Does a House Divided Stand? Kinship and the Continuity of Shared Living Arrangements

JENNIFER E. GLICK Arizona State University

JENNIFER VAN HOOK The Pennsylvania State University∗

Does a House Divided Stand? Kinship and the

Continuity of Shared Living Arrangements

Shared living arrangements can provide hous-ing, economies of scale, and other instrumentalsupport and may become an important resourcein times of economic constraint. But the extentto which such living arrangements experiencecontinuity or rapid change in composition isunclear. Previous research on extended-familyhouseholds tended to focus on factors that trig-ger the onset of coresidence, including lifecourse events or changes in health status andrelated economic needs. Relying on longitudi-nal data from 9,932 households in the Surveyof Income and Program Participation (SIPP),the analyses demonstrate that the distribu-tion of economic resources in the householdalso influences the continuity of shared livingarrangements. The results suggest that multi-generational households of parents and adultchildren experience greater continuity in com-position when one individual or couple has a dis-proportionate share of the economic resourcesin the household. Other coresidential house-holds, those shared by other kin or nonkin,experience greater continuity when resourcesare more evenly distributed.

School of Social and Family Dynamics, 951 S. Cady Mall,Box 873701, Arizona State University, Tempe, AZ85287-3701 ([email protected]).

*Department of Sociology, 611 Oswald Tower, ThePennsylvania State University, University Park, PA 16802.

This article was edited by Jay Teachman.

Key Words: economic support, event history analysis, kin-ship, life transitions, living arrangements, multigenerationalrelations.

Social ties may buffer stressful life events,including unemployment, housing insecurity, orother shocks (Wright, Avshalom, Caspi, Moffitt,& Silva, 1998). Coresidence with kin or nonkinis one strategy employed by individuals wheneconomic resources are constrained. Almost onefifth of individuals in the United States live withextended family members (i.e., those other thanspouses and minor children) or nonkin (authors’tabulations from U.S. Current Population Surveydata, 2000 – 2008). The proportion of individu-als in these types of households is even higheramong the poor, disabled, and elderly. Suchcomplex living arrangements likely will increasein the near term as individuals adapt to economicand housing constraints resulting from the cur-rent economic recession. Although scholars havedocumented differences in living arrangementsover time and across groups, we have less of anunderstanding of how continuous (i.e., long last-ing) such complex living arrangements may be(Bethencourt & Rios-Rull, 2009; Rendell, 2011).We may expect that such living arrangementscontinue as long as needed, but there is someevidence that economic constraints not onlymotivate coresidence but also make those liv-ing arrangements somewhat tenuous (Menjivar,2000).

Extensive scholarship on the nature of house-hold composition and change demonstrated theimportance of the many societal and individualfactors that facilitate and discourage particu-lar living arrangements over time and acrossplace (Goody, 1976; Hammel & Laslett, 1974;Rossi & Rossi, 1990). In this paper, we focuson the continuity of living arrangements over

Journal of Marriage and Family 73 (October 2011): 1149 – 1164 1149DOI:10.1111/j.1741-3737.2011.00869.x

1150 Journal of Marriage and Family

time and ask whether some households are morelikely to stay together than others. We con-sider two potential patterns for the continuity ofcomplex living arrangements. First, with someimportant exceptions, much prior research onextended-family living arrangements seemed topresuppose that households fulfill a ‘‘privatewelfare’’ function in which the living arrange-ments are formed so that better-off people cancare for needy family members. The underlyingassumption of this perspective is that inequal-ity among household members (i.e., householdscontaining both the needy and a potential bene-factor) is a key condition that justifies, andindeed may sustain, extended or complex livingarrangements. For ease of explication, we refer tothis private welfare approach as the functionalistperspective on living arrangements.

An alternative set of theoretical perspectivescan be broadly termed the contractual per-spectives. Here, the expectation is that livingarrangements are formed in a more contractualmanner in which everyone is expected to con-tribute and benefit from the living arrangementmore or less equally. In other words, resourceswithin the household will be comparatively equi-table across individuals in order to allow forcompatible exchanges or contributions amonghousehold members. For households formedunder these expectations, inequality within thehousehold may encourage individuals to seekalternative living arrangements.

To assess these alternatives, we explorewhether households with an unequal distributionof resources—in other words, households con-taining a potential benefactor and the potentiallyneedy—stay together longer than householdswith more equitable resource distribution. Indoing so, we push beyond the prior literatureon extended-family living arrangements in threemajor ways. First, we gain insight about thenature of coresidential relationships by follow-ing complex households over time rather thanrelying on cross-sectional snapshots of livingarrangements. Second, to discern between thefunctionalist and contractual perspectives, weexplore how continuity of residence in multigen-erational or other complex households is relatednot only to the amount, but also the distribu-tion of resources across adults living in thesenonnuclear family households (Bethencourt &Rios-Rull, 2009). Third, we explore diversityacross households in these relationships. Wehypothesize that multigenerational households,

those shared by parents and adult children,willoperate in accordance with the functionalist per-spective, whereas households containing othertypes of kin and nonkin will operate more consis-tently with the contractual perspective. Overall,the results provide insights about the conditionsthat bind extended kin and nonkin together underthe same roof, thus helping to assess the likeli-hood of continuity of support among family andfriends through coresidence.

COMPETING MODELS TO EXPLAINCORESIDENCE

By their very nature, coresident relationshipsare complex and influenced by economic status,family ties, gender, and age as well as religiousor cultural practices (Coleman & Ganong, 2008;Goode, 1982; Goody, 1976; Hammel & Laslett,1974; Shuey & Hardy, 2003; Silverstein, Gans,& Yang, 2006; Spitze & Trent, 2006; Wellman,1990). Living arrangements may be influencedby dependencies and inequalities among cores-ident family members as well (Folbre, 2004;Liversage & Jakobsen, 2010; Menjivar, 2000).Our focus here is on the role of one formof inequality among household members—eco-nomic inequality—on the likelihood that cores-idential relationships are sustained over time.There are two theoretical models that providesomewhat different expectations for the wayeconomic inequality among household membersmay be associated with household continuity.

First, economic inequality could increase themotivation for shared living arrangements. Thiswould be consistent with research coming outof a functionalist theoretical perspective thatexplains extended-family living arrangementsin terms of their social and economic func-tions (namely, meeting the economic and careneeds of family members in a cost-effectiveway; Goode, 1982). For example, extended-family living arrangements have been shownto be strongly associated with significant lifecourse events (e.g., birth of a child, divorce,completion of schooling, death of a spouse,illness, retirement), low income, and the lackof other economic resources (Beresford &Rivlin, 1966; Burr & Mutchler, 1992, 1993;McGarry & Schoeni, 2000; Michael, Fuchs, &Scott, 1980; Mutchler & Burr, 1991; Pampel,1983; Wolf & Soldo, 1988). Further evidencefor this perspective has come from researchsuggesting that the formation of extended-family

Continuity of Shared Living Arrangements 1151

households was a sign of shifting economic for-tunes. Better-off relatives tended to coresidewith less-well-off kin in harder economic times,but live independently when less-well-off familymembers became better able to support them-selves again (Bethencourt & Rios-Rull, 2009;Peek, Koropeckyj-Cox, Zsembik, & Coward,2004; Stack, 1975). For example, coresidencewith parents declined over the 20th century asolder adults became more economically inde-pendent (Ruggles, 2007), largely through theintroduction of Social Security (McGarry &Schoeni, 2000).

