Patterns of Marital Relationship Change Across the Transition FromOne Child to Two

Brenda L. VollingUniversity of Michigan

Wonjung OhTexas Tech University

Richard Gonzalez and Patty X. KuoUniversity of Michigan

Tianyi YuUniversity of Georgia

Patterns of marital change after the birth of a second child were explored in a sampleof 229 married couples, starting in pregnancy, and at 1, 4, 8 and 12 months postpartum.Five trajectory patterns that reflected sudden, persistent decline (i.e., crisis), sudden,short-term decline (i.e., adjustment and adaptation), sudden, short-term gain (i.e.,honeymoon effect), linear change, and no change were examined with dyadic, longi-tudinal data for husbands and wives. Six distinct latent classes emerged using growthmixture modeling: (a) wife decreasing positivity–husband honeymoon (44%), (b) wifeincreasing negativity–husband adjustment and adaptation (34.5%), (c) wife honey-moon–discrepant marital positivity (7.4%), (d) wife adjustment and adaptation (6.9%),(e) couple honeymoon with discrepant positivity and negativity (5.2%) and (f) husbandadjustment and adaptation (1.7%). Classes were distinguished by individual vulnera-bilities (i.e., depression, personality), stresses associated with the transition (i.e.,unplanned pregnancy), and adaptive processes (i.e., marital communication, socialsupport). Marital communication, parental depression, and social support emerged asimportant targets for intervention that can assist parents planning to have additionalchildren.

Keywords: family stress, marital adjustment, second-child, siblings, transition to parenthood

Normative transitions are common experi-ences across the family life cycle, and the tran-sition to parenthood is clearly one that requiresconsiderable adjustment and adaptation as part-ners become parents (Cowan & Cowan, 2000).

Significant declines in marital satisfaction havebeen noted after the birth of a first child (Mit-nick, Heyman & Smith Slep, 2009), and there issignificant variation in marital change trajecto-ries reflecting decline, growth, and continuity(Belsky & Rovine, 1990; Don & Mickelson,2014; Doss, Rhoades, Stanley, & Markman,2009). No study has examined variation in mar-ital trajectories after the birth of a second child,even though marital quality continues to declinewith the birth of each additional child (Twenge,Campbell, & Foster, 2003). Several studieshave examined differences in marital quality forprimiparous and multiparous couples, withsome finding greater declines in marital satis-faction and relationship quality for multiparouscouples (Belsky, Spanier, & Rovine, 1983;Krieg, 2007; Lindblom et al., 2014), whereasothers report equivalent declines across the twogroups (O’Brien & Peyton, 2002; Wilkinson,1995). The current study offered a unique op-

Brenda L. Volling, Department of Psychology, Univer-sity of Michigan; Wonjung Oh, Department of HumanDevelopment and Family Studies, Texas Tech University;Richard Gonzalez and Patty X. Kuo, Department of Psy-chology, University of Michigan; Tianyi Yu, Center forFamily Research, University of Georgia.

This research was supported by grants from the EuniceKennedy Shriver National Institute of Child Health andHuman Development (R01HD042607, K02HD047423) toBrenda L. Volling. We are grateful to the families of theFamily Transitions Study for their participation in this re-search program.

Correspondence concerning this article should be ad-dressed to Brenda L. Volling, Department of Psychology,University of Michigan, 530 Church Street, Ann Arbor, MI48109-5406. E-mail: volling@umich.edu

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

Couple and Family Psychology: Research and Practice © 2015 American Psychological Association2015, Vol. 4, No. 3, 177–197 2160-4096/15/$12.00 http://dx.doi.org/10.1037/cfp0000046

177

portunity to examine variation in marital rela-tionship change after the birth of a second childusing a longitudinal research design that startedin the last trimester of pregnancy and followedcouples throughout the year after the birth oftheir second child (1, 4, 8, and 12 months post-partum). Our main goal was to examine hus-bands’ and wives’ reports of marital changeafter the second birth in an effort to exploredyadic patterns of marital relationship function-ing. In addition, we examined various indicatorsof individual and family functioning before thebirth that would predict these patterns to iden-tify targets of intervention that could assist par-ents making the transition from one child totwo.

Defining Longitudinal Trajectory Patterns

The longitudinal research on marital changeacross the transition to parenthood finds de-clines in marital quality, on average, but there isalso considerable variation among couples (e.g.,Belsky & Rovine, 1990; Cox, Paley, Burchinal& Payne, 1999; Holmes, Sasaki, & Hazen,2013; O’Brien & Peyton, 2002). According totheories of family stress and resilience (e.g.,DeHaan, Hawley & Deal, 2002; Patterson,2002), the stresses associated with the birth of achild may bring about change, but resilient fam-

ilies will adjust and adapt successfully to pre-vent a family crisis, which is evident when thefamily is unable to return to levels of prebirthfunctioning. This perspective on family resil-ience requires a longitudinal design with assess-ments before, during, and after the stressfulevent that allows for an examination of in-creases and decreases in family functioningover time. Different longitudinal patterns de-scribing sudden change with short-term adjust-ment and adaptation (resilience) versus sudden,persistent change over time (crisis) are criticalto understand because they distinguish whichcouples are having the most difficulty adjustingto the birth of a second child. In the currentstudy, we tested five patterns of marital changethat took into consideration the different waysin which couples could adjust and adapt to thebirth of a second child. These patterns are de-picted in Figure 1.

Two patterns have been found previously instudies of the transition to parenthood: no-change (i.e., flat slope) or a pattern of linearchange, either a gradual or sudden decrease orincrease (Belsky & Rovine, 1990; Belsky &Hsieh, 1998). A third pattern describing suddenand persistent change (Doss et al., 2009), orwhat we refer to as the crisis model, reflects apattern of sudden decline that persists over theyear following the birth and indicates that cou-

Figure 1. Hypothesized patterns describing different patterns of change over the familytransition.

178 VOLLING, OH, GONZALEZ, KUO, AND YU

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

ples are unable to adjust and return to prebirthlevels (see Figure 1). Two additional patternshave rarely been examined. The first is a patternof short-term adjustment and adaptation, inwhich couples experience a sudden decline inmarital functioning in the month immediatelyfollowing the birth (adjustment), followed by aperiod of adaptation, whereby marital qualityreturns to prebirth levels. The second pattern isa honeymoon effect in which there is a suddenimprovement after the birth followed by a returnto prebirth levels when the “honeymoon” hasended (e.g., Miller & Sollie, 1980; Wallace &Gotlib, 1990).

To capture variation in marital change, per-son-focused, group-based trajectory analyseshave been conducted to examine different pat-terns of marital relationship change that classi-fied spouses showing similar trajectories (e.g.,linear decline, no change) into groups (e.g.,Belsky & Rovine, 1990; Belsky & Hsieh, 1998;Don & Mickelson, 2014). In most instances,data for husbands and wives were analyzedseparately. Because of the interdependence ofmarried spouses, marital data need to be ana-lyzed at a dyadic, not individual, level of anal-ysis (Kenny, Kashy, & Cook, 2006). The cur-rent study takes a dyadic approach inidentifying patterns of marital change for hus-bands and wives.

Marital trajectories may be similar or discrep-ant for husbands and wives after the birth of aninfant. For instance, Cowan and colleagues(1985) found that wives’ marital satisfaction, onaverage, declined suddenly after the first birth,whereas husbands’ decline was more gradual inthe 18 months after the birth, even though bothwere equally dissatisfied by 18 months. Otherstudies find that marital changes were similarfor husbands and wives when using dyadic,within-couple analyses using latent growthcurves where husbands’ and wives’ slopes andintercepts were often correlated (e.g., Cox et al.,1999; Holmes, Sasaki, & Hazen, 2013; Karney& Bradbury, 1995; O’Brien & Peyton, 2002).One of the central goals of this research was toidentify couples where husbands and wivesshowed similar relationship change trajectoriesfrom couples with discrepant marital trajecto-ries. Our first hypothesis was that different sub-groups of couples would emerge, some in whichthe husbands’ and wives’ marital change trajec-

tories would be similar and others in which theywould differ.

Positive and NegativeRelationship Dimensions

In most studies, a single global assessment,usually marital satisfaction, is used to assessmarital relationship change when, in fact, mar-ital quality may best be represented by bothpositive and negative dimensions (Fincham &Linfield, 1997; Mattson, Paldino, & Johnson,2007). Negative and positive dimensionschange differently across the transition to par-enthood for both husbands and wives (e.g., Bel-sky & Rovine, 1990; Don & Mickelson, 2014;Doss et al., 2009; Holmes et al., 2013), and thismay also be the case for the transition after thesecond child. Knowing whether change ismostly an increase in negative marital quality(e.g., poor communication patterns) or declinesin positive marital quality (e.g., decreased com-panionship) provides important information fordesigning couple-based interventions becausedifferent intervention approaches would beneeded (Cowan & Cowan, 1995; Pinquart &Teubert, 2010).

Ideally, statistical modeling of marital rela-tionship trajectories should take the interdepen-dencies of partner, time, and relationship di-mension into consideration, yet doing sopresents real challenges for researchers exam-ining couple and family relations. A recentstudy by Lindblom and colleagues (2014) high-lights the complexity of group-based analyseswhen multiple dimensions over multiple timepoints for multiple individuals were examined.The resulting groups were much smaller in sizethan those identified when analyzing a singledimension for one individual separately. In asample of 715 Finnish couples going throughthe transition to parenthood, they identified 11family system typologies. The largest group ofcohesive families represented 34% of the sam-ple, yet there were also smaller groups thatincluded 5% to 15% of families, and severaleven smaller groups with as few as 2% of fam-ilies. We expected a similar situation in thecurrent exploratory investigation examiningpositive and negative marital reports for hus-bands and wives over 5 time points, with sev-eral small classes (i.e., 2% to 5%), in addition toclasses with a larger percentage of couples.

179MARITAL CHANGE AFTER SECOND CHILD

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

These smaller classes are important to considerbecause their unique circumstances may repre-sent some of the riskier family situations in needof intervention (see also Rutter, 1996).

Although a recent meta-analysis showed thatmarital relationship functioning continued todeteriorate after the birth of subsequent children(see Twenge et al., 2003), none of the studiesexamined the transition from one child to two,but compared primiparous and multiparous (2or more children) couples. Only Krieg’s (2007)study specifically examined marital changefrom pregnancy to 1 month after birth for first-time mothers and second-time mothers, andfound that after controlling for the length ofmarriage, first-time mothers reported more pos-itive and less negative marital relations thansecond-time mothers. The current research ex-pands upon this prior work in three ways by (a)including husbands’ reports of marital quality,allowing (b) a dyadic examination of maritalrelationship change beyond (c) the first month(a period of adjustment) to include additionaltime points to test for adjustment and adapta-tion. Understanding marital relationship pat-terns takes on added significance for familieshaving a second child because of the knownnegative effects of openly, unresolved maritalconflicts for children’s development (Gordis,Margolin, & John, 2001; Kouros, Cummings, &Davies, 2010). All families in the current studyhad a firstborn child at the time of the transition,so any increases in marital conflict could bedetrimental to the firstborn’s well-being andcontribute to adjustment difficulties once theinfant was born (Volling, 2012).

