition for Husbands and Wives at All Assessment Periods

Condition Control PFP CS-PREP

Spouse Husbanda Wifeb Husbandc Wifed Husbande Wifef

Time of AssessmentBaseline �.145 (.255) �.120 (.271) �.195 (.207) �.552 (.262) �.052 (.200) �.471 (.208)Postintervention �.115 (.252) �.258 (.273) .158b (.184) .199b (.188) .246b (.196) �.325 (.190)Six-month �.009 (.266) �.259 (.296) .356b (.203) .021b (.197) .277b (.185) �.013b (.190)Twelve-month .017 (.262) �.201 (.309) .368b (.179) .205b (.188) .473b (.185) �.148b (.214)

Note. Standard errors for each cell are in parentheses. Cells within column with a subscript b are significantly differentfrom the baseline mean using paired t-tests.a Effect of time for husbands in the control group is not significant. b Effect of time for wives in the control group is notsignificant. cEffect of time for husband in PFP is significant, as is the linear trend. d Effect of time for wife in PFP issignificant, as is the linear trend. e Effect of time for husband in CS-PREP is significant, as is the linear trend. f Effectof time for wives in CS-PREP is significant, as is the linear trend.

209ENHANCING MARTIAL ENRICHMENT

ANOVAs to explicate the different patterns ofchange in response to intervention displayed byhusbands and wives. For husbands, there was asignificant effect of time F(2.88, 898) � 10.01,p � .001, �p

2 � .031, indicating a positiveresponse to intervention across conditions, andonly the linear component of change was sig-nificant, F(1, 312) � 25.3, p � .001. There wasno significant interaction of group with time forhusbands F(2.88, 898) � .435, ns, suggestingthat husbands experienced very similar patternsof change in response to both intervention for-mats over the time period studied. For wives,there was also a significant effect of timeF(2.86, 894) � 11.28, p � .001, �p

2 � .035,indicating a positive response to interventionacross conditions. However, there was also asignificant interaction of group with timeF(2.86, 894) � 3.42, p � .05, �p

2 � .011.The interaction of group with time for wives

is plotted in Figure 2. The figure shows similarstarting points but differing patterns in out-comes over time for wives assigned to eachintervention. Linear and quadratic componentsof the effect of time were significant for bothintervention groups but did not differ betweengroups. Only the cubic component of the effectwas significantly different for wives in the twotreatment groups. The significant cubic effect isthe result of significantly greater change from

baseline to posttest in PFP than in CS-PREP forwives, F(1, 312) � 9.44, p � .005, �p

2 � .029;no difference in change from baseline to6-month follow-up, F(1, 312) � .261, ns; and amarginally significant difference in changefrom baseline to 12-month follow-up for wivesin the two active treatment conditions F(1,312) � 3.72, p � .055, �p

2 � .012. The signif-icantly greater change for wives in PFP com-pared to those in CS-PREP from baseline topost is illustrated in Figure 2. For comparisonpurposes, overall outcome is also plotted for6-month and 12-month follow-ups for wives inboth treatment groups.

End Point Comparisons

Finally, to characterize difference in groupmeans at the 12-month follow-up assessmentwe computed Cohen’s d, comparing activetreatment conditions to the control condition.As would be expected given the patterns ofchange, Cohen’s d for the two husband compar-isons were similar in magnitude. For husbandsin PFP, Cohen’s d was 1.43 and for husbands inCS-PREP, Cohen’s d was 2.01. Using conven-tional terminology (Cohen, 1969), both effectswould be considered “large.” Conversely, forwives in PFP compared to the control group,Cohen’s d was also large, d � 1.58, whereas for

Figure 2. Differential change for pretreatment to posttreatment assessment for wives as-signed to either PFP or CS-PREP.

210 BEACH ET AL.

wives in CS-PREP compared to the controlgroup, Cohen’s d was small, d � .20.