The functionalist perspective identifiessocially enforced norms of obligation, such asfilial piety, as the primary reason people arewilling to assist needy family members throughcoresidence, even if the living arrangement is notobviously equitable or in everyone’s self-interestat the time. In other words, close kinship ties areexpected to garner altruism and unidirectionalassistance, for example, from parents to youngadult children or from middle-aged adults totheir aging parents (Coleman & Ganong, 2008).Family members may reciprocate this type ofsupport, but such reciprocity is not necessar-ily immediate or exchanged in kind (Rossi &Rossi, 1990).

The second, alternative perspective we iden-tify has tended to emphasize the contractualnature of shared living arrangements. In this con-tractual perspective, coresidential arrangementsare more likely to be maintained when all house-hold members contribute to and benefit from thearrangement more or less equally and simulta-neously. Menjivar (1997) highlighted this ideain her ethnographic study of Vietnamese, Sal-vadoran, and Mexican immigrants in San Fran-cisco. She found that the kinship networks (andextended-family living arrangements) among theVietnamese and Mexican immigrants offeredmore assistance and were longer lasting thanthose among Salvadoran immigrants. Menji-var’s (1997) explanation was that the steadysupply of resources available to the Vietnamesethrough refugee support and the communityresources among Mexican immigrants helpedfoster stronger kinship ties through the exchangeof resources and practices of reciprocity. This isconsistent with other research suggesting that inorder for extended-family households to be ben-eficial for participants, available coresidentialkin must be able to provide either economicsupport or care for dependent family members

so that the labor supply in the household canincrease (Hogan, Hao, & Parish, 1990; Tienda& Glass, 1985). An underlying assumption ofthe contractual perspective is that the stabil-ity of relationships, even family relationships,depends on the practice of ‘‘balanced reci-procity’’ (Sarkisian & Gerstel, 2004). Whenreciprocity cannot be maintained, these relation-ships are weakened and may become less tenableover time (Plickert, Cote, & Wellman, 2007).

Here we evaluate the contradictory expec-tations of the functionalist and contractual per-spectives by examining the relationship betweenhousehold income inequality and the continuityof shared living arrangements. The functional-ist perspective predicts that household incomeinequality leads to longer-lasting coresidentialliving arrangements. This will be particularlytrue for households that contain both individ-uals with resources and those who have morelimited resources, which may promote exchangeacross units or enhance the normative obliga-tions family members may feel towards kin withfewer resources than they possess (Coleman &Ganong, 2008). But from a contractual perspec-tive, income inequality is expected to underminethe cohesion of intrahousehold relationships andthus lead to less continuity of shared livingarrangements over time. Although it may also beinstructive to examine inequality in the exchangeof nonmonetary resources, we focus here on theextent to which income inequality serves as anindicator of the conditions that would supportthe practice of ‘‘balanced reciprocity’’ for thereasons discussed above. There is a pragmaticreason for focusing on income inequality as well:It can be consistently measured across a largerpopulation than has been previously assessed insmaller studies.

Although the prevalence of extended-familyhouseholds in the United States is substantial, wecurrently know very little about the continuityof these living arrangements over time. A lackof longitudinal data has meant that researchershave had to rely on the simple prevalence ofshared households as indicators of supportiveties. As a consequence, few studies go beyondcross-sectional analyses of household structureto observe how financial need or inequalityis related to household continuity (Smits, vanGaalen, & Mulder, 2010). Our major contribu-tion is that we gain insight about the natureof coresidential relationships by exploring thefactors that affect their persistence over time.


In doing so, we explore how the continuity ofshared living arrangements is related not only tothe amount but also the distribution of resourcesacross adults in the same household.

Does the Applicability of the Functionalist andContractual Models Depend on Relatedness?

Not all coresidential households are the same,and it seems likely that the role potential reci-procity plays in the continuity of these livingarrangements could depend on the type of rela-tionships involved. Although some of thesehouseholds may have been formed and oper-ate to fulfill a ‘‘private welfare’’ function (thefunctionalist perspective), others may dependon the practice of balanced reciprocity fortheir persistence (the contractual perspective).Drawing on theories developed to understandunion stability, we hypothesize that multi-generational households of parents and adultchildren correspond more closely with thefunctionalist perspective and that those con-taining other kin or nonkin have dynamicsthat are more consistent with the contractualperspective.

Economic theories and research on union sta-bility distinguish between marriage and cohabi-tation. This distinction is instructive because ofthe parallels between the more formalized rela-tionships of marriage and those expected frommultigenerational ties among parents and theirdescendants, as compared to the less formalizedrelationships involving cohabiting partners andthose of other kin such as siblings or cousins,or even relationships among nonkin, includingfriends and more distant acquaintances. Priorresearch showed that marital unions in theUnited States are destabilized when both spousesmake similar economic contributions (Becker,1981; Kalmijn, Loeve, & Manting, 2007;Oppenheimer, 1997), consistent with the expec-tations of the functionalist perspective. Inter-estingly, cohabiting unions have been foundto display an opposite pattern. Income equal-ity tended to increase the continuity of theshared living arrangement among cohabiters(Brines & Joyner, 1999), which is more con-sistent with the contractual perspective. Brinesand Joyner suggested that legal and norma-tive ties may encourage married couples totake the risks necessary to specialize andengage in the exchange of unlike goods andservices, which then reinforced the dependency

of spouses on one another and the stability ofthe union. But the contingent nature of cohabit-ing relationships discouraged specialization andinstead rewarded equality and exchange of likegoods (Brines & Joyner, 1999; Kalmijn et al.,2007).

To extend these ideas to other coresiden-tial relationships, we consider variations in thestrength of legal and normative ties amongkin and nonkin. Normative obligations mostoften extend to kin relationships before nonkinand to close kin before distant and extendedkin (Bianchi, Hotz, McGarry, & Seltzer, 2007;Burton & Stack, 1993; Ganong & Coleman,2006; Silverstein, Conroy, Wang, Giarrusso,& Bengston, 2002; Taylor, Chatters, & Mays,1988). Parents and their dependent children, likespouses, have considerable obligations to oneanother both informally and legally (Swartz,2009). Much of the research on living arrange-ments has focused on multigenerational cores-idence of parents and their adult children (i.e.,Burr & Mutchler, 1999; Choi, 2003; Ruggles,2007, Silverstein et al., 2006) and the extent towhich parents and their adult children rely onone another for instrumental and social support.Given that relationships among these familymembers may be guided by clearly definedsocial norms of responsibility, we may expectthat a household shared by parents and adultchildren will garner a greater willingness toprovide resources to family members, regard-less of potential for immediate reciprocation,than households with other types of kinship ties(Goode, 1982; Hamilton, 1964). In other words,we expect that multigenerational households ofparents and their adult children or grandchildren,or both, will operate in ways that, like marriages,are consistent with the functionalist perspective.We therefore hypothesize:

Hypothesis 1: Multigenerational extended-familyhouseholds will retain their structure wheneconomic inequality is high in comparison to thosehouseholds where the economic status is balancedacross household members.