Antecedents of Marital RelationshipChange Patterns

Once patterns of marital relationship changewere identified, our second goal was to under-stand what prebirth indicators would distinguishthe different marital change patterns. We reliedon the Vulnerability-Stress-Adaptation (VSA)model (Karney & Bradbury, 1995) which un-derscores three areas for explaining variabilityin marital relationship change: (a) enduring vul-nerabilities (e.g., intrapersonal characteristics);(b) the nature of the stressful event; and (c)adaptive processes (e.g., communication) thecouple uses to cope with the stressful event. Wefocus on the vulnerabilities, stresses, and adap-

tive processes from the prenatal period becauseour goal was to identify at-risk families beforethe birth so that interventions could eventuallybe developed that targeted the most influentialrisk factors.

Enduring Vulnerabilities

Enduring vulnerabilities, specifically de-pressed mood and personality traits, can in-crease the likelihood of marital dissatisfactionand decline. Maternal depression is one of themost frequent complications of pregnancy andbirth (Flynn, 2010), with approximately 23% ofwomen and 10% of men reporting depressivesymptoms in the postnatal period (Paulson &Bazemore, 2010). Depressed mothers and fa-thers have reported increased marital dissatis-faction and conflict, and less positive maritalinteractions, after the transition to parenthood(Bower, Jia, Schoppe-Sullivan, Mangelsdorf, &Brown, 2013; Cox et al., 1999; Holmes et al.,2013). No study has examined whether similarlinks exist during the transition to the secondchild, although enduring vulnerabilities shouldcontinue to place parents at-risk during otherstressful life events. We hypothesized thatmothers and fathers with higher depressivesymptoms would have lower marital quality,and quite possibly more discrepant relationshipsshould one spouse be depressed and the othernot.

One personality characteristic that has beenlinked consistently with marital relationshipfunctioning is neuroticism, which reflects anindividual’s tendency to respond negativelywith more avoidance and withdrawal from un-pleasant situations (e.g., marital conflict). In-deed, couples high on neuroticism are generallyless satisfied with their marriages and have lesssatisfied partners (e.g., Belsky & Hsieh, 1998;Bower et al., 2013). Bower et al. (2013) re-ported that mothers’ and fathers’ neuroticismwas related to their partners’ reports of maritalsatisfaction during the pregnancy for first-timeand non-first-time parents, but did not predictchange trajectories after the birth.

Nature of the Stressful Event

The VSA model also claims that variability inmarital change can be tied to the nature of thestressful event. The transition may differ de-pending on whether the pregnancy was planned,

180 VOLLING, OH, GONZALEZ, KUO, AND YU

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

the length of the marriage, and the birth intervalbetween the first and second children, withmore strain on the marriage when the pregnancyis unplanned (Belsky & Rovine, 1990; Cox etal., 1999; Lawrence, Rothman, Cobb, Rothman,& Bradbury, 2008), when couples have beenmarried fewer years (Belsky & Rovine, 1990;Doss et al., 2009; Mortensen, Torsheim, Melke-vik, & Thuen, 2012) and when the spacingbetween children is small.

Parenting stress and how hassled the parentsfeel in caring for the firstborn may also play arole in how couples adjust after the birth of thesecond child. Given the well-established linkbetween marital relations and parenting (Erel &Burman, 1995), prebirth parenting stress maypredict marital decline. Indeed, Ahlborg andcolleagues (2009) found that parents who had asecond child reported more strain from chil-dren’s behavior than parents who had one child,and this was more so with mothers than fathers.Stewart (1990) also reported that dealing withthe firstborn’s behavior was the most commonstress reported by second-time mothers.

The division of childcare also becomes moretraditional after the transition to parenthood,with mothers performing more childcare thanfathers (e.g., Goldberg & Perry-Jenkins, 2004;Sanchez & Thomson, 1997). Mothers’ violatedrole expectations and perceptions of unfairnesswith respect to the division of labor have beenlinked to marital dissatisfaction (Dew & Wil-cox, 2011), and husbands performing morechildcare were less satisfied with their mar-riages than husbands in more traditional mar-riages (Belsky & Hsieh, 1998). Fathers’ child-care participation, however, particularly withthe first child, after the second birth may helpease the adjustment for the first child, as well aschildcare demands on mothers, who are primar-ily responsible for the infants’ care in the earlymonths (Kreppner, Paulsen, & Schuetze, 1982;Stewart, 1990). Thus, we hypothesized that par-enting stress and women’s greater involvementin childcare with the firstborn would be relatedto lower marital quality.

Adaptive Processes

Adaptive processes such as marital commu-nication patterns and familial social support canhelp parents cope with the stressful event andpotentially prevent marital decline. In earlier

studies, poor marital communication patternsbefore birth predicted larger declines in maritalrelationship quality after the birth (Cox et al.,1999; Doss et al., 2009; Shapiro, Gottman, &Carrère, 2000), whereas instrumental and emo-tional support from family members was relatedto women’s parenting efficacy and emotionaladjustment to new parenthood (e.g., Haslam,Pakenham, & Smith, 2006; Leahy-Warren, Mc-Carthy, & Corcoran, 2012). We hypothesizedthat declines in marital relationship functioningwould be mitigated when couples had moreconstructive communication skills, engaged inmore positive marital interactions, and receivedmore support from family and friends.

The Current Study

The current study used a person-centered ap-proach to identify patterns of marital relation-ship change after the second birth for married,heterosexual couples starting in the last trimes-ter of the second pregnancy and over the courseof the year after the second child’s birth. Weincluded husbands’ and wives’ reports of bothpositive and negative marital relations simulta-neously to identify dyadic patterns of adjust-ment and adaptation that could describe differ-ent change trajectories. Finally, in line withVSA theory, we examined individual vulnera-bilities, the nature of the stress surrounding thetransition, and adaptive processes that were as-sociated with patterns of marital relationshipchange. All of these goals were designed toidentify couples experiencing marital difficul-ties and the risk factors that predicted thesedifficulties so that successful couple-based in-terventions could be developed to assist thesefamilies.

Method

Participants

Participants included 241 two-parent familiesconsisting of fathers, mothers, and first-bornchildren (mean age � 31.12 months, SD �10.12 at time of infant’s birth) recruited for alongitudinal study examining change in chil-dren’s adjustment and family relationships fol-lowing the birth of a second child. The recruitedsample was primarily European American(86.3% of fathers, 85.9% of mothers) with

181MARITAL CHANGE AFTER SECOND CHILD

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

13.7% of fathers and 14.1% of mothers repre-senting other racial and ethnic groups. Thelength of marriage ranged from .58 to 20 years(M � 5.77, SD � 2.74). The majority of fathers(79.2%) and mothers (83.9%) earned a bache-lor’s degree or higher, and the mode for annualhousehold income was $60,000 to $99,999(37.8%).

The study consisted of five longitudinal time-points, starting in the last trimester of the wom-en’s pregnancies with the second child and 1, 4,8, and 12 months after the infant’s birth. Datafor the present report included self-reports ofmarital relationship functioning obtained fromboth husbands and wives at each of the fivetime-points, prebirth reports on family demo-graphics, depressed mood, personality, maritalsatisfaction, the division of childcare (with thefirst child), marital communication, parentingstress, and social support, as well as a prebirthobservational assessment of marital interaction.

Of the initial 241 families recruited, 203 fam-ilies remained in the study and participated atthe 12-month time point. Thirty-eight familieshad missing data at 12 months. One family hadmissing data because they could not schedulefor the 12-month home visit within the requiredtime period; the other 37 families dropped for avariety of reasons (e.g., no longer interested,moving from the area, not enough time). The203 remaining families were not significantlydifferent from the recruited sample on years ofmarriage, wives’ or husbands’ ages, and wives’or husbands’ race/ethnicity. Remaining familieshad significantly higher incomes, �2(3) � 13.94, p � .01, and both wives, �2(2) � 7.90, p �.05, and husbands, �2(3) � 10.82, p � .05, werebetter educated. Wives remaining in the study at12 months had lower marital negativity scoresprenatally, t � 2.00, p � .05, whereas remain-ing fathers did not differ significantly on any ofthe marital measures.

Positive and Negative MaritalRelationship Quality

Husbands and wives completed the 25-itemIntimate Relations Questionnaire (IRQ: Braiker& Kelley, 1979), which assesses: love (“Towhat extent do you have a sense of belonging toyour spouse/partner?”), ambivalence (“Howconfused are you about your feelings towardyour spouse/partner?”), maintenance (“How

much do you and your spouse/partner talk aboutthe quality of your relationship”), and conflict(“How often do you feel angry or resentfultoward your partner?”). Items were rated on a9-point scale ranging from 1 (not at all/never) to9 (very much/extremely) and averaged for eachscale. Internal consistency across the five timepoints ranged from .64 to .89 for wives (M �.76), and from .63 to .88 for husbands (M �.75). The sum of the maintenance and love sub-scales comprised the positive marital relationscomposite, r � .48 to .58 for wives, .40 to .53 forhusbands; all ps � .001, whereas the negativemarital relations composite was the sum of theconflict and ambivalence subscales, r � .46 to .56for wives, .54 to .60 for husbands; all ps � .001.This measure has been used extensively in earlierstudies of marital relationship change across thetransition to parenthood, making it a perfectsource for comparison (e.g., Belsky & Rovine,1990; Holmes et al., 2013; Krieg, 2007; MacDer-mid, Huston, & McHale, 1990).

Prenatal Antecedents:Enduring Vulnerabilities

Personality. Both husbands and wives com-pleted the 12 items of the neuroticism subscale ofthe NEO Five-Factor Inventory-Short Form(Costa & McCrae, 1985). Couples rated eachstatement (e.g., neuroticism: “I often feel tenseand jittery,” from 1 (strongly disagree) to 5(strongly agree) and items were summed. TheNEO-FFI has good reliability and validity (Mc-Crae & Costa, 2004); internal consistency for neu-roticism was .87 for wives and .85 for husbands.

Depressed mood. Husbands and wives alsocompleted the Beck Depression Inventory (BDI:Beck, Ward, Mendelson, Mock, & Erbaugh,1961). Both spouses reported on the frequency of21 items, using a 0 to 3 scale, with 0 (no depres-sive symptoms) to 3 (severe depressive symptoms).The BDI is widely used and has well-documentedconcurrent and discriminant validity, with ade-quate internal consistency for the current study(� � .85 for wives, .79 for husbands).

Prenatal Antecedents: Nature of theStressful Event

Division of childcare. We assessed divi-sion of childcare for the firstborn using 11 itemsfrom the Child Care Checklist (Ehrenberg,Gearing-Small, Hunter & Small, 2001). During

182 VOLLING, OH, GONZALEZ, KUO, AND YU

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

a joint-couple, prenatal interview, both spouseswere asked to agree on who performed variouschildcare tasks (e.g., taking older child to thedoctor, putting older child to sleep at night) ona scale from 1 (almost always wife), 2 (usuallywife), 3 (both equally), 4 (usually husband), and5 (almost always husband). Items were aver-aged (� � .73).