Discussion

The current study is the first to examine theefficacy of brief PREP formats, modified to beculturally sensitive for African American cou-ples. The results suggest that the PREP pro-gram, delivered in a spiritual context with afocus on prayer, is as efficacious as a programfocused on traditional PREP material, presentedin a culturally sensitive framework, and may bemore efficacious for wives, particularly in termsof its immediate effects from baseline to post-test. In addition, both culturally sensitive ver-sions were more efficacious than the controlcondition in producing positive change frombaseline to postintervention and to 1-year fol-low-up. Participants indicated a high level ofsatisfaction with both formats and with the in-

clusion of a focus on spirituality and prayer.Likewise, the response of community leaders tothe overall ProSAAM program has been posi-tive (Hurt et al., 2006). In addition, because theprogram is an extension of PREP, and is pro-vided in a highly structured, self-contained pro-gram format, it has the potential to be dissem-inated widely with high fidelity either throughcommercially supported programming or bycharitable organizations and churches, and theprogram has the potential to be implemented ingroups of various sizes depending on localneeds. It should be noted, however, that thecurrent results do not indicate that culturallysensitive PREP is superior to standard formatPREP in terms of impact on couple functioningof African American couples because standardformat PREP was not one of the comparisonconditions. However, as suggested in earlierwriting (Beach et al., 2008a), even when out-

Figure 3. Consort flow chart.

211ENHANCING MARTIAL ENRICHMENT

comes are similar it may be useful to havealternative approaches that can be utilized withappropriate populations and this may have ben-efits in terms of community participation inmarital enrichment efforts.

The effect size of the intervention was char-acterized in several ways. However, the mosteasily interpretable metric is provided by Co-hen’s d at 12-month follow-up. Cohen’s d is themost widely used measure of standardized meandifference between two groups, and becauseCohen’s d takes into account the variance of thedependent variable, it provides informationabout effects that are comparable between dif-ferent indices of change. Typically, an effectsize of 0.2 is considered “small” whereas aneffect size greater than .8 is considered “large,”For comparison purposes, a large effect wouldbe comparable to the average difference inheights of 13-year-old versus 18-year-old girls(Cohen, 1969, p. 23). At 12-month assessment,both husbands and wives in both active inter-vention conditions showed significant improve-ment from baseline scores, and this was not truefor those in the control condition. Effect sizesfor husbands and wives in PFP were large andcomparable (Cohen’s d � 1.43 and 1.58 forhusbands and wives respectively), but were di-vergent for husbands and wives in CS-PREP(Cohen’s d � 2.01 and .20 for husbands andwives respectively). This result suggests thatwhereas more than 90% of husbands and wivesin the control group at 12-month follow-upscored below those in PFP, and below husbandsin CS-PREP, only 58% of wives in the controlgroup would score below wives in CS-PREP at12-month follow-up.

The differential pattern of response of wivesto the two intervention formats is of interest andsuggests the potential value of future work ex-plicating the role that spirituality may play inthe process of change in marital enrichmentprograms, particularly for women. Whereashusbands showed similar response to interven-tion regardless of format, wives showed en-hanced early response to the inclusion of prayerand spiritual themes. To the extent that thisdifference affects long-term couple outcomes, itmay have implications for maintenance of gainsfor men as well as women at longer-term fol-low-ups. In addition, if the two formats arecomparable in effect for men, but have differenteffects for women, couple-based formats may

need to attend to the characteristics that forecastbetter response among wives.

The presence of different patterns of responseover time to PFP and CS-PREP for wives sug-gests the possibility of somewhat different ac-tive ingredients or processes of change in thetwo conditions. In particular, it is noteworthythat the PFP condition compared to the CS-PREP condition produced significantly greaterinitial change (i.e., baseline to post), with vari-ability across the follow-up, but relatively littlenet additional improvement across the remain-der of the follow-up period. Conversely, theCS-PREP condition showed relatively modestinitial change, and then continuing modest im-provements across the follow-up period. This in-teresting difference in pattern of change for wivessuggests the possibility of underlying differencesin process of change between the two conditions.