But social norms for behavior and altruismare less clearly defined in the case of rela-tionships with other kin or nonkin (Coleman& Ganong, 2008; Rossi & Rossi, 1990). Likecohabiting partners, siblings and friends maybe important sources of social and instrumentalsupport, but their role is less clearly prescribed


and their contact over the life course may beviewed as contingent on the quality of the rela-tionship and treated in more voluntary ratherthan obligatory terms (Spitze & Trent, 2006;Stewart-Williams, 2007; White, 2001). In theabsence of institutional regulation or normsthat monitor long-term relationships, unbalancedresources would represent too great of a riskof nonreciprocity (Molm, Collett, & Schaefer,2007). In other words, we expect their house-holds, like cohabiting unions, to conform to theexpectations of the contractual perspective. Wetherefore further hypothesize:

Hypothesis 2: Coresidential households other thanmultigenerational extended-family households,such as those shared by siblings, other extendedkin, or nonkin, will have lower continuity ofmembership when economic inequality is highthan those households where the economic statusis balanced across household members.

There are, of course, other factors that mayencourage continuity of shared living arrange-ments over time, and our analyses take manyof these into consideration in addition to ourprimary focus on the relatedness and distribu-tion of economic resources among householdmembers. For example, the overall level ofresources in the household may be importantbecause lower levels of income could add stressto households regardless of composition (Burton& Stack, 1993; Menjivar, 1997, 2000; Portes& Sensenbrenner, 1993; Roschelle, 1997) andreduce incentives to remain in the same livingarrangement. Likewise, the dependent care andhealth needs of household members are likely tobe associated with living arrangements (Hogan,Eggebeen, & Clogg, 1993; Martikainen, Nihh-tila, & Moustgaard, 2008; Speare & Avery,1993). Previous research also demonstratedhigher prevalence of extended-family house-holds among some racial and ethnic groupswhen compared to non-Hispanic Whites as wellas among recent immigrants when compared tothose of longer duration in the United Statesor those born in the United States (Van Hook& Glick, 2007). The analyses presented belowtake all of these characteristics into accountwhen examining the role of the relative dis-tribution of economic resources on householdcontinuity.

METHOD

Data

We use the pooled 1990, 1991, 1992, 1993, and1996 panels of the Survey of Income and Pro-gram Participation (SIPP), a longitudinal survey,to study the association between intrahouseholdincome inequality and the continuity of livingarrangements in the United States. Each of thefive SIPP panels includes a separate, indepen-dent sample that is interviewed every 4 monthsfor roughly 3 to 4 years. For example, the 1990Panel includes individuals who were interviewedup to eight times over a period of 32 monthsstarting in 1990, and the 1991 panel includes anentirely new sample that was interviewed up toeight times over a period of 32 months startingin 1991. The respondents in the 1992, 1993, and1996 panels were interviewed every 4 monthsover 40, 36, and 48 months, respectively. Crucialto our study, the SIPP follows individuals overtime even if they leave their original householdsand form new ones and includes time-varyinginformation on living arrangements and standardsocial, demographic, and economic variables.

By combining five SIPP panels, we amass asufficiently large sample to examine the durationof extended-family or nonkin households thatcontain at least one adult age 18 or older atfirst interview (10,224 households). Using thepooled data, we construct a longitudinal datafile that includes an observation for each time ahousehold is followed up or until the householdexperiences a transition in its structure as definedbelow. We drop the first interview because weuse lagged variables in our analysis and mostof our lagged variables are unobserved for thefirst interview. This leaves 40,954 household-interview observations. The U.S. Census Bureauuses the ‘‘hot deck’’ method to impute missingvalues for most but not all variables in the SIPP.We dropped 292 households (2.9%) and 1,256(3.1%) household-interviews from the analyticsample with nonimputed missing values on oneor more of the analytic variables, leaving afinal sample of 9,932 households and 39,698household-interviews.

Even though the SIPP offers unique advan-tages, it suffers the disadvantage of moderatelyhigh attrition rates. In a typical SIPP panel, 18%to 22% of the original sample drops out duringthe course of the study period; roughly half ofall attrition occurs between the first and secondinterviews (Jabine, 1990). Even large amounts


of attrition will not bias results unless cases fallout of the sample in a nonrandom manner suchthat attrition is significantly associated with theerror term (Fitzgerald, Gottschalk, & Moffitt,1998). Attrition rates (i.e., the percentage thatleft the sample before the end of the study)were significantly higher for Blacks (31.2%)and other racial minority groups (34.6%) thanfor non-Hispanic Whites (20.8%), but werenot significantly different between multigenera-tional households (21.7%) and other coresiden-tial households (23.9%). Attrition rates did notdiffer significantly by household structure withinracial and ethnic groups.

Extended-Family and Nonkin HouseholdStructure

We define household structure based on thenumber and relationships among minimalhousehold units (MHU). The MHU refers tosmaller identifiable units within householdsbased on marriage or cohabitation and parentageof minor children (Biddlecom, 1994; Ermisch &Overton, 1985; Glick, Bean, & Van Hook, 1997;Glick & Van Hook, 2002). For our purposes,married or cohabiting couples and parents withunmarried, childless children younger than 25are defined as belonging to the same MHU.Young adults aged 24 and younger who aremarried or cohabiting or have children of theirown are classified in their own MHU withtheir spouse or partner and children, if any. Inaddition, single adults age 25 or above withoutminor children make up an individual MHU.Finally, minor children not living with a parent(such as foster children) are classified in thesame MHU as the householder. For example,a household containing a man, his wife, hismother-in-law, his 20-year-old single, childlessdaughter, and a boarder would include threeseparate MHUs: (a) the man, wife, and daughter,(b) the mother-in-law, and (c) the boarder. If thedaughter had a young child or was married, shewould be put in a separate MHU along with herchild and husband, if any.

We identify coresidential living arrangementsas households containing at least two MHUs.Among these households, we distinguish multi-generational households—those that containMHUs from multiple generations, such as house-holds including adult children and their elderlyparents—from all other types of extended-familyor nonkin households. Following prior work,

we infer the relationships among MHUs basedon relationship to the householder (Coward,Cutler, & Schmidt, 1989; Glick et al., 1997;Schmertmann, Boyd, Serow, & White, 2000).For example, in a household with three MHUs,if the first MHU head was the householder, thesecond was the child of the householder, andthe third was the sibling of the householder,we would code the third MHU as the uncle oraunt of the second MHU. The entire householdwould be classified as multigenerational becauseit contains a parent – child relationship betweenthe first and second MHUs. Seventy-four per-cent of the extended-family households in oursample are multigenerational.

