Which of the Following Families Would Be Included in the U.s. Census Bureau's Definition of Fam

Demography. Writer manuscript; available in PMC 2013 Nov i.

Published in final edited form as:

PMCID: PMC3496021

NIHMSID: NIHMS396029

Measuring Cohabitation and Family Structure in the United states: Assessing the Impact of New Data From the Electric current Population Survey

S. Kennedy

Minnesota Population Middle, University of Minnesota, 50 Willey Hall, 225 19th Ave S., Minneapolis, MN, 55455, Usa ude.nmu@305ennek

C. Fitch

Minnesota Population Center, University of Minnesota, Us

Abstruse

In 2007, the Current Population Survey (CPS) introduced a measure that identifies all cohabiting partners in a household, regardless of whether they depict themselves equally "unmarried partners" in the relationship to householder question. The CPS at present also links children to their biological, footstep-, and adoptive parents. Using these new variables, we clarify the prevalence of cohabitation every bit well equally the demographic and socioeconomic characteristics of different-sex cohabiting couples during the years 2007–2009. Estimates of cohabitation produced using just single partnerships miss 18 % of all cohabiting unions and 12 % of children residing with cohabiting parents. Although differences between unmarried partners and nearly newly identified cohabitors are small, newly identified cohabitors are older, on average, and are less probable to exist raising shared biological or adopted children. These new measures also allow us to identify a small number of young, disadvantaged couples who primarily reside in households of other family unit members, nearly commonly with parents. We conclude with an examination of the circuitous living arrangements and poverty status of American children, demonstrating the broader value of these new measures for research on American family and household structure.

Keywords: Cohabitation, Measurement, Living Arrangements, Stepfamilies, Poverty

Introduction

The rise of cohabitation has dramatically reshaped American family life. About non-existent in 1960, the number of cohabiting couples increased to 7.5 million past 2010 (Fitch et al. 2005; Kreider 2010). More than two-thirds of American adults cohabit earlier they marry, and virtually xl % of children live in a cohabiting family unit during babyhood (Kennedy and Bumpass 2008). The evolution of big and consequent data sources collected at regular intervals to study cohabitation has lagged behind these shifts in family unit structure (Casper and Hofferth 2007). The paucity of regular information on cohabiting families is particularly problematic for the study of children'southward living arrangements and the assessment of children'south well-beingness.

The Almanac Social and Economic Supplement (ASEC) to the Current Population Survey (CPS) is the primary source of annual data on the structure and economic well-beingness of American families (U.S. Census Bureau 2010). Until 2007, but persons who reported themselves as "unmarried partners" of the householder ^ane could exist identified as cohabitors. This measure has two limitations. First, couples who do not recognize the term "unmarried partner" can be missed. 2d, unions in which neither partner is identified equally the householder are excluded. Measurement of family relationships in the CPS was greatly improved in 2007 past the introduction of directly questions identifying all cohabiting couples and linking children to their biological, footstep-, and adoptive parents.

Using the 2007–2009 ASEC surveys, we examine the bear upon of these family unit relationships measures on estimates of the prevalence and characteristics of cohabiting families and couples. Considering unmarried partners are the only identifiable cohabiting couples in the decennial demography, the American Community Survey (ACS), and earlier years of the CPS, we as well assess whether couples identified using the unmarried partner measurement differ demographically or socioeconomically from newly identified cohabiting couples. We examine in detail the living arrangements of cohabiting couples, including residence with children, parents, other relatives, and nonrelatives. In improver, nosotros use the detailed measures of income and family structure in the ASEC to examine the economical well-being of cohabiting couples. Finally, we document the diverseness of children'south living arrangements by describing family construction, stepfamily and extended family unit residence, and child poverty rates in married, cohabiting, and single-parent families.

Background

Much of what we know near cohabitation comes from family surveys. Prior to 1987, no nationally representative statistics existed on cohabitation. The 1987 National Survey of Families and Households was the first survey to collect detailed cohabitation histories. The periodic National Survey of Family Growth has collected cohabitation histories since 1988, but data until recently were limited to women of reproductive age. The Fragile Families survey, a written report of children built-in to urban unmarried parents during 1998–2000, provides important longitudinal data on these families. Although valuable sources of cohabitation data, family surveys are often limited by small or nonnationally representative samples of cohabitors, infrequent data drove, or limited information on family economic well-being. In addition, because family unit surveys often employ different methods for identifying cohabitors, results may not be directly comparable across surveys (Hayford and Morgan 2008; Knab and McLanahan 2007; Pollard and Harris 2007; Teitler et al. 2006).