Parenting stress. Parenting stress was as-sessed with 14 items from the Parenting DailyHassles scale (PDH: Crnic & Greenberg, 1990),which measured how hassled each parent feltwhen taking care of their firstborn child (e.g.,“You continually have to clean up after yourolder child’s messes”), using a 5-point scale,ranging from 1 (no hassle) to 5 (huge hassle).Mean scores across items were calculated forwives (� � .84) and husbands (� � .83).

Length of marriage/interbirth interval/planned pregnancy. During a joint coupleinterview at the first prenatal home visit,spouses reported on the length of their marriagein years. The interbirth interval (in months) wascalculated from the child’s and the infant’sbirthdates. Spouses were also asked whethertheir second pregnancy was planned using fivecategories: (a) both parents wanted and plannedfor second child; (b) both parents wanted sec-ond child, but not right now; (c) neither parenthad planned nor wanted second child; (d) wifewanted second child, but husband did not; (e)husband wanted second child, but wife did not.Because most couples reported that both parentswanted and planned for the second child (196 of229, or 85.6%), we collapsed the remaining fourcategories to form an unplanned group (n �33).

Prenatal Antecedents: Adaptive Processes

Marital satisfaction. Spouses also com-pleted the 3-item Kansas Marital SatisfactionScale (KMSS: Schumm et al., 1986) and ratedeach item from 1 (not at all) to 5 (extremely).The KMS has good concurrent and discriminantvalidity (Schumm et al., 1986). Mean scoreswere calculated across items for husbands (� �.92) and wives (� � .92).

Social support. During the prenatal coupleinterview, social support was measured usingthe 12-item parenting support scale (Bonds,Gondoli, Sturge-Apple, & Salem, 2002). Itemswere rated from 1 (never) to 5 (quite often), and

averaged to assess the types of assistance thatspouses find helpful from others (e.g., “To whatextent do you feel you have someone to talk toabout things that worry you,” � � .86 for wivesand .86 for husbands).

Marital communication. Spouses’ percep-tions of dyadic communication were measuredwith a revised version of the CommunicationPatterns Questionnaire (Heavey, Larson, Zum-tobel, & Christensen, 1996). Spouses reportedseparately on how they typically dealt with dis-agreements surrounding the division of house-hold labor and childcare, on a scale from 1 (veryunlikely) to 9 (very likely). Items were summedto create composites of constructive communi-cation (e.g., “When you disagree about child-care and household tasks, both of you try todiscuss the problem,” 3 items, � � .75 forhusbands, � � .81 for wives) and destructivecommunication subscales (e.g., “When discuss-ing disagreements about childcare or householdtasks, both of you blame, accuse, and criticizeeach other,” 4 items, � � .76 for husbands, .78for wives).

Home observation of marital interaction.At the first prenatal home visit, husbands andwives were asked to engage in a 10-min maritaldiscussion in which they were instructed to dis-cuss their day as would be typical for them (e.g.,what they did that day, anything about whichthey were particularly excited or upset). Wechose this discussion format over the standardproblem-solving discussion because we did notwant to provoke marital conflict during homeobservations with a child present. Discussionswere videotaped and husbands’ and wives’ be-haviors and affect were later coded by trainedindependent coders using the Interactional Di-mensions Coding System (Kline et al., 2004).Each 10-min interaction was broken down intothree equal segments of three minutes and 20seconds, and each segment was rated on a9-point rating scale from 1 (extremely unchar-acteristic) to 9 (extremely characteristic) to as-sess positive affect – the positivity of facialexpressions, tone of voice, and body language;negative affect – the negativity of facial expres-sions, tone of voice, and body language; domi-nance – control that one partner has over theother; support validation – positive listeningand speaking skills that demonstrated support ofthe other partner; conflict – expressed strugglebetween two individuals; withdrawal – avoid-

183MARITAL CHANGE AFTER SECOND CHILD

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

ance of the interaction; and communicationskills – an individual’s ability to conveythoughts and feelings in a clear, constructivemanner. Means across segments were calcu-lated for each code. Interrater reliability (intra-class correlations) ranged from .88 to .95 (M �.91) for wives and .78 to .92 (M � .88) forhusbands. A composite score for positive inter-action was created by summing positive affect,support validation, and communication skills,� � .59 for wives and .70 for husbands, and fornegative interaction by summing negative af-fect, dominance, conflict, and withdrawal, � �.71 for wives and .73 for husbands. Higherscores indicated more positive and more nega-tive marital interaction.

Data Analysis Plan: Defining MaritalChange Using Linear andNonlinear Models

To maintain a multidimensional, dyadic rela-tionship perspective (Kenny, Kashy, & Cook,2006), we first fit an unconditional latent growthcurve model (LGCM) with four parallel pro-cesses (positive and negative for wives, positiveand negative for husbands) to test the hypothe-sized linear and nonlinear growth trajectories,using particular orthogonal polynomial con-trasts to determine the overall general pattern ofchange for the entire sample: the latent lineargrowth curve model (random intercept and lin-ear slope), the crisis model (random interceptand linear slope; fixed overall quadratic effectover the 5 time points); the adaptation andadjustment (AA) and honeymoon models wereboth tested by including a random intercept andlinear slope and a fixed effect local quadraticassessing deflection at one month postpartumrelative to the prenatal and 4-month time points.Time was centered across all models at theprenatal time point. The paths from the latentlinear slope to the observed items were con-strained to be 0, 1, 2.5, 4.5, and 6.5 to takeinto account the uneven intervals between theprenatal, 1-, 4-, 8- and 12-month time points,respectively. The paths from the AA effect tothe observed items were constrained to be �1(prenatal), 2.5 (1-month), �1 (4-months), 0(8 months), and 0 (12 months) so that thecontrast was orthogonal to the linear slope,which allowed us to test whether this effectwas in addition to the linear pattern. The

paths for the crisis model (overall quadraticeffect) to the observed items were constrainedto be the square of the linear contrast paths asis typically done in latent growth curve mod-eling. Models were estimated in Mplus Ver-sion 6.0 (Muthén & Muthén, 1998 –2010) us-ing full-information maximum likelihood(FIML) estimation for missing data, resultingin 229 families for analyses (of the 241 fam-ilies recruited, 10 excluded because of miss-ing data across all time-points and 2 outliers).

Because the unconditional model indicatedsignificant variance around the growth parame-ters (see below), we then used Growth MixtureModeling (GMM) as a means of identifyingperson-centered, group-based trajectory pat-terns. In the final step, we conducted mixedmodel (spouse x class) ANOVAs, with spouseas a repeated factor to describe how each mar-ital class differed on prenatal VSA indicators.

Results

We present results from the unconditionallatent growth curve model (LGCM) with fourparallel processes first and then move into thepresentation of the growth mixture model(GMM). The fit indices for the unconditionalLGCM follow: (a) Linear, �2(166) � 300.17,p � .001, CFI � .960, RMSEA � .059; (b)Crisis, �2(162) � 285.56, p � .001, CFI �.964, RMSEA � .058; (c) Adjustment and Ad-aptation (AA), �2(162) � 270.61, p � .001,CFI � .968, RMSEA � .054. The growth pa-rameters for husbands’ and wives’ marital pos-itivity and negativity are shown in Table 1. Thefixed effect for the linear slope of wives’ maritalnegativity curve is significant indicating a linearincrease in negativity after the birth of the sec-ond child, but the fixed effects for the linearslopes are nonsignificant for husbands’ negativ-ity, and both husbands’ and wives’ positivity,indicating no change. The random effects, how-ever, indicate there is substantial variabilityaround the intercepts and slopes. Comparing thetheoretical models with the unconditional linearmodel reveals support for the adjustment andadaptation (AA) model. Chi-square tests indi-cated that the unconditional crisis model,��2(4) � 14.61, p � .01, and the unconditionalAA model, ��2(4) � 29.55, p � .001, were betterfitting models than the unconditional linear model;and the unconditional AA model was a better

184 VOLLING, OH, GONZALEZ, KUO, AND YU

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

fitting model than the unconditional crisis model(crisis model AIC � 14362.976 vs. unconditionalAA model, AIC � 4348.036). Thus, the uncon-ditional AA model was chosen as our final modelto be used in the Growth Mixture Modeling thatfollowed, as based on the recommendations ofMuthén and Muthén (2000).

The fixed effects of the GMM (i.e., intercept,linear slope, and AA contrast) were freely esti-mated for each class and the random variance ofgrowth parameters (i.e., intercept, linear slope)were constrained to be equal across classes. Theresidual variances were estimated freely foreach time point, but were constrained to beequal across classes. The intercepts and linearslopes were allowed to correlate across the fourparallel processes. When estimating the full-information maximum likelihood (FIML) mix-ture models, we increased the random start val-ues as needed to ensure a global maximumsolution (see Nylund, Asparouhov, & Muthén,2007). GMMs were run until the model solu-tions were replicated with different starting val-ues (range of random start values was 100–1700) to avoid a local maximum likelihoodsolution and ensure a global optimum. Maritalnegativity scores were reverse-coded to easeconvergence of the GMM (Kline, 2010), but forease of interpretability, the marital negativityscores were plotted in Figure 2 using their orig-inal scaling. We followed standard GMM mod-eling procedures and evaluated models with dif-ferent numbers of latent classes to determinewhich model provided the best fit to the data.We estimated fit indices for 1- (unconditionalmodel) to 8-class-solution models, and used

recommended standards by Masyn (2013) forclass enumeration.

Because models with different numbers ofclasses were not nested, a model comparisonwas conducted using a set of fit indices, in-cluding the Bayesian Information Criterion(BIC; Schwarz, 1978), the sample size ad-justed BIC (SABIC; Sclove, 1987), and theAkaike Information Criterion (AIC; Akaike,1987); lower scores represent better fittingmodels. To further guide model selection, wealso used the Lo-Mendell-Rubin (LMR) like-lihood ratio test of model fit and the entropymeasure, which refers to the average classifi-cation accuracy in assigning individuals toclasses; values range from zero to 1, withhigher scores reflecting better accuracy inclassification of class membership.

Based on fit indices, the six-class solutionwas considered the best fitting model because ithad a lower AIC and SABIC, AIC � 14218.00,SABIC � 14253.16, than the 5 class, AIC �14230.07, SABIC � 14261.80, and 7 classmodels, AIC � 14239.73, SABIC � 14278.32,and higher entropy (.844) than the 7-classmodel (.787). The estimated growth curves andobserved means for husbands’ and wives’ mar-ital negativity and positivity for the six classesare shown in Figure 2. Table 2 presents esti-mates and standard errors for the fixed effectsfor each class, which also assists in interpreta-tion of the changes observed within each class(a table of random effects is available on re-quest). Fixed effects estimates are raw coeffi-cients so in the present model can be interpreted

Table 1Growth Parameter Estimations of the Unconditional Latent Growth Curve Model With FourParallel Processes

Spouse

Growth parameters of marital positivity Growth parameters of marital negativity

Intercept Linear slope AA Intercept Linear slope AA

WifeMean (SE) 13.302��� (.12) �.007 (.02) .056� (.02) 6.223��� (.13) .061��� (.02) .014 (.03)Variance (SE) 2.863��� (.32) .019�� (.01) 0a 3.111��� (.35) .018� (.01) 0a

HusbandMean (SE) 13.109��� (.12) �.014 (.02) .092��� (.03) 6.248��� (.13) �.007 (.02) .086�� (.03)Variance (SE) 2.589��� (.29) .017�� (.01) 0a 3.390��� (.38) .020�� (.01) 0a

Note. AA � Adjustment and Adaptation testing deflection of 1 month from prenatal and 4 months.a Random variance for the AA effect was constrained to 0.� p � .05. �� p � .01. ��� p � .001.