Household structure and living arrangementsare treated as time-varying across the SIPPpanel. Among households that are ever classifiedas coresident during the SIPP study period,we model household change. These transitionscan involve either changes in the householdcomposition without a transition to a simplehousehold structure or transitions to a simplehousehold structure (i.e., no longer living withextended-family or nonkin). To distinguishchanges in the household composition arisingfrom birth, adoption, or changes in marital statusfrom other types of change, we do not countchanges arising from additions or departures ofchildren under age 15 or of spouses as changesin the household roster.

Duration of Ongoing Spells

We must account for the length of time extended-family or nonkin living arrangements haveexisted. For coresidential households that wereformed during the SIPP panel, it is straight-forward to measure the duration of the liv-ing arrangement (in months since formation).For coresidential households that were formedbefore the first interview, we use retrospectivedata on place of residence to deduce when the liv-ing arrangement was formed. The SIPP includesthe month and year that each person age 15 orolder moved into the household. We use thisinformation to reconstruct households back intime in order to estimate how long adult (age25 or older) family members had been livingtogether. About half of the ongoing spells were10 years or less in duration at the time of thefirst interview (M = 14 years). The start timeestimates are only approximations and proba-bly underestimate the duration of coresidential


spells because they do not incorporate time spentin the household by MHUs who moved awayprior to the first interview.

Intrahousehold Income Distribution

To measure the distribution of economicresources within the household at each inter-view, we use a measure that stems from DeMarisand Longmore’s (1996) study of couple-levelexchange. In a household containing n MHUs,each MHU can be paired with (n − 1) otherMHUs in the household. For each MHUj , weestimated income contribution (Cj) as the ratioof its income (Ij) to the total income from allsources of each MHU pair, averaged across allpairs:

Cj =∑n

k=1,k �=j

(Ij

Ij +Ik

)

n − 1

For example, if MHU a makes $100, bmakes $50, and c makes $0, then the averagecontribution would be .83 for a ((100/150 +100/100)/2); .67 for b; and 0 for c. If incomewere distributed equally across MHUs in thehousehold and each MHU made $50, then theaverage contributions of each MHU would be.50. Conversely, if MHU a contributed all of theincome, the average contribution would be 1.0for a and 0 for b and c. We then subtracted 0.5from the MHU-level indicators of contribution(i.e., the difference from an equal contribution),squared it (to eliminate negatives), and averagedacross all MHUs to obtain a household measureof inequality (Qi):

Qi =∑n

k=1 (Ck − .5)2

n

In the case of the first household describedabove, Q = .129. In the case of the householdin which all three MHUs contribute equally,Q = 0. For households in which one MHUcontributes all of the income, Q = .25. Thus, ourtime-varying measure of intrahousehold incomeinequality ranges from 0 to .25 (M = 0.12,SD = .11).

Other Variables

In our multivariate analysis of the continuityof coresidential living arrangements, we control

for several factors that may be associated withhousehold transitions. All are permitted to varyover time. We take into account the size andselected characteristics of the household. Wecontrol for number of adults and the ratio ofthe household income to the federal povertythreshold. We adjust for potential needs fordependent care with dummy variables indicatingthe presence of certain types of persons inthe household: at least one elderly person age65 or older (41.5% of sample households), atleast one child age 0 – 4 (14.7%), at least oneperson who is divorced or widowed (57.3%),and at least one unmarried parent of a minorchild (age 0 – 17; 5.9%). Because life coursetransitions may be associated with changes inliving arrangements, we further control for thenumber of children born to household members,the number of adults experiencing a change inmarital status, and the number of children whoturned 18 since the previous interview. We alsoinclude measures for the nativity compositionof the household. We contrast households withnew immigrants (i.e., in the United States forfewer than 10 years) to those with immigrantswhose arrival in the United States occurred 10or more years ago and those with no immigrants.

Finally, we account for characteristics of thehouseholder in order to control for the ability(or opportunity) of the householder to take inkin: age, gender, educational attainment (lessthan high school, high school, some college, andcollege or more), number of children ever borneor fathered, and race and ethnicity. These factorshave been observed to be associated with livingarrangements and kin availability.

Data Analysis

We analyze the (dis)continuity of coresidentialliving arrangements using life tables fordescriptive analyses and discrete-time eventhistory models for multivariate analyses. The lifetables are used to estimate the probability thatcoresidential households experience a transitionin composition within 1, 3, and 5 years since theformation of the living arrangement (Preston,Heuveline, & Guillot, 2001). The life tableis a model that simulates reductions overtime in the size of a synthetic cohort of1,000 coresidential households. Decrementsfrom the synthetic cohort occur as householdsexperience a transition either by transitioningto a simple household or experiencing change


in the household composition. The probabilityof household transition is calculated directlyfrom the SIPP data as the proportion makingthe transition by duration of coresidence amongthose that had not made an earlier transition.The life tables are estimated separately byliving arrangement (multigenerational and othercoresidential households) and income inequality(above and below the median).

We also estimate discrete-time event historymodels (Allison, 1995; Guo, 1993) of householdtransition. These are well suited for modelingthe timing of events like death, divorce, orany other type of transition. Just like lifetables, they divide the time at risk into smallintervals (e.g., months, years), and then modelthe probability of experiencing the event withineach interval, given that the event has notalready occurred and conditional on the durationof time already at risk. As described above,our analytic sample includes one observationfor each 4-month period between interviewsuntil and including the interview at which thehousehold roster changes (apart from changesdue to births or adoptions) or is right-censored.We use logistic regression to model the loggedodds of transition in living arrangement (Lit )within each time interval as a function of theduration of the spell (Dt ) and lagged household(Ht−1) characteristics (measured at time t − 1):

Ln(Lit/(1 − Lit )) = α + φDt + δHit−1

We use modeling procedures designed by StataCorporation (1997; also Levy and Lemeshow,1999) to take into account clustering withinsampling strata and PSUs.

RESULTS

Table 1 provides means or percentage distribu-tions for all of the variables used in the analysisfor the entire sample of households as well asseparately for multigenerational and other cores-idential households. Multigenerational house-holds were more likely to include elderly anddivorced or widowed persons and tended tohave older, less educated female householderswith more children than other households. Othercoresidential households were more likely toinclude recent immigrants and to have male andHispanic householders.

Table 2 presents the life table analysis ofthe proportion of households with consistent

composition over time. These estimates sug-gest that approximately 38% of all householdsremained intact after 1 year, 12% were intactafter 3 years, but only 7% retained the same com-position at the end of 5 years. This means that themajority of households experienced a transitionin the first year and very few remained with thesame composition continuously beyond 3 years.