Several possibilities are consistent with theobserved patterns and are worthy of continuedexploration. First, PFP might have produced aninitial additive effect on early response to inter-vention. This could result, for example, if PFPand CS-PREP had similar effects on change inmarital skills, but PFP had a greater impact onthe extent to which couples saw their relation-ship as having a spiritual and church-basedfoundation. In this case, PFP might have anextra source of impact on satisfaction that couldexert its effect rapidly. Alternatively, the PFPcondition might provide added benefits primarilyby enhancing initial positive expectations forchange or mobilization of hope, particularly forwives. If so, this would also be expected to lead toa pattern of rapid initial gains, although one wouldexpect such effects to diminish over time witheventual convergence with changes produced inCS-PREP. Both these conjectures are consistent inbroad brush with the observed outcomes. Accord-ingly, more detailed processes analyses are in or-der to help further refine the use of these programsin the future.

It will also be useful to examine other pro-cesses that have been shown to be influenced byprayer and that may be relevant to long-termrelationship functioning such as changes inforgiveness (Lambert, Fincham, Stillman,Graham, & Beach, 2010), infidelity (Fincham,Lambert, & Beach, 2010), and gratitude (Lam-bert, Fincham, Braithwaite, Graham, & Beach,2009). Such changes may or may not be relatedto initial changes in satisfaction and marital

212 BEACH ET AL.

interaction, but may nonetheless exert an influ-ence on outcomes over time, or influence resil-ience in response to relationship problems orexternal stressors. Accordingly, these potentialfacets of marital change also deserve attentionat longer-term follow-ups.

Are Effects Limited to PREP Derivatives?

It should not be assumed that the addition ofprayer and spirituality or other culturally sensi-tive adaptations can only work in combinationwith PREP or similar skills-based programs.The current results suggest the addition of spir-ituality and prayer for the partner may be suit-able for a range of marital enhancement pro-grams and allow these programs to be moreculturally sensitive for religiously committedcouples without loss of program effectiveness.Because the prayers are presented by the groupfacilitators as being suggestions only, partici-pants were given strong encouragement to makethe prayers their own and to incorporate thecontent in their own way. This allowed individ-uals with a range of religious commitments toparticipate in the program without feeling pres-sured or proselytized. This is likely to be an-other important consideration for community-based intervention that incorporates a spiritualdimension (see Beach et al., 2008b; Marks,2008; Sullivan & Karney, 2008; Worthington,2008). In addition, the PFP format should allowa wide range of African American religiousgroups to endorse the program, potentially in-creasing community engagement (Hurt et al.,2006).

Limitations

There are several limitations to the currentstudy. In particular, it will be important to di-rectly compare CS-PREP and PFP to PREP inits usual format in order to identify whether theyare best thought of as alternative, equivalent,formats for marital enrichment, or whether ele-ments of these programs, particularly the inclu-sion of spirituality, should be considered “addi-tive” in its impact on some dimensions. Further,the current report provides only self-reportedchange across a range of facets of the maritalrelationship, and in future work it will be im-portant to examine impact on observationalmeasures to confirm that change occurs in areas

not dependent on self-report. Similarly, the cur-rent study focused on change in three areas,overall positive evaluation of the relationship,communication and listening, and positive in-tentions toward the partner and relationship, allreflecting a positive focus on relationship out-comes. This positive focus is quite appropriatein the context of relationship enhancement withcommunity couples. However, over longer timeframes, a broader measurement strategy will beimportant, including measures that may bettercapture the emergence of dysfunction, and assess-ments of negative outcomes such as divorce/separation and problems of intimate partner vio-lence. Finally, because the study focused on Afri-can American couples only, it is not yet possibleto tell whether similar modifications would pro-vide similarly positive outcomes for equally reli-giously committed couples of other ethnic groups.