185MARITAL CHANGE AFTER SECOND CHILD

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

as the change in the dependent variable for aone month change in time. For example, thesignificant �.092 linear slope for mother’s pos-itivity in Class 1 is interpreted as a drop of .09in the positivity score per month. In this way theraw score coefficients serve as the effect sizeestimates. Because of the exploratory nature ofthe analyses, each class was labeled in line withthe overall change patterns describing husbandsand wives, and whether trajectories were similaror discrepant across spouses.

Trajectory Classes of MaritalRelationship Change

The largest class (C1) was denoted as the wifedecreasing positivity– husband honeymoon(44.1%, n � 101). As shown in Table 2 andFigure 2, both husbands and wives in C1 mar-riages reported high levels of marital positivityat the prenatal time point (see intercepts inTable 2). Although there was a relatively small,but significant linear decline in wives’ marital

Figure 2. Estimated mean trajectories of GMM 6-class solution for marital positivity(black) and negativity (grey) for wives (solid lines) and husbands (dashed lines) withobserved means.

186 VOLLING, OH, GONZALEZ, KUO, AND YU

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

positivity over time (see slope in Table 2), therewas no change in husbands’ marital positivitythroughout the first year after the birth of thesecond child. Husbands and wives both reportedlow levels of marital negativity at the prenataltime-point; the linear effects were not signifi-cantly different from 0. Husbands showed ahoneymoon effect; a small, significant decreasein marital negativity from the prenatal to1-month time points, followed by an increasefrom 1 to 4 months, with no change throughoutthe last half of the year.

The second class (C2) was labeled wife in-creasing negativity–husband adjustment andadaptation (34.5%, n � 79). This class wascharacterized by husbands who showed a sig-nificant adjustment and adaptation effect formarital positivity and wives who showed a sig-nificant linear increase in marital negativity (seeTable 2). These couples appeared to have amore difficult transition than C1 couples be-cause wives reported increasing marital nega-tivity and their husbands showed a significantAA effect; an immediate, but short-lived, de-cline in positive marital relations following thebirth, with adaptation and a return to prebirthlevels of positive marital relations again by 4months (see also Figure 2).

In the third class (C3) of couples (7.4%, n �17), wives showed an initial “honeymoon ef-fect” in that there was a decrease in maritalnegativity from prenatal to 1 month, followedby an increase in negativity from 1 to 4 months,and a linear increase throughout the year. Hus-bands reported a significant linear increase overtime on marital positivity. Unlike C1 and C2couples, there was a greater discrepancy in pos-itivity between husbands and wives in C3 cou-ples, with wives reporting more marital positiv-ity than husbands (Wald Z � 2.49, p � .05). Welabeled these couples wife honeymoon–discrep-ant marital positivity.

A fourth class (C4, 6.9%, n � 16) differedfrom other classes in that the transition seemedto affect the wives more than the husbands.Although husbands and wives reported simi-larly high levels of marital positivity and simi-larly low levels of negativity prior to the birth,wives experienced a severe adjustment period(sudden and dramatic increase in marital nega-tivity and concurrent decrease in marital posi-tivity from prebirth to 1 month), but recovered(i.e., adapted) by 4 months, as well as a signif-T

able

2T

he6-

Cla

ssSo

luti

onof

GM

M:

Fix

edE

ffec

tsP

aram

eter

Est

imat

esan

d(S

tand

ard

Err

ors)

for

Fix

edE

ffec

ts(N

�22

9)

Eff

ect

Cla

ss1

(n�

101)

(44.

1%)

Cla

ss2

(n�

79)

(34.

5%)

Cla

ss3

(n�

17)

(7.4

%)

Cla

ss4

(n�

16)

(6.9

%)

Cla

ss5

(n�

12)

(5.2

%)

Cla

ss6

(n�

4)(1

.7%

)

Wif

eH

usba

ndW

ife

Hus

band

Wif

eH

usba

ndW

ife

Hus

band

Wif

eH

usba

ndW

ife

Hus

band

Mar

ital

posi

tivity

Inte

rcep

t14

.075

��

�(.

131)

13.4

89�

��

(.24

0)12

.694

��

�(.

322)

12.9

59�

��

(.51

7)14

.086

��

�(.

437)

12.5

98�

��

(.68

5)12

.051

��

�(.

625)

12.3

91�

��

(.81

2)10

.292

��

�(.

734)

12.7

61�

��

(.42

0)15

.907

��

�(.

481)

12.7

81�

��

(.51

4)Sl

ope

�.0

92�

�(.

035)

.014

(.03

6).0

10(.

035)

�.0

97(.

082)

.085

(.05

4).1

18�

(.05

8).1

55�

(.06

6).0

34(.

057)

.234

��

(.07

9)�

.141

�(.

069)

.025

�(.

011)

.344

��

�(.

089)

AA

a.0

34(.

038)

.037

(.03

1).0

56(.

049)

.172

��

�(.

047)

�.0

56(.

088)

.002

(.09

9).4

31�

�(.

139)

.016

(.08

9)�

.090

(.07

9)�

.104

(.05

6)�

.034

(.07

9)1.

460�

��

(.22

4)M

arita

lne

gativ

ityb

Inte

rcep

t13

.219

�(.

197)

13.3

95�

��

(.33

9)13

.413

��

�(.

288)

12.5

17�

��

(.50

2)11

.221

��

�(1

.321

)11

.140

��

�(1

.195

)12

.075

��

�(.

525)

12.8

76�

��

(.71

9)8.

604�

��

(.50

5)11

.493

��

�(.

605)

13.7

11�

��

(.03

6)12

.489

��

�(.

449)

Slop

e.0

15(.

027)

.001

(.03

6)�

.213

��

�(.

031)

�.0

17(.

063)

�.2

10�

(.10

2).2

11(.

218)

.047

(.08

6)�

.049

(.10

2).2

57�

�(.

082)

�.0

87(.

067)

�.0

75�

�(.

022)

.227

��

�(.

025)

AA

b.0

20(.

035)

�.0

96�

(.04

8)�

.047

(.08

5)�

.074

(.07

5)�

.509

��

�(.

115)

�.1

12(.

112)

.704

��

�(.

099)

.010

(.08

8)�

.362

��

(.11

5)�

.335

�(.

145)

.168

(.24

0).4

15�

��

(.11

8)

Not

e.C

lass

1�

wif

ede

crea

sing

posi

tivity

–hu

sban

dho

neym

oon;

Cla

ss2

�w

ife

incr

easi

ngne

gativ

ity–

husb

and

adju

stm

enta

ndad

apta

tion;

Cla

ss3

�w

ife

hone

ymoo

n–di

scre

pant

mar

ital

posi

tivity

;C

lass

4�

wif

ead

just

men

tan

dad

apta

tion;

Cla

ss5

�co

uple

hone

ymoo

nw

ithdi

scre

pant

mar

ital

nega

tivity

and

posi

tivity

;C

lass

6�

husb

and

adju

stm

ent

and

adap

tatio

n.a

Apo

sitiv

eA

Ava

lue

for

mar

ital

posi

tivity

mea

nsth

atpo

sitiv

ityin

itial

lyde

clin

edfr

ompr

enat

alto

1m

onth

and

then

incr

ease

d(a

djus

tmen

tan

dad

apta

tion)

,w

here

asa

nega

tive

valu

ein

dica

tes

that

ther

ew

asan

initi

alin

crea

sein

posi

tivity

from

pren

atal

to1

mon

th,

with

asu

bseq

uent

decl

ine

(i.e

.,ho

neym

oon

effe

ct).

bM

arita

lne

gativ

itysc

ores

wer

ere

vers

e-co

ded

inth

eG

MM

soa

nega

tive

slop

ein

the

tabl

ein

dica

tes

anin

crea

sein

nega

tivity

.Apo

sitiv

eA

Ava

lue

for

mar

ital

nega

tivity

mea

nsne

gativ

ityin

crea

sed

from

pren

atal

to1

mon

thth

ende

clin

edby

4m

onth

s(a

djus

tmen

tand

adap

tatio

n),a

nda

nega

tive

AA

valu

ein

dica

tes

that

nega

tivity

initi

ally

decl

ined

from

pren

atal

to1

mon

than

dth

enin

crea

sed

by4

mon

ths

(i.e

.,ho

neym

oon

effe

ct).

�p

�.0

5.��

p�

.01.

���

p�

.001

.

187MARITAL CHANGE AFTER SECOND CHILD

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

icant linear increase in positivity thereafter. Incontrast, husbands’ reports of marital positivityand marital negativity were unchanged through-out the first year after the birth. We labeled thisclass the wife adjustment and adaptation class.

For the fifth class (C5, 5.2%, n � 12) wiveshad very low levels of marital positivity andvery high levels of marital negativity before thebirth (see Figure 2 and intercepts in Table 2).Wives reported significant improvement in mar-ital positivity, and declines in marital negativ-ity, over the year following birth, but still re-ported lower levels of marital positivity at 12months in comparison to the other classes. Hus-bands, by contrast, showed moderate levels ofpositivity with a significant decrease in positiv-ity over time. Although wives reported im-provement over time in positive marital rela-tions, their husbands actually reported declinesin marital positivity over this same period.There was also a honeymoon effect for bothhusbands’ and wives’ negativity, in which mar-ital negativity declined from prenatal to 1month, only to increase again at 4 months. Forwives, this increase was followed by a lineardecline in marital negativity throughout the re-maining year, yet their levels of negativity at 12months were still relatively high. Husbands’negativity remained unchanged from 4 monthsto the end of the year with the highest level ofnegativity at 12 months than any other class.Husbands and wives in this class also reporteddiscrepant experiences, with husbands reportingmore positive and less negative experiencesthan their wives. We labeled this class couplehoneymoon–discrepant marital negativity andpositivity,

The sixth and smallest class (C6, 1.7%, n �4) was unique in that it was clear that thetransition was experienced by the husbands dif-ferently than the wives, given the significantadjustment and adaptation response for hus-bands’ marital positivity and negativity. Hus-bands experienced a significant decrease in mar-ital positivity and increase in negativity after thebirth but recovered by 4 months, followed by anincrease in marital positivity and decrease inmarital negativity across the remainder of theyear. Wives in these marriages reported signif-icant linear increases in both marital positivityand marital negativity. We labeled this classhusband adjustment and adaptation.