We estimate household continuity for theupper and lower halves of the sample by incomeinequality (above and below about .12) and findsmall differences in the percentage remainingtogether across time periods; those in less equalhouseholds were slightly more likely to haveremained together after 3 and 5 years. When wedisaggregate the sample by household structure,larger differences emerge that are consistentwith our theoretical expectations. Multigener-ational households were much more likely toremain together than other types of coresidentialliving arrangements (45% vs. 25% of non-multigenerational households remained togetherafter 1 year). Furthermore, among multigener-ational households, income inequality is asso-ciated with a greater probability of retaininghousehold structure than among those withequitable distributions of income across units.Among other coresidential households, incomeinequality is associated with a lower proba-bility of retaining household structure. This isalso depicted in Figure 1, which graphs the pro-portion of households estimated to retain thesame composition by duration of living arrange-ment. The sharp drop indicates less continuityin household composition in the first year withfewer transitions after 3 years.

The results presented in Figure 1 providesome initial support for our hypotheses. First,the probability of continuity in multigenerationalextended-family households was greater thanin the other coresidential households. Further,we observe a different pattern based on thedistribution of resources in the household.Multigenerational households with more equalincome distribution across household memberswere more likely to experience a change incomposition than those with less equitabledistributions of income. The opposite patternis observed among the other coresidentialhouseholds. Here the potential for balancedreciprocity appears associated with increasedcontinuity of composition.

To test these ideas further, we use multivariatediscrete-time hazard models of household


Table 1. Sample Means or Percentages With Standard Deviations, All Coresidential Households (N = 9,932 Householdsand 39,698 Observations)

All Coresidential HHs Multigenerational HHs Other Coresidential HHs

Variables M or % SD M or % SD M or % SD

HH transitioned since lastinterview (%)

12.6 11.3 16.2

Multigenerational HH (%) 73.6 100.0 0.0Income inequality 0.120 0.107 0.133 0.106 0.084 0.100Duration of living

arrangement (years)12.1 13.0 10.6 10.9 16.2 17.0

Number of adults in HH 2.8 1.0 3.0 1.0 2.4 0.9Income-to-poverty ratio 3.9 2.8 4.0 2.8 3.7 2.9HH contains immigrant

newcomers (0 – 9 years inUnited States) (%)

3.0 2.1 5.6

HH contains otherimmigrants (%)

6.2 6.5 5.4

At least one HH member is:Age 65 or older (%) 41.5 46.6 27.3Has child age 0 – 4 (%) 14.7 15.5 12.6Divorced or widowed (%) 57.3 58.9 52.9Single parent (%) 5.9 5.9 5.8

Life course transitions in HHsince last interview:No. of persons whochanged marital status

0.017 0.146 0.017 0.142 0.018 0.156

Births 0.040 0.363 0.046 0.390 0.024 0.274No. of persons who turnedage 18

0.014 0.184 0.014 0.186 0.015 0.180

Characteristics of HHreference personAge 55.6 15.3 59.0 13.0 46.0 17.3Male (%) 58.1 59.2 54.9

EducationHigh school graduate (%) 31.1 32.0 28.4Some college (%) 24.7 23.5 28.2College graduate (%) 18.8 16.2 26.2Number of children borneor fathered

2.4 1.9 2.9 1.8 0.9 1.3

Race/ethnicityBlack (%) 7.0 7.7 4.9Mexican (%) 2.4 2.3 2.5Other Hispanic (%) 2.1 2.1 2.3Asian (%) 1.7 1.7 1.7

n observations 39,698 29,669 10,029n households 9,932 7,029 2,903

Note. The data are the 1990, 1991, 1992, 1993, and 1996 SIPP panels.The sample includes all coresidential households.Each household contributes one observation per interview until either the end of the survey or the household dissolves orotherwise changes composition. HH = household.


Table 2. Continuity of Coresidential Living Arrangements by Income Distribution and Household Structure (N = 9,932Households and 39,698 Observations)

Proportion Remaining in Living Arrangement After

1 year 3 years 5 years

All coresidential HHs 0.383 0.124 0.065Less equal income distributiona 0.395 0.137 0.071More equal income distributionb 0.370 0.114 0.059

Difference: less − more equal 0.025 0.023∗ 0.111∗

Multigenerational HHs 0.447 0.163 0.097Other coresidential HHs 0.255 0.053 0.016Difference: multigenerational − other 0.191∗ 0.109∗ 0.081∗

Multigenerational HHsLess equal income distribution 0.462 0.178 0.104More equal income distribution 0.427 0.147 0.088

Difference: less − more equal 0.034 0.031∗ 0.015∗

Other coresidential HHsLess equal income distribution 0.223 0.038 0.007More equal income distribution 0.281 0.063 0.022

Difference: less − more equal −0.058∗ −0.025∗ −0.015∗

Note: See Table 1 for description of data and sample. HH = household.aUpper half of sample on HH inequality scale. bLower half of sample on HH inequality scale.∗Difference is statistically significant (p < .05).

change that take into account the other factorsassociated with the probability that a householdhas continuous composition or experiences achange in composition. Results are displayed inTable 3. Unlike the life table estimates, whichthus far provide a picture of duration in thesame household (i.e., continuity), these modelspredict changes in household composition. Thus,positive coefficients indicate a greater likelihoodof household change and negative coefficientsindicate a lower likelihood.

Model 1 includes all of the independentvariables, including duration of living arrange-ment, income inequality, household structure,race/ethnicity, and all of the control variables.Duration of living arrangement is associatedwith lower probability of transition (i.e., thelikelihood of transition was highest amongnewly formed living arrangements and declinedover time). And larger households, not sur-prisingly, experienced a greater likelihood of achange in composition. Somewhat surprisingly,we observe no significant effect of householdpoverty on the probability that a household expe-rienced change. Households containing recentimmigrants, persons age 65 or older, young chil-dren, and widowed or divorced persons weremore likely to experience a change and those

including single parents were less likely. Lifecourse transitions occurring in the household,such as a child reaching age 18, are also associ-ated with a transition in household composition.

There is some variation in the timing of house-hold change when we look at the householder’scharacteristics. Householder’s age is associ-ated with a decreased probability of householdchange. Households with male householderswere slightly more likely to experience a changein composition. We observe considerable racialand ethnic variation in household transitions aswell. Black householders and householders ofMexican and Asian origin were less likely toexperience a change in composition than non-Hispanic Whites.

To test our hypotheses, Model 1 includes themeasure for household type and income inequal-ity. As predicted, multigenerational extended-family households were less likely to experiencetransition in the observed time period. Overall,however, income inequality is not significantlyassociated with transitions. The next step inour analysis is to determine whether the effectsof inequality differ by type of household assuggested by the descriptive results.