There were also several strengths and severalpossible limitations related to our decision touse only African American group leaders. Mostnotably, because all group leaders were of Africanancestry, we were not able to directly examine ourassumption that ethnicity of the group leader is animportant consideration in providing culturallysensitive adaptations of marital enrichment pro-grams. Likewise we were not able to assessdirectly the impact of knowledge regarding cul-turally specific patterns of communication andinteraction. It is possible that behavioral char-acteristics such as warmth, humor, and positiveengagement with couples are more importantthan shared ethnicity in predicting positive re-sponse. Likewise, we do not have detailed codingof group leader feedback to particular participants,limiting our ability to examine the dynamics ofinteractions between couples and group leaders. Inaddition, participants reported valuing the infor-mal interaction that occurred around shared meals.Again, this may be an important focus for futurework on the process of effective dissemination ofmarital enrichment programs.

Practical and Ethical Issues in the Useof Prayer

It is important to consider ethical issues regard-ing the incorporation of prayer into relationshipenhancement programs. One potential concern isthat some people may incorporate problematicstyles of relating into their prayers, raising thequestion of how one might address such problems

213ENHANCING MARTIAL ENRICHMENT

without inappropriately infringing on the religiousbelief systems of spouses. Another question ishow one might respond to and be inclusive ofparticipants whose beliefs or prayer practices arevery different from others in the group. We haveaddressed a range of similar issues more abstractlyelsewhere (Beach et al., 2008a; 2008b). However,it is instructive to examine these issues in thecontext of the current study.

With regard to issues of style and content ofprayers, we found it useful to highlight thevalue of prayers for the spouse that included afocus on themes of beneficence and agape andthat avoided themes of retribution and change.This was not viewed as intrusive with regard toreligious belief because it was narrowly focusedon the expectation that some types of prayers(and not others) were most strongly related topositive couple outcomes (see Fincham et al.,2010; Lambert et al., 2009). Accordingly, groupleaders were encouraged to function only intheir role as providers of advice regarding mar-ital enhancement, and not to adopt the role ofreligious leader.

Although we had no reported instances ofpeople with beliefs so different than others inthe group that they were viewed with suspicionor concern, given the open group structure, thisis a possibility we anticipated. In part, our suc-cess in this regard may have resulted from ourapproach of having group leaders present them-selves as experts on marital enhancement, noton matters of religious belief. Accordingly,group leaders modeled an accepting, open atti-tude toward all participants’ religious beliefs forthe purposes of the group meetings, and noproselytizing or effort to influence the religiousbeliefs of others in the group was allowed. Thismay have limited opportunity for theologicaldisputes, particularly given the relatively briefformat we utilized. Although it did not provenecessary, we had anticipated consulting withpastors of the same denomination if we encoun-tered rituals or patterns of prayer that seemedincompatible with the marital enhancement fo-cus of the program. At least within the briefframework of the current program, couplesreadily accepted the suggested focus on benef-icence and agape, and the utility of prayer di-rected to the spouse’s welfare and benefit. Con-sequently, doctrinal disputes did not pose aproblem for participants or group leaders, and it

did not prove necessary to consult with pastorson issues of doctrine or religious ritual.

Conclusions

Despite the widespread use of prayer(McCullough & Larson, 1999) and its professedinfluence in people’s lives (McCaffrey & Eisen-berg, 2004), prayer has received remarkablylittle attention from marital researchers. Thisrelative lack of attention may be a particularlyimportant obstacle when marital and familytherapists attempt to work with some currentlyunderserved populations. In our sample, thevast majority of participants were already pray-ing regularly for their spouse in some form,suggesting that the spiritual dimension of theirrelationship was already salient to them. Ourwork conducting marital enrichment programsfor African American couples in Georgia, andthe reactions we have received from the com-munities we serve, suggest that inclusion ofprayer may be helpful in enhancing communityengagement in some areas and with somegroups (Hurt et al., 2006), and this is clearly atopic that deserves continued attention in thecontext of marital enhancement programs moregenerally (Wolfinger & Wilcox, 2008).