Prenatal Antecedents of MaritalRelationship Change

Preliminary analyses examining demograph-ics such as household income, husbands’ andwives’ ages, and husbands’ and wives’ educa-tion revealed no significant differences by class.Chi-square analyses also revealed no class dif-ferences on firstborn gender. In an effort todetermine whether there were meaningful dif-ferences across the marital classes before thebirth that might help account for the differentpatterns, we examined prenatal indicators ofenduring vulnerabilities of the couple (e.g., per-sonality, depressed mood), the nature of thestressful event (e.g., marriage length, interbirthinterval, division of childcare, planned preg-nancy, parenting stress), and adaptive processes(e.g., marital satisfaction, marital communica-tion and interaction, social support). Althoughwe expected that the two classes with discrepantpositive and negative marital experiences (C3and C5) would evince greater husband-wife dif-ferences in their enduring vulnerabilities andwould, overall, be lower on adaptive maritalprocesses, we advanced no specific hypothesesabout how each of the classes may differ withrespect to prenatal antecedents given the explor-atory nature of the current study and the factthat no prior study has examined marital changeacross the year following the transition fromone child to two. Given that we do not advancepredictions about potential antecedent predic-tors we focus on comparing means acrossclasses for each variable as a way to describehow the classes differed.

In these analyses, we dropped the smallest C6class with only 4 families from consideration instatistical testing and conducted 2 (spouse) � 5(class) mixed model ANOVAs with spouse as arepeated factor, and our prenatal measures asdependent variables, in an effort to provide adescription of the couples in each class. Weacknowledge the differences in sample sizeacross classes and the possibility of increasedprobability of Type I errors of these post-hoccomparisons; all remaining classes, however,constituted greater than 5% of the sample, inline with Lindblom et al.’s (2014) decision toretain classes larger than 4%. In the ANOVAtesting, we used Tukey’s post-hoc tests to re-duce risk of Type I error because of multipletesting when comparing means across classes.

188 VOLLING, OH, GONZALEZ, KUO, AND YU

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

Enduring vulnerabilities. With respect toenduring vulnerabilities, there were significantspouse main effects for neuroticism, F(1,216) � 16.13, p � .001, p

2 � .07, and de-pressed mood, F(1, 215) � 40.23, p � .001,p

2 � .16. In general, wives reported more neu-roticism and greater depressed mood than theirhusbands. Significant main effects for classwere found for neuroticism and depressed mood(see Table 3). Tukey’s post-hoc comparisons(all ps � .05) revealed that C1 couples (wifedecreasing positivity-husband honeymoon) hadsignificantly lower neuroticism scores than C5couples (couple honeymoon-discrepant positiv-ity and negativity). C5 couples also had signif-icantly higher depressive symptoms thanspouses in all other classes. Spouses in Classes1 and 2 (wife increasing negativity–husbandadjustment and adaptation) had the lowest de-pressive symptoms, whereas C3 couples (wifehoneymoon–discrepant positivity) had signifi-cantly more depressive symptoms than the C1couples, but not as high as C5 couples.

The main effect for depression was qualifiedfurther by a Spouse � Class interaction, F(1,215) � 2.75, p � .05, p

2 � .05. Means in Table3 reveal that wives in C5 (couple honeymoon-discrepant positivity and negativity) reportedgreater depressed mood than all other husbandsand wives, with wives in C3 (wife honeymoon-discrepant positivity) reporting moderately highlevels of depressed mood. C3 husbands alsoreported the highest depressed mood than allother husbands. Wives in C1 (wife decreasingpositivity-husband honeymoon) reported lessdepressed mood than wives in other classes.

Nature of the stressful event. Table 3 in-dicates that there was a significant main effectof class for the length of the marriage, with C4couples (wife adjustment and adaptation) mar-ried longer than C3 (wife honeymoon-discrep-ant positivity), and a significant spouse effectfor parenting stress, F(1, 220) � 8.93, p � .01p

2 � .04, with wives reporting more stress thantheir husbands. Interbirth interval, parentingstress, and the division of childcare did notdiffer across classes. The chi-square for preg-nancy planning approached significance,�2(5) � 9.53, p � .07. A greater percentage(35%) of the second pregnancies were un-planned for couples in Class 3 (wife honey-moon–discrepant positivity) and Class 5 (cou-ple honeymoon–discrepant negativity and

positivity. 25%) compared to Classes 1 (wifedecreasing positivity– husband honeymoon:10%), 2 (wife increasing negativity–husbandadjustment and adaptation: 15%), and 4 (wifeadjustment and adaptation: 12.5%).

Adaptive processes. There were signifi-cant main effects of class for social support,marital satisfaction, destructive marital commu-nication, and positive and negative observedinteraction (see Table 3). In general, couples indiscrepant marriages (C3 and C5) used moredestructive communication patterns when set-tling childcare disagreements than the otherthree classes. Further C5 (couple honeymoon–discrepant negativity and positivity) couples re-ported less social support than C1 (wife de-creasing positivity– husband honeymoon)couples, and C3 couples (wife honeymoon–discrepant positivity) engaged in significantlyless positive and more negative marital discus-sions than C1 couples. Wives also reported sig-nificantly more social support, F(1, 220) �9.57, p � 01 p

2 � .40, and less marital satis-faction, F(1, 215) � 27.64, p � .001 p

2 � .11,than their husbands.

These findings were qualified further by sig-nificant Spouse � Class interactions for socialsupport, F(4, 220) � 2.97, p � .05 p

2 � .05,marital satisfaction, F(4, 215) � 9.20, p � .001,p

2 � .15, and destructive communication, F(4,214) � 5.16, p � .01, p

2 � .09. Follow-upcontrasts (all ps � .05) revealed that both C5(couple honeymoon–discrepant negativity andpositivity) wives and husbands reported less socialsupport than C1 (wife decreasing positivity–husband honeymoon) and C2 (wife increasingnegativity–husband adjustment and adaptation)wives and husbands (see Table 3). C1 wivesalso reported more social support than C3 (wifehoneymoon–discrepant positivity) wives (seeTable 2). Additionally, C1 husbands reportedmore social support than C4 husbands (wifeadjustment and adaptation). Within C1, C2, andC4 couples, wives reported more social supportthan their husbands.

C5 wives engaged in more destructive com-munication than all other wives, although C3wives also used more destructive communica-tion than C1 and C2 wives. C5 husbands usedmore destructive communication than C1 hus-bands, and C3 husbands used more destructivecommunication than C1, C2, and C4 husbands.As for within-couple differences, C2 husbands

189MARITAL CHANGE AFTER SECOND CHILD

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

Tab

le3

Mea

nsan

dSt

anda

rdD

evia

tion

sof

Pre

nata

lA

ntec

eden

tsby

Tra

ject

ory

Cla

sses

and

Spou

se(H

usba

nd/W

ife)

Wit

hSi

gnifi

cant

Con

tras

ts

Ant

eced

ent

Cla

ss

F

p2C

1M

(SD

)C

2M

(SD

)C

3M

(SD

)C

4M

(SD

)C

5M

(SD

)

End

urin

gvu

lner

abili

ties

Dep

ress

ion

Hus

band

�5.

20a

(.47

)�

6.13

(.53

)�

8.31

b(1

.16)

6.81

(1.1

6)�

7.25

(1.3

4)W

ife

7.74

a(.

52)

8.90

ab(.

58)

11.6

3 b(1

.28)

8.88

ab(1

.28)

16.0

8 c(1

.48)

Cla

ss6.

47a

(.37

)7.

51(.

42)

9.97

b(.

93)

7.84

(.93

)11

.67 c

(1.0

7)7.

55.1

2N

euro

ticis

mH

usba

nd27

.28

(.73

)28

.73

(.83

)29

.06

(1.8

3)29

.69

(1.8

3)29

.25

(2.1

1)W

ife

30.6

0(.

75)

30.7

3(.

86)

32.5

0(1

.88)

31.4

4(1

.88)

38.6

7(2

.17)

Cla

ss28

.94 a

(.53

)29

.73

(.60

)30

.78

(1.3

3)30

.56

(1.3

3)33

.96 b

(1.5

3)2.

69.0

5St

ress

ful

even

tM

arri

age

leng

thC

lass

5.77

(.26

)6.

08(.

30)

4.23

a(.

64)

6.94

b(.

66)

4.94

(.76

)2.

69.0

5In

terb

irth

inte

rval

Cla

ss30

.95

(1.0

2)32

.57

(1.1

9)28

.53

(2.6

4)30

.44

(2.5

6)30

.67

(2.9

5)N

.S.

Div

isio

nof

child

care

Cla

ss2.

32(.

05)

2.33

(.06

)2.

45(.

12)

2.40

(.13

)2.

33(.

15)

N.S

.Pa

rent

ing

stre

ssH

usba

nd2.

20(.

05)

2.21

(.06

)2.

22(.

12)

2.29

(.12

)2.

36(.

14)

Wif

e2.

31(.

05)

2.38

(.06

)2.

23(.

13)

2.34

(.13

)2.

48(.

15)

Cla

ss2.

26(.

05)

2.30

(.05

)2.

23(.

12)

2.31

(.12

)2.

42(.

14)

N.S

.A

dapt

ive

proc

esse

sSo

cial

supp

ort

Hus

band

�3.

43a

(.07

)�

3.34

ab(.

08)

3.23

(.18

)�

2.97

b(.

18)

2.79

c(.

21)

Wif

e3.

51a

(.07

)3.

48a

(.08

)3.

10b

(.17

)3.

28(.

18)

2.95

b(.

21)

Cla

ss3.

47a

(.07

)3.

41(.

08)

3.16

(.17

)3.

13(.

18)

2.87

b(.

20)

2.99

.05

Mar

ital

satis

fact

ion

Hus

band

4.53

a(.

07)

4.29

b(.

08)

3.85

c(.

17)

4.15

bc

(.17

)�

3.58

c(.

20)

Wif

e4.

50a

(.07

)4.

22b

(.07

)3.

81c

(.16

)3.

81c

(.16

)2.

33d

(.19

)C

lass

4.52

a(.

06)

4.25

b(.

07)

3.83

b(.

14)

3.98

b(.

14)

2.96

c(.

17)

23.4

8.3

0C

onst

ruct

ive

com

mun

icat

ion

Hus

band

20.5

1(.

42)

19.7

8(.

48)

18.4

4(1

.04)

18.4

4(1

.04)

17.8

3(1

.21)

Wif

e20

.64

(.47

)19

.40

(.53

)20

.50

(1.1

7)19

.31

(1.1

7)15

.00

(1.3

5)C

lass

20.5

7 a(.

37)

19.5

9 a(.

42)

19.4

7(.

92)

18.8

8(.

92)

16.4

2 b(1

.06)

4.02

.07

Des

truc

tive

com

mun

icat

ion

Hus

band

8.28

a(.

51)

�9.

58ac

(.58

)14

.31 b

c(1

.27)

9.3 a

c(1

.27)

�11

.58 c

(1.4

7)W

ife

7.94

a(.

44)

8.40

a(.

50)

11.8

8 b(1

.10)

10.1

9 ab

(1.1

0)16

.83 c

(1.2

7)C

lass

8.12

a(.

41)

8.99

a(.