Until 1990, researchers using population censuses and surveys had to infer cohabitation status based on the coresidence of people of the opposite sex, which proves to be an extremely imprecise method (Casper and Cohen 2000; Fitch et al. 2005). In 1990, the U.S. Demography Bureau added "unmarried partner" as a category in the question asking "relationship to the householder." When the CPS was updated with similar linguistic communication in 1995, detailed data on the prevalence of cohabiting families became available annually. Yet, this arroyo still failed to place some cohabiting couples. Because nigh cohabitors do not use the term "single partner" to describe their relationship—instead, preferring identifiers such as "boyfriend" or "fiancée"— many couples are potentially missed by this measure out (Manning and Smock 2005). In addition, unions non involving the householder (e.m., couples residing with parents or roommates) could not be identified. The new question in the CPS allows us to assess how many couples fall into these ii categories.

CPS and demography data were also limited in their ability to measure the living arrangements and economic well-being of cohabiting families. Prior to 2007, almanac Census Bureau estimates of children's family structure counted children of cohabiting couples equally though they were raised by a single parent (Kreider 2008). Currently, about 40 % of cohabiting couples are raising resident children, and these households would have been classified as female parent-only or begetter-only families. This definition is still in use today by the Census Agency when calculating official family poverty statistics. Treating cohabiting families as unmarried-parent families excludes the income of the cohabiting partners from poverty calculations and substantially underestimates the economic well-being levels in cohabiting families (Iceland 2007; Manning and Brown 2006). The near contempo attempts to calculate cohabiting family incomes and associated poverty levels are a decade out of date, when data sets containing detailed information on income and cohabiting family relationships were last available (Iceland 2007; Manning and Smock 2005).

In 2007, the CPS questionnaire was revised to amend the measurement of cohabitation and family relationships (Kreider 2008). The CPS questionnaire begins past enumerating all usual residents of the sampled household ² equally well every bit persons with no usual residence who are staying in the household. Each person is assigned a line number that represents their position (or line) on the household roster. Afterwards, the interviewer collects demographic data on household members, including relationship to the householder, historic period, and sexual practice (U.South. Census Bureau 2008a). A directly question on cohabitation was added to this section of the interview. In households with unrelated adults, unmarried respondents are asked, "Practise yous take a boyfriend, girlfriend or partner in this household?" If the response was yes, the respondent was and so asked to identify the cohabiting partner, and the interviewer recorded the partner's line number. The same question was posed about all other unmarried adults in the household except persons identified as an unmarried partner in the relationship to householder variable. Estimates of the number of different-sex activity cohabiting couples increased by more than than xx % as a result of the new cohabitation question (Kreider 2008). ⁱⁱⁱ

Because the cohabitation question is asked only of household members, information technology will non identify couples who live together some of the fourth dimension just maintain carve up residences. The more than restrictive definition of the household membership in the CPS approach volition yield lower estimates of cohabitation than surveys that include office-time or visiting relationships (Knab and McLanahan 2007; Pollard and Harris 2007).

Earlier 2007, the CPS also nerveless express information on parent-child relationships: interviewers recorded the line number of 1 parent or stepparent based on relationship to householder and the interviewer'south "noesis of the family unit structure." Thus, researchers could not make up one's mind whether a child was related biologically to both partners or to only one partner. In 2007, the Census Bureau expanded the data nerveless on parent-child relationships. The CPS at present includes both mother and father line numbers and distinguishes between biological, step-, and adoptive parents (Kreider 2008). ⁴ With these new variables, the CPS provides detailed almanac information on the cohabitation experiences of children and adults.

Table 1 illustrates two hypothetical cohabiting household rosters to demonstrate the new family locator variables. ⁵ The top household presents a newly identified cohabiting union involving the householder: prior to the availability of the new cohabitation question, the union between the woman and her male person roommate was invisible. Now they are linked by the partner locator variable. In add-on, it is axiomatic from the parent locator variables that i of the children is the shared biological child of the cohabiting couple. A second child is linked to the female person partner but non to the male partner, and is likely the female parent's child from a previous human relationship. In the second household, the new questions reveal a cohabiting couple residing with their ain child within the household of the female person partner'south parents.