To this end, Model 2 adds an interactionterm between income inequality and household


FIGURE 1. PROBABILITY OF CONTINUITY OF CORESIDENTIAL LIVING ARRANGEMENT BY DURATION.

structure comparing the multigenerationalhouseholds to others. The interaction effect issignificant and negative (p < .001). To interpretthe results, Figure 2 graphs predicted probabil-ities of transitions by inequality and householdstructure. The predicted probabilities are gen-erated by substituting mean values of all thecontrol variables into the estimated model,but varying values for inequality and house-hold structure. The resulting estimates—pre-dicted logged odds—are then converted topredicted probabilities (DeMaris, 1992). Theresults clearly show that income inequalityis associated with an increased likelihood ofchange among nonmultigenerational coresidenthouseholds (p < .001). Income inequality is notsignificantly associated with household changeamong multigenerational households (the slopeappears negative but postestimation tests showthat this association is not significant; p = .13).In other words, consistent with the findings fromFigure 1 but now adjusting for the compositionof the different households, the association of

income inequality with the continuity of livingarrangements depends on the relationships ofthose sharing the household.

DISCUSSION

The results of our analyses demonstrate the highfrequency of moves among those in multigener-ational and other coresidential living arrange-ments. These living arrangements change incomposition frequently as kin and nonkin moveinto and out of shared households. Some ofthese changes are triggered by normative lifecourse events such as marital transitions, births,and entering adulthood. Nonetheless, sharinga household with people other than parents,adult children, or grandchildren is associatedwith an even greater probability of transitionthan sharing a household with other multigen-erational kin; these results persist even in thepresence of controls for the life course anddependency characteristics that may be expectedto trigger the departure from or persistence in


Table 3. Logistic Regression Models of Transition Among Coresidential Households (N = 9,932 Households and 39,698Observations)

Model 1 Model 2

Predictor β OR β OR

Multigenerational HHa −0.51∗∗∗ 0.60 −0.31∗∗ 0.74Income inequalityb 0.32 1.38 1.56∗ 4.74Income inequality × multigenerational HH −1.78∗∗ 0.17Duration of living arrangement (years) −0.14∗∗∗ 0.87 −0.14∗∗∗ 0.87Duration of living arrangement (squared) 0.00∗∗∗ 1.00 0.00∗∗∗ 1.00No. of adults in HH 0.12∗∗ 1.12 0.11∗∗ 1.12Income-to-poverty ratio 0.01 1.01 0.01 1.01HH contains immigrant newcomersa 0.15 1.16 0.13 1.14HH contains other immigrantsa 0.08 1.08 0.08 1.08At least one HH member is

Age 65 or oldera 0.36∗∗∗ 1.44 0.39∗∗∗ 1.44Has child age 0 – 4a 0.18∗ 1.19 0.18∗ 1.20Divorced or widoweda 0.38∗∗∗ 1.46 0.37∗∗∗ 1.44Single parenta −0.14 0.87 −0.14 0.87

Life course transitions occurring in the HH since the last interviewNumber of persons who changed marital status 2.09∗∗∗ 8.12 2.09∗∗∗ 8.10Births 0.03 1.03 0.03 1.03Number of persons who turned age 18 0.68∗∗∗ 1.97 0.68∗∗∗ 1.98

Characteristics of HH reference personAge −0.02∗∗∗ 0.98 −0.02∗∗∗ 0.98Malea (reference = female) 0.13∗ 1.14 0.13∗ 1.13

Education (reference = not a high school graduate)High school graduatea −0.05 0.95 −0.05 0.95Some collegea −0.05 0.95 −0.05 0.95College graduatea −0.07 0.93 −0.05 0.95Number of children borne or fathered 0.08∗∗∗ 1.08 0.08∗∗∗ 1.08

Race/ethnicity (reference = non-Hispanic White)Blacka −0.14∗ 0.87 −0.13∗ 0.87Mexicana −0.28∗∗ 0.76 −0.27∗ 0.77Other Hispanica −0.09 0.91 −0.09 0.91Asiana −0.17 0.84 −0.18 0.84

Intercept −0.80∗∗∗ 0.45 −0.85∗∗∗ 0.43N 39,698 39,698

Note: See Table 1 for description of data and sample. OR = odds ratio.aCoded as 1 = yes and 0 = no. bIncome inequality ranged from 0 (low inequality) to .25 (high inequality).∗p < .05. ∗∗p < .01. ∗∗∗p < .001.

these households. Distant kin or friends mayshare living arrangements temporarily, but thissort of coresidence is often short lived (Fertig &Reingold, 2008).

In addition, we find partial support forour hypotheses, based on a functionalist per-spective, that multigenerational households areless likely to transition when resources areinequitably distributed. In fact, we find nosignificant association between intrahousehold

income inequality and the continuity of multi-generational living arrangements, but the rela-tionship ran in the expected direction. Thismay derive from the relative needs of house-hold members for in-kind assistance as wellas normative expectations for coresidence. Forexample, elderly household members may notbe expected to work, so households contain-ing elderly parents and their adult children mayretain their composition over time even with


FIGURE 2. PROBABILITY OF HOUSEHOLD TRANSITION BY HOUSEHOLD STRUCTURE AND INCOME INEQUALITY (BASED ON

TABLE 3, MODEL 2).

greater income inequality. Likewise, householdswith many young children may contain adultswho do not contribute earned income in order toprovide child care. In other words, these house-holds may contain more opportunities to provideimmediate reciprocity in the form of in-kindexchanges such as care for dependent fam-ily members. Unbalanced economic resourcesin these situations may be more normativelyacceptable than if the household were com-prised entirely of working-age adults. This isalso consistent with the idea that close kin rela-tionships are guided by norms for reciprocitythat carry out over the life course (Silversteinet al., 2002).

The opposite is the case for householdswith other types of kin and nonkin. Thesehouseholds are much more likely to experiencea change in composition in the absence ofbalanced resources (with the greater risk ofnonreciprocity) as suggested by the contractualperspective. Thus, it appears that householdsmade up of siblings or more distantly relatedkin or nonkin are likely to be more sensitive tothe economic resources of household members.

These results are consistent with the economictheoretical perspectives on union stability inwhich relationships with fewer prescribednorms carry greater expectations for similarcontributions from all parties.

One of the primary limitations of our analysesis that we can only measure the balanceof economic resources between individuals orcouples in the household. As noted, individualsmay provide in-kind services such as childcare or household labor in exchange forinstrumental support, including housing. Inaddition, we cannot observe how economicresources are distributed within the MHUs oreven across minimal household units. Thus, wecannot observe whether those bringing economicresources to the household are able to controlhow those resources are spent or distributed.Another related limitation in our analyses is thatwe can only observe coresidential relationshipsand not those relationships with individualsoutside the household. Kin and nonkin networksextend beyond the household and resourcesand assistance may flow between as wellas within households. But the formation and


continuity of the extended-family householdprovides an excellent opportunity to observethe mobilization of resources in the form ofshared housing (Aquilino, 1990; De Vos, 2000;Ruggles, 2007). Despite data limitations thatpreclude a direct test of the extent to whichresources are directly exchanged within orbeyond households or whether other factorssuch as preferences or power differentials amongindividuals in the households also influence theduration of shared living arrangements (Bianchiet al., 2007; Folbre, 2004; Kenney, 2006), theresults presented here are consistent with thehypothesis that the potential for reciprocity ofexchange is important for understanding livingarrangements. And the strength of expectationsfor reciprocity depends, in part, on the types ofrelationships involved.