References

Acitelli, L. K., Douvan, E., & Veroff, J. (1997).The changing influence of interpersonal percep-tions on marital well-being among Black andWhite couples. Journal of Social and PersonalRelationships, 14, 291–304. doi:10.1177/0265407597143001

Adelmann, P. K., Chadwick, K., & Baerger, D. R.(1996). Marital quality of Black and White adultsover the life course. Journal of Social and Per-sonal Relationships, 13, 361–384. doi:10.1177/0265407596133004

Beach, S. R. H., Fincham, F. D., Hurt, T. R., McNair,L. M., & Stanley, S. M. (2008a). Prayer and mar-ital intervention: A conceptual framework. Journalof Social and Clinical Psychology, 27, 641–669.doi:10.1521/jscp.2008.27.7.641

Beach, S. R. H., Fincham, F. D., Hurt, T. R., McNair,L. M., & Stanley, S. M. (2008b). Prayer and mar-ital intervention: Toward an open, empirically-grounded dialogue. Journal of Social and ClinicalPsychology, 27, 693–710. doi:10.1521/jscp.2008.27.7.693

Brody, G. H., Stoneman, Z., Flor, D., & McCrary, C.(1994). Religion’s role in organizing family rela-

214 BEACH ET AL.

tionships: Family process in rural, two-parent Af-rican American families. Journal of Marriage &the Family, 56, 878–888. doi:10.2307/353600

Brown, E., Orbuch, T. L., & Bauermeister, J. A.(2008). Religiosity and marital stability amongBlack and White American couples. Family Rela-tions, 57, 186 –197. doi:10.1111/j.1741–3729.2008.00493.x

Bryant, C. M., Taylor, R., Lincoln, K. D., Chatters,L. M., & Jackson, J. S. (2008). Marital satisfactionamong African Americans and Black Caribbeans:Findings from the national survey of Americanlife. Family Relations, 57, 239–253. doi:10.1111/j.1741–3729.2008.00497.x

Chatters, L. M., Taylor, R. J., Jackson, J. S., &Lincoln, K. D. (2008). Religious coping amongAfrican Americans, Caribbean Blacks, and Non-Hispanic Whites. Journal of Community Psychol-ogy, 36, 371–387.

Chatters, L. M., Taylor, R. J., & Lincoln, K. D.(1999). African American religious participation:A multi-sample comparison. Journal for the Sci-entific Study of Religion, 38, 132–145. doi:10.2307/1387589

Cohen, J. (1969). Statistical power analysis for thebehavioral sciences. New York: Academic Press.

Ellison, C. G., & Taylor, R. J. (1996). Turning toprayer: Social and situational antecedents of reli-gious coping among African Americans. Review ofReligious Research, 38, 111–132.

Fincham, F. D., Beach, S. R., & Kemp-Fincham, S. I.(1997). Marital quality: A new theoretical perspec-tive. In R. J. Sternberg & M. Hojjat (Eds.), Satis-faction in close relationships (pp. 275–304). NewYork: Guilford Press.

Fincham, F. D., & Beach, S. R. H. (1999). Conflict inmarriage. Annual Review of Psychology, 50, 47–77. doi:10.1146/annurev.psych.50.1.47

Fincham, F. D., Lambert, N. M., & Beach, S. R. H.(2010). Faith and unfaithfulness: Can praying foryour partner reduce infidelity? Journal of Person-ality and Social Psychology.

Fowler, D. N., & Hill, H. M. (2004). Social supportand spirituality as culturally relevant factors incoping among African American women survivorsof partner abuse. Violence against Women, 10,1267–1282. doi:10.1177/1077801204269001

Hunt, L. L., & Hunt, M. O. (2001). Race, religion,and religious involvement: A comparative study ofWhites and African Americans. Social Forces, 80,605–631. doi:10.1353/sof.2001.0098

Hurt, T. R., Franklin, K. J., Beach, S. R. H., Murry,V. B., Brody, G. H., McNair, L. D., & Fincham,F. D. (2006). Dissemination of couples interven-tions among African American populations: Expe-riences from ProSAAM. Couples Research &Therapy Newsletter, 12, 13–16.