46)

13.0

9 b(1

.01)

9.75

a(1

.01)

14.2

1 b(1

.16)

10.3

1.1

6Po

sitiv

em

arita

lin

tera

ctio

nH

usba

nd16

.90

(.27

)16

.07

(.31

)15

.20

(.66

)15

.98

(.68

)15

.69

(.79

)W

ife

16.9

9(.

23)

16.9

0(.

27)

15.2

2(.

57)

16.7

3(.

58)

16.2

5(.

68)

Cla

ss16

.95 a

(.21

)16

.49

(.23

)15

.21 b

(.50

)16

.35

(.51

)15

.97

(.59

)3.

05.0

5N

egat

ive

mar

ital

inte

ract

ion

Hus

band

6.06

(.27

)6.

54(.

30)

7.16

(.65

)7.

69(.

67)

6.97

(.77

)W

ife

6.19

(.27

)6.

78(.

31)

8.65

(.66

)6.

75(.

68)

7.53

(.78

)C

lass

6.12

a(.

22)

6.66

(.26

)7.

90b

(.55

)7.

22(.

56)

7.25

(.65

)3.

11.0

5

Not

e.C

1�

wif

ede

crea

sing

posi

tivity

–hu

sban

dho

neym

oon;

C2

�w

ife

incr

easi

ngne

gativ

ity–

husb

and

adju

stm

ent

and

adap

tatio

n;C

3�

wif

eho

neym

oon–

disc

repa

ntm

arita

lpo

sitiv

ity;

C4

�w

ife

adju

stm

ent

and

adap

tatio

n;C

5�

coup

leho

neym

oon-

disc

repa

ntm

arita

lne

gativ

ityan

dpo

sitiv

ity;

C6

�hu

sban

dad

just

men

tan

dad

apta

tion.

Fva

lues

only

repo

rted

ifp

�.0

5.M

eans

desi

gnat

edw

ithdi

ffer

ent

subs

crip

tsar

esi

gnifi

cant

lydi

ffer

ent

from

each

othe

rba

sed

onpo

st-h

ocT

ukey

com

pari

sons

(p�

.05)

.�{

Des

igna

tes

sign

ifica

ntw

ithin

-cou

ple

gend

erdi

ffer

ence

s.

190 VOLLING, OH, GONZALEZ, KUO, AND YU

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

used more destructive communication than theirwives, but C5 wives were more destructive intheir communications than their husbands.

As for marital satisfaction, C1 couples (wifedecreasing positivity–husband honeymoon) re-ported significantly higher marital satisfactionthan all other couples, and C5 couples (couplehoneymoon–discrepant negativity and positiv-ity) reported significantly lower marital satisfac-tion than all other couples before the birth, withC5 wives reporting the lowest marital satisfac-tion than all others. C5 husbands reported lowermarital satisfaction than other husbands. Hus-bands and wives within Classes 1, 2, and 3 hadsimilar scores and were mutually satisfied, eventhough couples in Class 1 were more satisfiedthan couples in Class 2, who, in turn, were moresatisfied than couples in Class 3. Wives andhusbands had more discrepant marital satisfac-tion scores in Class 5.

Discussion

The transition to parenthood is a stressful lifeevent for many couples. Theories of familystress and resilience indicate that a stressful lifeevent can turn into a family crisis if families areunable to adjust and adapt over time. The maingoal of the current study was to examine dyadicpatterns of marital relationship change after thebirth of a second child using relationship-centered, trajectory-based analyses. One of thestrengths of this work was the direct testing ofseveral potential change trajectories that wouldexplain whether a marriage was in crisis (i.e.,sudden and persistent change), or whether cou-ples experienced an initial adjustment periodfollowed shortly afterward by a period of adap-tation. In general, married couples goingthrough the transition from one child to twochildren experienced a period of adjustmentshortly after the birth (1 month), but adapted by4 months, when their reports of marital qualityreturned to prebirth levels.

Similar to earlier group-based analyses lookingat marital trajectories after the birth of a first child,we found evidence for six different couple-levelpatterns of marital relationship change, indicatingthat husbands (his) and wives (her) do have dif-ferent marital experiences after the birth of a sec-ond infant, although some couples have more dis-crepant marital experiences than others. Weacknowledge that these patterns may be unique to

the period following the second birth, and may notgeneralize to and describe marital change duringother developmental transitions. In contrast tofirst-time parents, couples expecting a secondchild must contend with the care of two youngchildren and organize their childcare responsibil-ities accordingly. The division of childcare didnot, however, differ across the classes. The divi-sion of childcare may not be as critical for thetransition with the second child because spouseshave already negotiated childcare roles after thefirst child and so there are no surprises or violatedexpectations with the more traditional division oflabor that often accompanies the birth of the firstchild.

Our results indicated that it was the manner inwhich couples communicated when there weredisagreements about the division of childcarethat differentiated the marital classes, not whowas doing what and how much. Specifically,discrepant couples with vastly different per-spectives on the positive and negative quality oftheir marriages (C5 and C3), were more likelyto use destructive marital communication(blaming, yelling, and threats) to try and settledisagreements than couples sharing similar per-spectives on marital quality. Further, C3 cou-ples (wife honeymoon–discrepant marital pos-itivity) engaged in more negative and lesspositive marital interaction during observedprebirth marital discussions than C1 couples.

The majority of our couples (C1) reportedhigh positive marital relations and low negativemarital relations over the year after the birth,with only a slight decrease in marital positivityfor wives and evidence of a honeymoon effectfor husbands. Thus, most couples had little dif-ficulty managing the transition to the secondchild. Other couples did have more difficulties,underscoring the degree of individual differ-ences within couples undergoing the transitionfrom one to two children. C2 couples, for in-stance, were similarly high on marital positivityand low on marital negativity initially as the C1couples, but there were more disruptive changesin their marriages with a steady increase inwives’ negativity over time and evidence of aninitial decline in husbands’ positivity shortlyafter birth, with evidence of adaptation by 4months. C4 (wife adjustment and adaptation)couples represented only 7% of the sample butthere appeared to be more marital stress forwives after the birth than their husbands given

191MARITAL CHANGE AFTER SECOND CHILD

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

that there was an initial decline in marital pos-itivity and an increase in marital negativity 1month after the birth. There was also a verysmall group (C6: husband adjustment and ad-aptation) in which husbands displayed a declinein their marital positivity and an increase inmarital negativity after the birth, although therewas little change for wives. Even though thetransition period from one child to two maystress marital relations for some families, it isimportant to note that there was no evidencethat the birth of a second child gave rise to afamily crisis, with a pattern of sudden and per-sistent decline in marital relations over time.Indeed, our findings suggest that in most casesthe marital relationship was resilient andspouses adapted shortly after the transition.

For two of the marital classes, however, thetransition was quite difficult. C5 couples startedoff with a rocky transition having lower levels ofmarital positivity and higher levels of negativitythan all other couples even before the birth. C5wives actually reported higher negativity thanpositivity scores, which was a very rare occur-rence in the current sample. Furthermore, C5 hus-bands and wives had very discrepant perspectives,with husbands reporting more positive and lessnegative marital relations than their wives. Wivesin these couples were very high on neuroticismand depression which may also explain, in part,their negative evaluations of their marriage. Thesecouples also reported the lowest marital satisfac-tion before birth. Despite the accumulation of sig-nificant prebirth risk factors (i.e., neuroticism, ma-ternal depression, marital dissatisfaction), whichcould be seen as a “pile-up of stressors” (McCub-bin & Patterson, 1983) that should set the C5couples up for a marital crisis once the infant wasborn, these marriages actually improved over timewith wives reporting more positivity and less neg-ativity by 12 months. Yet, even with improve-ment, these marriages were still higher on maritalnegativity and lower on marital positivity one yearafter the birth than other marriages. Regardless,early indications of maternal depression pairedwith marital dissatisfaction before the birth maybe enough to warrant further concern about mar-ital well-being, and recommendations for couple-based interventions for these couples.

C3 and C5 couples were further distinguishedfrom other classes by the clear discrepancies inhusbands’ and wives’ reports of positive andnegative marital quality. These couples were

more likely to report the second pregnancy wasunplanned, which could be one reason for thediscrepancies. C3 husbands were clearly lesspositive about their marriages than their wivesinitially and this remained the case over theyear. C3 and C5 couples also used more de-structive communication than other coupleswhen dealing with childcare disagreements. C3and C5 wives experienced a honeymoon effectwith reported declines in marital negativityshortly after birth, but this honeymoon did notlast long and by 4 months, wives were onceagain reporting higher levels of negativity thatappeared to escalate over time, which may beattributable to the greater use of destructivecommunication by these couples.

The present study focused on patterns of mar-ital change during the transition from one childto two. Although there is some evidence thatmultiparous couples with two or more childrenhave lower marital functioning than first-timeparents, this study was the first to examinelongitudinal dyadic change patterns in maritalrelationships for an entire year after the secondinfant’s birth. Although couples already had onechild, our results indicated that parenting stress,the division of childcare, and the age of thefirstborn did not explain the different maritalpatterns. Instead, it was indicators of adaptivemarital processes (e.g., marital interaction, de-structive communication, and social support)and enduring vulnerabilities of spouses (i.e.,personality, depression) that discriminatedamong the different marital patterns. Most cou-ples manage the transition just fine with eitherlittle decline or short-term adjustment difficul-ties indicating that universal prevention focusedon marital decline is probably not needed forcouples expecting their second child. Some cou-ples, however, may benefit from couple-basedinterventions that focus on strengthening pat-terns of constructive marital communication,particularly around child rearing responsibili-ties. Further, maternal depression and a lack ofsocial support with parenting responsibilities inthe prenatal period may be signs of future prob-lems that bode poorly for the marriage and forthe two young children in the family.

There were several unique methodologicalstrengths to the current work that serve as recom-mendations for future research. First, we exam-ined both positive and negative dimensions ofmarital relationship quality simultaneously by in-

192 VOLLING, OH, GONZALEZ, KUO, AND YU

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

cluding the perspectives of both husbands andwives. Second, we had repeated assessmentsacross five time points in the course of a year tocapture complex patterns of marital change withthe arrival of the second child. One of the inno-vations of our modeling strategy was the inclusionof an orthogonal, polynomial contrast that allowedus to test specifically for evidence of both a hon-eymoon effect and an adjustment and adaptation(AA) effect, along with an orthogonal quadraticpolynomial to ascertain whether families under-went a crisis of a sustained period of decline. Theresults clearly indicated that the AA was the bestfitting model for our data. The timing of ourassessments proved critical for uncovering boththe honeymoon effect and the adjustment andadaptation effect because without the 1-month as-sessment, we would have “missed” changes thatoccurred shortly after the birth. By 6 months,when many transition to parenthood studies maketheir first, postbirth contact, families had alreadyadapted to the transition.

Despite the many strengths of this research,there are also several limitations. The study wasdesigned specifically to examine the role of fa-thers in supporting the firstborns’ adjustment afterthe sibling’s birth, so included married, heterosex-ual couples with biological fathers. Couples werealso predominantly white and middle-class. Thereis a clear need to develop further studies usingmore diverse samples of families (e.g., single-parent; adoptive, same-sex couples) from differentsocioeconomic and ethnic backgrounds (Volling,2012). Relationship dynamics and the transition toa second birth may be very different in thesefamily arrangements.