As the population ages, there is greaterconcern that family members and significantothers may be called upon for instrumental andsocial support (Bengston, 2001). The resultspresented here suggest multigenerational tiesare the ones most likely to maintain coresidentialarrangements. This may place greater demandson adult child – parent relationships at a timewhen there are more older adults with feweradult children. On the other hand, many ofthe other coresidential households also containolder adults, and our results also suggest thatthese more contractually based coresidentialrelationships can also be maintained. But therole that individuals play in these householdsis likely to be quite different from thoseexperienced in households shared with parents,children, and grandchildren. Further, our resultssuggest the fragility of some extended orcomplex households formed in response tosevere economic constraints (e.g., Menjivar,2000; Wright et al., 1998) may stem, in part,from the difficulty of sustaining equitablecontributions by household members.

NOTE

This research was supported in part by a grant from theEunice Kennedy Shriver National Institute of Child Healthand Human Development (R03-HD44700).

REFERENCES

Allison, P. D. (1995). Survival Analysis Using theSAS System: A Practical Guide. Cary, NC: SASPublishing.

Aquilino, W. S. (1990). The likelihood of parent –adult child coresidence: Effects of family structureand parental characteristics. Journal of Marriageand the Family, 52, 405 – 419.

Becker, G. S. (1981). Altruism in the family andselfishness in the market place. Economica, 48,1 – 15.

Bengston, V. L. (2001). The Burgess Award lecture:Beyond the nuclear family: The increasingimportance of multigenerational bonds. Journalof Marriage and Family, 63, 1 – 16.

Beresford, J. C., & Rivlin, A. M. (1966). Privacy,poverty and old age. Demography, 3, 247 – 258.

Bethencourt, C., & Rios-Rull, J. V. (2009). Onthe living arrangements of elderly widows.International Economic Review, 50, 773 – 801.

Bianchi, S. M., Hotz, J., McGarry, K., & Seltzer, J. A.(2008). Intergenerational ties: Alternative theories,empirical findings and trends, and remaining chal-lenges. In A. Booth, A. C. Crouter, S. M. Bianchi,& J. A. Seltzer (Eds.), Intergenerational caregiv-ing (pp. 3 – 43). Washington, DC: Urban Institute.

Biddlecom, A. E. (1994, May). Immigration andcoresidence in the U.S. since 1960. Paper presentedat the annual meeting of the Population Associationof America, Miami, FL.

Brines, J., & Joyner, K. (1999). The ties that bind:Principles of cohesion in cohabitation andmarriage. American Sociological Review, 64,333 – 355.

Burr, J. A., & Mutchler, J. E. (1992). The living ar-rangements of unmarried elderly Hispanic females.Demography, 29, 93 – 112.

Burr, J. A., & Mutchler, J. E. (1993). Nativity, ac-culturation, and economic status—explanations ofAsian-American living arrangements in later life.Journals of Gerontology, 48, S55 – S63.

Burr, J. A., & Mutchler, J. E. (1999). Race and ethnicvariation in norms of filial responsibility amongolder persons. Journal of Marriage and the Family,61, 674 – 687.

Burton, L., & Stack, C. B. (1993). Conscripting kin:Reflections on family, generation and culture. InP. A. Cowan, D. Field, D. A. Hansen, A. Skolnick,& G. E. Swanson (Eds.), Family, self and soci-ety (pp. 103 – 113). New York: Routledge.

Choi, N. G. (2003). Coresidence between unmarriedaging parents and their adult children—who movedin with whom and why? Research on Aging, 25,384 – 404.

Coleman, M., & Ganong, L. (2008). Normativebeliefs about sharing housing with an older familymember. International Journal of Aging andHuman Development, 66, 49 – 72.

Coward, R. T., Cutler, S. J., & Schmidt, F. E. (1989).Differences in the household composition ofelders by age, gender, and area of residence.Gerontologist, 29, 814 – 821.


DeMaris, A. (1992). Logit modeling: Practical ap-plications. Newbury Park, CA: Sage.

DeMaris, A., & Longmore, M. (1996). Ideology,power, and equity: Testing competing explanationsfor the perception of fairness in household labor.Social Forces, 74, 1043 – 1071.

De Vos, S. (2000). Kinship ties and solitary livingamong unmarried elderly women in Chile andMexico. Research on Aging, 22, 262 – 289.

Ermisch, J. F., & Overton, E. (1985). Minimal House-hold Units: A new approach to the analysisof household formation. Population Studies, 39,33 – 54.

Fertig, A. R., & Reingold, D. A. (2008). Homeless-ness among at-risk families with children intwenty American cities. Social Service Review,82, 485 – 510.

Fitzgerald, J., Gottschalk, P., & Moffitt, R. (1998).An analysis of sample attrition bias in panel data:The Michigan Panel Study of Income Dynamics.Journal of Human Resources, 33, 251 – 299.

Folbre, N. (2004). Disincentives to care: A cri-tique of U.S. family policy. In D. P. Moynihan,T. M. Smeeding, & L. Rainwater (Eds.), The futureof the family (pp. 231 – 260). New York: RussellSage Foundation.

Ganong, L., & Coleman, M. (2006). Patterns ofexchange and intergenerational responsibilitiesafter divorce and remarriage. Journal of AgingStudies, 20, 265 – 278.

Glick, J. E., Bean, F. D., & Van Hook, J. (1997).Immigration and changing patterns of extendedhousehold/family structure in the United States:1970 – 1990. Journal of Marriage and the Family,59, 177 – 191.

Glick, J. E., & Van Hook, J. (2002). Parents’ coresi-dence with adult children: Can immigration explainrace and ethnic variation? Journal of Marriage andFamily, 64, 240 – 253.

Goode, W. J. (1982). The family (2nd ed.). Engle-wood Cliffs, NJ: Prentice Hall.

Goody, J. (1976). Production and reproduction:A comparative study of the domestic domain.Cambridge, UK: Cambridge University Press.

Guo, G. (1993). Event-history analysis for left-truncated data. Sociological Methodology, 23,217 – 243.

Hamilton, W. (1964). The genetical evolution ofsocial behavior: I & II. Journal of TheoreticalBiology, 7, 1 – 52.

Hammel, E. A., & Laslett, P. (1974). Comparinghousehold structure over time and betweencultures. Comparative Studies in Society andHistory, 16, 73 – 109.

Hogan, D. P., Eggebeen, D. J., & Clogg, C. C. (1993).The structure of intergenerational exchanges inAmerican families. American Journal of Sociol-ogy, 98, 1428 – 1458.

Hogan, D. P., Hao, L. X., & Parish, W. L. (1990).Race, kin networks and assistance to mother-headed families. Social Forces, 68, 797 – 812.