Jacobson, N. S. (1985). The role of observation mea-sures in marital therapy outcome research. Behav-ioral Assessment, 7, 287–308.

Jenkins, N., & Saiz, C. C. (1995). The communica-tion skills test. Unpublished manuscript, Univer-sity of Denver, Denver, CO.

Johnson, C. A., Stanley, S. M., Glenn, N. D., Amato,P. A., Nock, S. L., Markman, H. J., & Dion, M. R.(2002). Marriage in Oklahoma: 2001 baselinestatewide survey on marriage and divorce (S02096OKDHS). Oklahoma City, OK: Oklahoma Depart-ment of Human Services.

Karney, B. R., Garvan, C., & Thomas, M. (2003).Family formation in Florida: 2003 Baseline surveyof attitudes, beliefs, and demographics relating tomarriage and family formation. Gainesville, FL:University of Florida.

Kazdin, A. E., & Bass, D. (1989). Power to detectdifferences between alternative treatments in com-parative psychotherapy outcome research. Journalof Consulting and Clinical Psychology, 57, 138–147. doi:10.1037/0022-006X.57.1.138

Lambert, N. M., Fincham, F. D., Stillman, T. F.,Graham, S. M., & Beach, S. R. M. (2010). moti-vating change in relationships: Can prayer increaseforgiveness? Psychological Science, 21, 126–132.

Lambert, N. M., Fincham, F. D., Braithwaite, S. R.,Graham, S. M., & Beach, S. R. H. (2009). Canprayer increase gratitude? Psychology of Religionand Spirituality, 1, 39–49. doi:10.1037/a0016731

Markman, H. J., Stanley, S. M., & Blumberg, S. L.(2001). Fighting for your marriage. San Francisco:Jossey-Bass.

Markman, H. J., Stanley, S. M., Blumberg, S. L.,Jenkins, N. H., & Whiteley, C. (2004). 12 hours toa great marriage. San Francisco: Jossey-Bass.

Marks, L. (2008). Prayer and marital intervention:Asking for divine help or professional trouble?Journal of Social and Clinical Psychology, 27,678–685. doi:10.1521/jscp.2008.27.7.678

Mattis, J. S. (2002). Religion and spirituality in themeaning making and coping experiences of Afri-can American women: A qualitative analysis. Psy-chology of Women Quarterly, 26, 309–321.

Mattis, J. S., Fontenot, D. L., & Hatcher-Kay, C. A.(2003). Religiosity, racism, and dispositional opti-mism among African Americans. Personality &Individual Differences, 34, 1025. doi:10.1016/S0191-8869(02)00087–9

McCaffrey, A. M., & Eisenberg, D. M. (2004).Prayer for health concerns: Results of a nationalsurvey of prevalence and patterns of use. Archivesof Internal medicine, 164, 858–862.

McCullough, M. E., & Larson, D. B. (1999). Prayer.In W. R. Miller (Ed.), Integrating spirituality intotreatment: Resources for practitioners (pp. 85–110). Washington, DC: American PsychologicalAssociation.

215ENHANCING MARTIAL ENRICHMENT

Murry, V. M., Kotchick, B. A., Wallace, S., Ketchen,B., Eddings, K., Heller, L., & Collier, I. (2004).Race, culture, and ethnicity: Implications for acommunity intervention. Journal of Child & Fam-ily Studies, 13, 81–100. doi:10.1023/B:JCFS.0000010492.70526.7d

Newlin, K., Knafl, K., & Melkus, G. D. E. (2002).African-American spirituality: A concept analysis.Advances in Nursing Science, 25, 57–70.

Norton, R. (1983). Measuring marital quality: A crit-ical look at the dependent variable. Journal ofMarriage and the Family, 45, 141–151.