As with many person-centered analysis strate-gies, GMM is an exploratory analytic tool thatallows the identification of couples sharing similartrajectories in a sample. Several of the resultingclasses were quite small (n � 4), which we antic-ipated might be the case when trying to modeldyadic, multidimensional, longitudinal trajecto-ries. These small classes included the discrepantcouples who in most cases were at higher risk ofexperiencing difficulties, as well as couples show-ing pronounced adjustment and adaptation re-sponses. Even though few in number, it is thesecouples who may be most in need of couple-basedinterventions to strengthen marital communica-tion; they should not be dismissed purely for sta-tistical reasons. We believe larger samples andfurther replications will demonstrate that these

smaller classes exist in larger numbers than wewere able to uncover here.

Further, most parents had wanted andplanned the second pregnancies so the patternsof change found here may not reflect patternsfound in samples where pregnancies are unin-tended or unwanted (Barber & East, 2011). Wemust also acknowledge that our results for mar-ital change for second-time parents may be at-tributable, in part, to selection effects as coupleswith distressed marriages may have decided notto have a second child or may have separatedafter the birth of the first child. Couples in thecurrent study were married, on average, fornearly 6 years, and it is estimated that approx-imately 20% of couples in the U.S. divorce orseparate within 5 years of their first marriages(Blaisure & Geasler, 2006). Further, coupleshaving a second child are often different fromcouples deciding to have only one child. Forinstance, Dyrdal and Lucas (2013) recently re-ported that parents having a second child werehappier than parents having one child even be-fore the birth of their first child. More satisfiedspouses were also more likely to have childrenearlier in their marriage (Lawrence et al., 2008;Shapiro et al., 2000). Thus, studies using sam-ples with different background characteristicsand family structure may or may not find similarpatterns as those described here. Clearly, thereis a need for more research examining the tran-sition after the birth of a second child.

Some may argue that we had no matchedcontrol group of first-time parents to deter-mine whether any noted declines were due tothe transition or to normative, developmentalchanges in marital functioning. Our goal wasto understand heterogeneity in the transitionfrom one child to two, not to make compari-sons across first- and second-time parents.Besides, it is unlikely one could find amatched control group of one-child familiesbecause families choosing to have one childdiffer notably from families choosing to havetwo children, particularly with respect towomen’s career aspirations and employment(Knox & Wilson, 1978). The strategic timingof our assessments, specifically the 1-monthtime point, revealed significant changes afterthe transition, even if short-lived, that cannotbe easily explained as a normative develop-mental progression of marital functioning.The first contact with our families, however,

193MARITAL CHANGE AFTER SECOND CHILD

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

occurred during the last trimester of preg-nancy, which is standard practice in transitionto parenthood studies. Yet, one could arguethat the transition may have been well under-way by this time, perhaps from the momentthe couple knew the wife was pregnant. Thispossibility may be even more pronouncedwith the birth of the second child when par-ents spend a considerable amount of timeduring the pregnancy preparing the firstbornand themselves for the infant’s birth and thetask of caring for two young children (Rich-ardson, 1983).

The transition to parenthood marks a sig-nificant developmental milestone for manyadults. Researchers have spent decades exam-ining whether the birth of an infant spells ruinfor the marriage. Perhaps the question is notwhether marriages have been compromisedwith the birth of a child, but how adults makethe transition from partners to parents as partof the adult life course. In a 22-year longitu-dinal study of 3,672 German men and women,Dyrdal and Lucas (2013) found an increase inlife satisfaction (i.e., happiness) for husbandsand wives in pregnancy and the year afterbirth, with mothers more satisfied than fa-thers, suggesting that parenting brings joy tofamilies even if marital satisfaction declines.In addition, Ahlborg et al., (2009) reportedthat 79% of parents in their sample either hada second child or were pregnant with thesecond child by the time the firstborn was 4years old, underscoring that couples continueto have children beyond the first child despitedeclines in marital satisfaction.

The birth of the first child marks the begin-ning of parenthood, but certainly not the end.Few studies consider a life course develop-mental perspective when examining maritalchange after the birth of an infant. The currentstudy was our attempt at doing so by focusingon patterns of marital change for couples hav-ing their second child. As expected, there wassignificant heterogeneity in marital changepatterns, which often differed for husbandsand wives. Although there was evidence ofmarital disruption for some couples, mostcouples appeared to manage the transition andadapt to changes. Even when there was sig-nificant change, it was often short-lived, at-testing to family resilience rather than crisisin the face of change.

References

Ahlborg, T., Misvaer, N., & Möller, A. (2009). Per-ception of marital quality by parents with smallchildren: A follow-up study when the firstborn is 4years old. Journal of Family Nursing, 15, 237–263. http://dx.doi.org/10.1177/1074840709334925

Akaike, H. (1987). Factor analysis and AIC. Psy-chometrika, 52, 317–332. http://dx.doi.org/10.1007/BF02294359

Barber, J. S., & East, P. L. (2011). Children’s expe-riences after the unintended birth of a sibling.Demography, 48, 101–125. http://dx.doi.org/10.1007/s13524-010-0011-2

Beck, A. T., Ward, C. H., Mendelson, M., Mock, J.,& Erbaugh, J. (1961). An inventory for measuringdepression. Archives of General Psychiatry, 4,561–571. http://dx.doi.org/10.1001/archpsyc.1961.01710120031004

Belsky, J., & Hsieh, K. H. (1998). Patterns of maritalchange during the early childhood years: Parentpersonality, coparenting, and division-of-laborcorrelates. Journal of Family Psychology, 12, 511–528. http://dx.doi.org/10.1037/0893-3200.12.4.511

Belsky, J., & Rovine, M. (1990). Patterns of maritalchange across the transition to parenthood: Preg-nancy to three years postpartum. Journal of Mar-riage and the Family, 52, 5–19. http://dx.doi.org/10.2307/352833

Belsky, J., Spanier, G. B., & Rovine, M. (1983).Stability and change in marriage across the transi-tion to parenthood. Journal of Marriage and theFamily, 45, 567–577. http://dx.doi.org/10.2307/351661

Blaisure, K. R., & Geasler, M. J. (2006). Educationalinterventions for separating and divorcing parents.In M. A. Fine & J. H. Harvey (Eds.), Handbook ofdivorce & dissolution of romantic relationships(pp. 575–602). Mahwah, NJ: Erlbaum.

Bonds, D. D., Gondoli, D. M., Sturge-Apple, M. L.,& Salem, L. N. (2002). Parenting stress as a me-diator of the relation between parenting supportand optimal parenting. Parenting: Science andPractice, 2, 409–435. http://dx.doi.org/10.1207/S15327922PAR0204_04

Bower, D., Jia, R., Schoppe-Sullivan, S. J., Mangels-dorf, S. C., & Brown, G. L. (2013). Trajectories ofcouple relationship satisfaction in families withinfants: The roles of parent gender, personality,and depression in first-time and experienced par-ents. Journal of Social and Personal Relation-ships, 30, 389 – 409. http://dx.doi.org/10.1177/0265407512457656

Braiker, H. B., & Kelley, H. H. (1979). Conflict inthe development of close relationships. In R. L.Burgess & T. L. Huston (Eds.), Social exchange indeveloping relationships (pp. 135–168). New

194 VOLLING, OH, GONZALEZ, KUO, AND YU

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

York, NY: Academic. http://dx.doi.org/10.1016/B978-0-12-143550-9.50011-2

Costa, P. T., & McCrae, R. R. (1985). The NEOpersonality inventory: Manual, form S and form R.Lutz, FL: Psychological Assessment Resources.

Cowan, C., & Cowan, P. A. (1995). Interventions toease the transition to parenthood: Why they areneeded and what they can do. Family Relations,44, 412–423. http://dx.doi.org/10.2307/584997

Cowan, C., & Cowan, P. A. (2000). When partnersbecome parents: The big life change for couples.Mahwah, NJ: Erlbaum Publishers.

Cowan, C. P., Cowan, P. A., Heming, G., Garrett, E.,Coysh, W. S., Curtis-boles, H., & Boles, A. J., III.(1985). Transitions to parenthood: His, hers, andtheirs. Journal of Family Issues, 6, 451– 481.http://dx.doi.org/10.1177/019251385006004004

Cox, M. J., Paley, B., Burchinal, M., & Payne, C. C.(1999). Marital perceptions and interactions acrossthe transition to parenthood. Journal of Marriageand the Family, 61, 611–625. http://dx.doi.org/10.2307/353564

Crnic, K. A., & Greenberg, M. T. (1990). Minorparenting stresses with young children. Child De-velopment, 61, 1628 –1637. http://dx.doi.org/10.2307/1130770

De Haan, L., Hawley, D. R., & Deal, J. (2002).Operationalizing family resilience: A methodolog-ical strategy. American Journal of Family Ther-apy, 30, 275–291. http://dx.doi.org/10.1080/01926180290033439

Dew, J., & Wilcox, W. (2011). If momma ain’thappy: Explaining declines in marital satisfactionamong new mothers. Journal of Marriage andFamily, 73, 1–12. http://dx.doi.org/10.1111/j.1741-3737.2010.00782.x

Don, B. P., & Mickelson, K. D. (2014). Relationshipsatisfaction trajectories across the transition to par-enthood among low-risk parents. Journal of Mar-riage and Family, 76, 677–692. http://dx.doi.org/10.1111/jomf.12111

Doss, B. D., Rhoades, G. K., Stanley, S. M., &Markman, H. J. (2009). The effect of the transitionto parenthood on relationship quality: An 8-yearprospective study. Journal of Personality and So-cial Psychology, 96, 601–619. http://dx.doi.org/10.1037/a0013969

Dyrdal, G. M., & Lucas, R. E. (2013). Reaction andadaptation to the birth of a child: A couple-levelanalysis. Developmental Psychology, 49, 749 –761. http://dx.doi.org/10.1037/a0028335

Ehrenberg, M. F., Gearing-Small, M., Hunter, M. A.,& Small, B. J. (2001). Childcare task division andshared parenting attitudes in dual-earner familieswith young children. Family Relations, 50, 143–153. http://dx.doi.org/10.1111/j.1741-3729.2001.00143.x

Erel, O., & Burman, B. (1995). Interrelatedness ofmarital relations and parent-child relations: Ameta-analytic review. Psychological Bulletin, 118,108 –132. http://dx.doi.org/10.1037/0033-2909.118.1.108

Fincham, F. D., & Linfield, K. J. (1997). A new lookat marital quality: Can spouses feel positive andnegative about their marriage? Journal of FamilyPsychology, 11, 489 –502. http://dx.doi.org/10.1037/0893-3200.11.4.489-502

Flynn, H. A. (2010). Depression and postpartum dis-orders. In B. Levin & M. Becker (Eds.), A publichealth perspective of women’s mental health (pp.109–120). New York, NY: Springer Science Business Media. http://dx.doi.org/10.1007/978-1-4419-1526-9_6