Jabine, T. (1990). Survey of Income and ProgramParticipation quality profile. Washington, DC:U.S. Government Printing Office.

Kalmijn, M., Loeve, A., & Manting, D. (2007). In-come dynamics in couples and the dissolutionof marriage and cohabitation. Demography, 44,159 – 179.

Kenney, C. T. (2006). The power of the purse:Allocative systems and inequality in couplehouseholds. Gender and Society, 20, 354 – 381.

Levy, P. S., & Lemeshow, S. (1999). Sampling ofpopulations (3rd ed.). New York: John Wiley &Sons.

Liversage, A., & Jakobsen, V. (2010). Sharingspace—gendered patterns of extended householdliving among young Turkish marriage migrants inDenmark. Journal of Comparative Family Studies,41, 693 – 715.

Martikainen, P., Nihhtila, E., & Moustgaard, H.(2008). The effects of socioeconomic status andhealth on transitions in living arrangements andmortality: A longitudinal analysis of elderlyFinnish men and women from 1997 to 2002.Journals of Gerontology, Series B: PsychologicalSciences and Social Sciences, 63, S99 – S109.

McGarry, K., & Schoeni, R. F. (2000). Social secu-rity, economic growth, and the rise in elderlywidows’ independence in the twentieth century.Demography, 37, 221 – 236.

Menjivar, C. (1997). Immigrant kinship networks:Vietnamese, Salvadoreans and Mexicans in com-parative perspective. Journal of Comparative Fam-ily Studies, 28, 1 – 24.

Menjivar, C. (2000). Fragmented ties: Salvadoranimmigrant networks in America. Berkeley: Uni-versity of California Press.

Michael, R. T., Fuchs, V. R., & Scott, S. R. (1980).Changes in the propensity to live alone—1950 –1976. Demography, 17, 39 – 56.

Molm, L. D., Collett, J. L., & Schaefer, D. R. (2007).Building solidarity through generalized exchange:A theory of reciprocity. American Journal ofSociology, 113, 205 – 242.

Mutchler, J. E., & Burr, J. A. (1991). A longitudinalanalysis of household and nonhousehold livingarrangements in later life. Demography, 28,375 – 390.

Oppenheimer, V. K. (1997). Women’s employmentand the gain to marriage: The specialization andtrading model. Annual Review of Sociology, 23,431 – 453.

Pampel, F. C. (1983). Changes in the propensity tolive alone—Evidence from consecutive cross-sectional surveys, 1960 – 1976. Demography, 20,433 – 447.


Peek, C. W., Koropeckyj-Cox, T., Zsembik, B. A., &Coward, R. T. (2004). Race comparisons of thehousehold dynamics of older adults. Research onAging, 26, 179 – 201.

Plickert, G., Cote, R. R., & Wellman, B. (2007). It’snot who you know, it’s how you know them: Whoexchanges what with whom? Social Networks, 29,405 – 429.

Portes, A., & Sensenbrenner, J. (1993). Embedded-ness and immigration: Notes on the social deter-minants of economic-action. American Journal ofSociology, 98, 1320 – 1350.

Preston, S. H., Heuveline, P., & Guillot, M. (2001).Demography: Measuring and modeling populationprocesses. Oxford, UK: Blackwell Publishing.

Rendell, M. (2011). Breakup of New Orleanshouseholds after Hurricane Katrina. Journal ofMarriage and Family, 73, 654 – 668.

Roschelle, A. (1997). No more kin: Exploring race,class and, gender in family networks. ThousandOaks, CA: Sage.

Rossi, A., & Rossi, P. H. (1990). Of human bonding:Parent – child relations across the life course. NewYork: A. de Gruyter.

Ruggles, S. (2007). The decline of intergenerationalcoresidence in the United States, 1850 – 2000.American Sociological Review, 72, 964 – 989.

Sarkisian, N., & Gerstel, N. (2004). Kin supportamong Blacks and Whites: Race and familyorganization. American Sociological Review, 69,812 – 837.

Schmertmann, C. P., Boyd, M., Serow, W., &White, D. (2000). Elder – child coresidence in theUnited States. Research on Aging, 22, 23 – 42.

Shuey, K., & Hardy, M. (2003). Assistance to agingparents and parents-in-law: Does lineage affectfamily allocation decisions. Journal of Marriageand Family, 65, 418 – 431.

Silverstein, M., Conroy, S. P., Wang, H., Giar-russo, R., & Bengston, V. L. (2002). Reciprocityin parent – child relations over the adult life course.Journal of Gerontology: Social Sciences, 57,S3 – S13.

Silverstein, M., Gans, D., & Yang, F. (2006). Inter-generational support to aging parents: The role ofnorms and needs. Journal of Family Issues, 27,1068 – 1084.

Smits, A., van Gaalen, R., & Mulder, C. H. (2010).Parent – child coresidence: Who moves in withwhom and for whose needs? Journal of Marriageand Family, 72, 1022 – 1033.

Speare, A., & Avery, R. (1993). Who helps whomin older parent – child families? Journal ofGerontology, 48, S64 – S73.

Spitze, G., & Trent, K. (2006). Gender differencesin adult sibling relations in two-child families.Journal of Marriage and Family, 68, 977 – 992.

Stack, C. B. (1975). All our kin: Strategies for survivalin a Black community. New York: Harper and Row.

Stata Corporation. (1997). Stata user’s guide, Release5. College Station, TX: Stata Press.

Stewart-Williams, S. (2007). Altruism among kin vs.nonkin: Effects of cost of help and reciprocalexchange. Evolution and Human Behavior, 28,193 – 198.

Swartz, T. T. (2009). Intergenerational family rela-tions in adulthood: Patterns, variations and impli-cations in the contemporary United States. AnnualReview of Sociology, 35, 191 – 212.

Taylor, R. J., Chatters, L. M., & Mays, V. M. (1988).Parents, children, siblings, in-laws, and nonkinas sources of emergency assistance to BlackAmericans. Family Relations, 37, 298 – 304.

Tienda, M., & Glass, J. (1985). Household structureand labor-force participation of Black, Hispanic,and White mothers. Demography, 22, 381 – 394.

Van Hook, J., & Glick, J. E. (2007). Immigration andliving arrangements: Moving beyond economicneed versus acculturation. Demography, 44,225 – 249.

Wellman, B. (1990). The place of kinfolk in personalcommunity networks. Marriage and FamilyReview, 15, 195 – 228.

White, L. (2001). Sibling relationships over the lifecourse: A panel analysis. Journal of Marriage andFamily, 63, 555 – 568.

Wolf, D. A., & Soldo, B. J. (1988). Household com-position choices of older unmarried women.Demography, 25, 387 – 403.

Wright, B., Avshalom, R. E., Caspi, T., Moffitt, E.,& Silva, P. A. (1998). Factors associated withdoubled-up housing: A common precursor tohomelessness. Social Service Review, 72, 92 – 111.

Documents

Does a House Divided Stand? Kinship and the Continuity of Shared Living Arrangements