Ooms, T. J., & Wilson, P. C. (2004). The challengesof offering couples and marriage education to lowincome couples. Family Relations, 53, 440–446.

Pew Charitable Trust. (2009). A religious portrait ofAfrican-Americans. Retrieved from http://pewforum.org/docs/?DocID�389

Roebuck, B. J., & Brown, S. (2007). Race-ethnicdifferences in marital quality and divorce. SocialService Review, 36, 945–967. doi.org/10.1016/j.ssresearch.2006.04.001

Schafer, J. L. (1997). Analysis of incomplete multi-variate data. New York: Chapman and Hall.

Stanley, S. M., Allen, E. S., Markman, H. J., Saiz,C. C., Bloomstrom, G., Thomas, R., . . . Baily,A. E. (2005). Dissemination and evaluation ofmarriage education in the Army. Family Process, 44,187–201. doi:10.1111/j.1545–5300.2005.00053.x

Stanley, S. M., Amato, P. R., Johnson, C. A., &Markman, H. J. (2006). Premarital education, mar-ital quality, and marital stability: Findings from alarge, random, household survey. Journal of Fam-ily Psychology, 20, 117–26. doi: 10.1037/0893–3200.20.1.117

Stanley, S. M., Markman, H. J., Prado, L. M., Olmos-Gallo, P. A., Tonelli, L., St. Peters, M, . . . Whit-ton, S. (2001). Community based premarital pre-vention: Clergy and lay leaders on the front lines.Family Relations: Journal of Applied Family &Child Studies, 50, 67–76. doi: 10.1111/j.1741–3729.2001.00067.x

Stanley, S. M., Markman, H. J., St. Peters, M., &Leber, B. D. (1995). Strengthening marriages andpreventing divorce: New directions in preventionresearch. Family Relations: Journal of Applied

Family & Child Studies, 44, 392– 401. doi:10.2307/584995

Stanley, S. M., Trathen, D., McCain, S., & Bryan, M.(1998). A lasting promise: A Christian guide tofighting for your marriage. San Francisco, CA:Jossey-Bass.

Sullivan, K. T., & Karney, B. J. (2008). Incorporatingreligious practice in marital interventions: To prayor not to pray? Journal of Social and ClinicalPsychology, 27, 670 – 677. doi:10.1521/jscp.2008.27.7.670

Taylor, R. J., Chatters, L. M., & Leven, T. S. (2004).Religion in the lives of African Americans. Thou-sand Oaks, CA: Sage.

Taylor, R. J., Mattis, J., & Chatters, L. M. (1999).Subjective religiosity among African Americans:A synthesis of findings from five national samples.Journal of Black Psychology, 25, 524–543. doi:10.1177/0095798499025004004

U.S. Census Bureau. (2005, June 29). America’s fam-ilies and living arrangements: 2004. Retrievedfrom http://www.census.gov/population/www/socdemo/hh-fam/cps2004.html

U.S. Census Bureau. (2006). Educational attainmentin the United States: 2006. Black alone Table3–09. Educational attainment of the population 15years and over, by marital status, age, sex, race,and Hispanic origin: 2006. Downloaded fromhttp://www.census.gov/population/socdemo/education/cps2006/tab03– 09.xls

Upchurch, S., & Mueller, W. H. (2005). Spiritualinfluences on ability to engage in self-care ac-tivities among older African Americans. Inter-national Journal of Aging & Human Develop-ment, 60, 77–94.

Wolfinger, N. H., & Wilcox, W. B. (2008). Happilyever after? Religion, marital status, gender, andrelationship quality in urban families. SocialForces, 86, 1311–1337.

Worthington, E. (2008). Can prayer be long andstrong enough to matter? Journal of Social andClinical Psychology, 27, 686–692. doi:10.1521/jscp.2008.27.7.686

Received October 24, 2009Revision received May 25, 2010

Accepted October 12, 2010 �

216 BEACH ET AL.