Goldberg, A. E., & Perry-Jenkins, M. (2004). Divi-sion of labor and working-class women’s well-being across the transition to parenthood. Journalof Family Psychology, 18, 225–236. http://dx.doi.org/10.1037/0893-3200.18.1.225

Gordis, E. B., Margolin, G., & John, R. S. (2001).Parents’ hostility in dyadic marital and triadic fam-ily settings and children’s behavior problems.Journal of Consulting and Clinical Psychology,69, 727–734. http://dx.doi.org/10.1037/0022-006X.69.4.727

Haslam, D. M., Pakenham, K. I., & Smith, A. (2006).Social support and postpartum depressive symp-tomatology: The mediating role of maternal self-efficacy. Infant Mental Health Journal, 27, 276–291. http://dx.doi.org/10.1002/imhj.20092

Heavey, C. L., Larson, B. M., Zumtobel, D. C., &Christensen, A. (1996). The Communication Pat-terns Questionnaire: The reliability and validity ofa constructive communication subscale. Journal ofMarriage and the Family, 58, 796–800. http://dx.doi.org/10.2307/353737

Holmes, E., Sasaki, T., & Hazen, N. L. (2013).Smooth versus rocky transitions to parenthood:Family systems in developmental context. FamilyRelations, 62, 824–837. http://dx.doi.org/10.1111/fare.12041

Karney, B. R., & Bradbury, T. N. (1995). The lon-gitudinal course of marital quality and stability: Areview of theory, method, and research. Psycho-logical Bulletin, 118, 3–34. http://dx.doi.org/10.1037/0033-2909.118.1.3

Keizer, R., & Schenk, N. (2012). Becoming a parentand relationship satisfaction: A longitudinal dyadicperspective. Journal of Marriage and Family, 74,759 –773. http://dx.doi.org/10.1111/j.1741-3737.2012.00991.x

Kenny, D. A., Kashy, D. A., & Cook, W. L. (2006).Dyadic data analysis. New York, NY: GuilfordPress.

195MARITAL CHANGE AFTER SECOND CHILD

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

Kline, G. H., Julien, D., Baucom, B., Hartman, S.,Gilbert, K., Gonzales, T., & Markman, H. J.(2004). The interactional dimensions coding sys-tem: A global system for couple interactions. InP. K. Kerig & D. H. Baucom (Eds.), Couple ob-servational coding systems (pp. 113–126). Mah-wah, NJ: Erlbaum.

Kline, R. B. (2010). Principles and practice of struc-tural equation modeling (3rd ed.). New York, NY:Guilford Press.

Knox, D., & Wilson, K. (1978). The differencesbetween having one and two children. The FamilyCoordinator, 27, 23–25. http://dx.doi.org/10.2307/582722

Kouros, C. D., Cummings, E. M., & Davies, P. T.(2010). Early trajectories of interparental conflictand externalizing problems as predictors of socialcompetence in preadolescence. Development andPsychopathology, 22, 527–537. http://dx.doi.org/10.1017/S0954579410000258

Kreppner, K., Paulsen, S., & Schuetze, Y. (1982).Infant and family development: From triads totetrads. Human Development, 25, 373–391. http://dx.doi.org/10.1159/000272821

Krieg, D. B. (2007). Does motherhood get easier thesecond-time around? Examining parenting stressand marital quality among mothers having theirfirst or second child. Parenting: Science and Prac-tice, 7, 149 –175. http://dx.doi.org/10.1080/15295190701306912

Lawrence, E., Rothman, A. D., Cobb, R. J., Rothman,M. T., & Bradbury, T. N. (2008). Marital satisfac-tion across the transition to parenthood. Journal ofFamily Psychology, 22, 41–50. http://dx.doi.org/10.1037/0893-3200.22.1.41

Leahy-Warren, P., McCarthy, G., & Corcoran, P.(2012). First-time mothers: Social support, mater-nal parental self-efficacy and postnatal depression.Journal of Clinical Nursing, 21(3–4), 388–397.http://dx.doi.org/10.1111/j.1365-2702.2011.03701.x

Lindblom, J., Flykt, M., Tolvanen, A., Vänskä, M.,Thtinen, A., Tulppala, M., & Punamäki, R.-L.(2014). Dynamic family system trajectories frompregnancy to child’s first year. Journal of Mar-riage and Family, 76, 796–807. http://dx.doi.org/10.1111/jomf.12128

MacDermid, S. M., Huston, T. L., & McHale, S. M.(1990). Changes in marriage associated with thetransition to parenthood: Individual differences asa function of sex-role attitudes and changes in thedivision of household labor. Journal of Marriageand the Family, 52, 475–486. http://dx.doi.org/10.2307/353041

Masyn, K. E. (2013). Latent class analysis and finitemixture modeling. In T. D. Little (Ed.), The Oxfordhandbook of quantitative methods (Vol. 2): Statis-tical analysis (pp. 551–611). New York, NY: Ox-

ford University Press. http://dx.doi.org/10.1093/oxfordhb/9780199934898.013.0025

Mattson, R. E., Paldino, D., & Johnson, M. D.(2007). The increased construct validity and clin-ical utility of assessing relationship quality usingseparate positive and negative dimensions. Psy-chological Assessment, 19, 146–151. http://dx.doi.org/10.1037/1040-3590.19.1.146

McCrae, R. R., & Costa, P. T., Jr. (2004). A contem-plated revision of the NEO five-factor inventory.Personality and Individual Differences, 36, 587–596. http://dx.doi.org/10.1016/S0191-8869(03)00118-1

McCubbin, H. I., & Patterson, J. M. (1983). Thefamily stress process: The double ABCX model ofadjustment and adaptation. Marriage & FamilyReview, 6, 7–37. http://dx.doi.org/10.1300/J002v06n01_02

Miller, B. C., & Sollie, D. L. (1980). Normal stressesduring the transition to parenthood. Family Rela-tions, 29, 459 – 465. http://dx.doi.org/10.2307/584459

Mitnick, D. M., Heyman, R. E., & Smith Slep, A. M.(2009). Changes in relationship satisfaction acrossthe transition to parenthood: A meta-analysis.Journal of Family Psychology, 23, 848 – 852.http://dx.doi.org/10.1037/a0017004

Mortensen, Ø., Torsheim, T., Melkevik, O., &Thuen, F. (2012). Adding a baby to the equation.married and cohabiting women’s relationship sat-isfaction in the transition to parenthood. FamilyProcess, 51, 122–139. http://dx.doi.org/10.1111/j.1545-5300.2012.01384.x

Muthén, L. K., & Muthén, B. O. (1998–2010). Mplususer’s guide (6th ed.). Los Angeles, CA: Author.

Nylund, K. L., Asparouhov, T., & Muthén, B. O.(2007). Deciding on the number of classes in latentclass analysis and growth mixture modeling: AMonte Carlo simulation study. Structural EquationModeling, 14, 535–569. http://dx.doi.org/10.1080/10705510701575396

O’Brien, M., & Peyton, V. (2002). Parenting atti-tudes and marital intimacy: A longitudinal analy-sis. Journal of Family Psychology, 16, 118–127.http://dx.doi.org/10.1037/0893-3200.16.2.118

Patterson, J. M. (2002). Integrating family resilienceand family stress theory. Journal of Marriage andFamily, 64, 349–360. http://dx.doi.org/10.1111/j.1741-3737.2002.00349.x

Paulson, J. F., & Bazemore, S. D. (2010). Prenataland postpartum depression in fathers and its asso-ciation with maternal depression: A meta-analysis.JAMA: Journal of the American Medical Associa-tion, 303, 1961–1969. http://dx.doi.org/10.1001/jama.2010.605

Pinquart, M., & Teubert, D. (2010). A meta-analyticstudy of couple interventions during the transitionto parenthood. Family Relations, 59, 221–231.

196 VOLLING, OH, GONZALEZ, KUO, AND YU

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.

http://dx.doi.org/10.1111/j.1741-3729.2010.00597.x

Richardson, P. (1983). Women’s perceptions ofchange in relationships shared with children duringpregnancy. Maternal-Child Nursing Journal, 12,75–88.

Rutter, M. (1996). Transitions and turning pointsin developmental psychopathology: As appliedto the age span between childhood and midadult-hood. International Journal of Behavioral De-velopment, 19, 603– 626. http://dx.doi.org/10.1177/016502549601900309

Sanchez, L., & Thomson, E. (1997). Becoming moth-ers and fathers: Parenthood, gender, and the divi-sion of labor. Gender & Society, 11, 747–772.http://dx.doi.org/10.1177/089124397011006003

Schumm, W. R., Paff-Bergen, L. A., Hatch, R. C.,Obiorah, F. C., Copeland, J. M., Meens, L. D., &Bugaighis, M. A. (1986). Concurrent and discrim-inant validity of the Kansas marital satisfactionscale. Journal of Marriage and the Family, 48,381–387. http://dx.doi.org/10.2307/352405

Schwarz, G. (1978). Estimating the dimensions of amodel. Annals of Statistics, 6, 461–464. http://dx.doi.org/10.1214/aos/1176344136

Sclove, S. L. (1987). Application of model-selectioncriteria to some problems in multivariate analysis.Psychometrika, 52, 333–343. http://dx.doi.org/10.1007/BF02294360

Shapiro, A. F., Gottman, J. M., & Carrère, S. (2000).The baby and the marriage: Identifying factors that

buffer against decline in marital satisfaction afterthe first baby arrives. Journal of Family Psychol-ogy, 14, 59–70. http://dx.doi.org/10.1037/0893-3200.14.1.59

Stewart, R. B. (1990). The second child: Familytransition and adjustment. Newbury Park, CA:Sage.

Twenge, J. M., Campbell, W. K., & Foster, C. A.(2003). Parenthood and marital satisfaction: Ameta-analytic review. Journal of Marriage andFamily, 65, 574–583. http://dx.doi.org/10.1111/j.1741-3737.2003.00574.x

Volling, B. L. (2012). Family transitions followingthe birth of a sibling: An empirical review ofchanges in the firstborn’s adjustment. Psychologi-cal Bulletin, 138, 497–528. http://dx.doi.org/10.1037/a0026921

Wallace, P. M., & Gotlib, I. H. (1990). Marital ad-justment during the transition to parenthood: Sta-bility and predictors of change. Journal of Mar-riage and the Family, 52, 21–29. http://dx.doi.org/10.2307/352834

Wilkinson, R. B. (1995). Changes in psychologicalhealth and the marital relationship through child-bearing: Transition or process as stressor? Austra-lian Journal of Psychology, 47, 86–92. http://dx.doi.org/10.1080/00049539508257505

Received February 20, 2015Revision received July 3, 2015

Accepted July 10, 2015 �

197MARITAL CHANGE AFTER SECOND CHILD

Thi

sdo

cum

ent

isco

pyri

ghte

dby

the

Am

eric

anPs

ycho

logi

cal

Ass

ocia

tion

oron

eof

itsal

lied

publ

ishe

rs.

Thi

sar

ticle

isin

tend

edso

lely

for

the

pers

onal

use

ofth

ein

divi

dual

user

and

isno

tto

bedi

ssem

inat

edbr

oadl

y.