Table one

Households illustrating the new CPS cohabiting partner and parent locator variables

				New family locator variables
Relationship to Householder	Line Number	Age	Sex	Cohabiting Partner Location	Female parent'southward Location	Mother'southward Relationship	Father's Location	Begetter'southward Relationship
1. Newly Identified Householder Unionsa
Householder	1	32	F	2
Roommate	2	thirty	Yard	i
Child	3	5	F		one	Biological
Child	4	1	M		i	Biological	2	Biological
2. Newly Identified Subfamily Unionsb
Relationship to Householder
Householder	i	47	M
Spouse	2	46	F
Child	3	18	F	4	ii		1
Nonrelative	4	18	One thousand	3
Grandchild	5	0	M		3	Biological	four	Biological

Data and Methods

We use data from the 2007–2009 ASEC of the CPS, provided by the Integrated Public Utilise Microdata Series at http://cps.ipums.org (Male monarch et al. 2010). The ASEC collects detailed information on income, employment, and noncash benefits, and is the source for annual census reports on Families and Living Arrangements. Our analysis includes nearly 9,000 different-sex cohabiting couples and more 95,000 children. ⁶

Our goals are to describe the prevalence and characteristics of cohabiting couples and families with children and to assess the impact of the new cohabitation and family relationship variables. Consequently, the methods used are descriptive in nature. We consider a broad array of demographic and socioeconomic correlates of cohabitation and incorporate information on all family members.

Variables

Cohabitation Measurement

Our assay distinguishes the newly identified cohabiting unions from unmarried partnerships. We separate cohabiting unions into the following categories:

1. Householder unions: Relationships involving the householder, the person (or one of the people) in whose name the housing unit is endemic or rented.

   (1a) Unmarried partner unions: Unions betwixt the householder and his or her unmarried partner (identified through relationship to the householder variable). These cohabitors could exist identified before the introduction of a direct question on cohabitation.

   (1b) Newly identified householder unions: Unions between the householder and a nonrelative in the household who is non identified every bit an single partner. These couples are identified only as a result of the direct question on cohabitation.

2. Subfamily cohabiting unions: Newly identified unions between two people in the household, neither of whom is the householder. These subfamilies should not be confused with census-defined subfamilies, which would not include cohabiting partners.

Previous research has revealed differences between unmarried partner unions (1a) and the newly identified unions (1b and 2 combined) (Kreider 2008), only we differentiate beyond all three types. Nosotros expect significant differences between the two major categories: householder (1a and 1b) and subfamily unions (2). The distinction that nosotros make between the two types of householder unions (1a and 1b) is the consequence of measurement and is not meant to imply conceptually different types of relationships. Still, it is critical to examine how representative unmarried partners are of all cohabitors because the single partner variable is the only mensurate of cohabitation in the ACS.

Parent-Child Relationships

Our analysis distinguishes between couples raising shared children and couples who are raising the children of one partner but. We define "shared children" to be children who are biologically related to or adopted by both cohabiting partners. "Stepchildren" are children who are identified as the biological kid or adopted child of 1 parent and as the stepchild of or equally unrelated to the cohabiting partner. ^vii Nosotros and then categorize couples into six mutually exclusive groups:

1. No children: The couple accept no children.

2. Shared children only: The couple are raising only shared biological or adopted children.

3. Shared and stepchildren: The couple are raising both shared and stepchildren.

4. Female person partner's simply: The couple are raising the biological or adopted child(ren) of the female partner only and have no shared children.

5. Male person partner's only: The couple are raising the biological or adopted child(ren) of the male partner only and accept no shared children.

6. Stepchildren of both partners: The couple are raising the biological or adopted child(ren) of the male partner and the biological or adopted child(ren) of the female partner. The couple have no shared children.

Demographic Characteristics of Cohabiting Partners

Nosotros examine variation between couples in partner ages and marital condition. We also compare the race, ethnicity, and nascency of each partner as well as metropolitan status, geographic region, and residential history.

Although the CPS allows respondents to written report multiple races, the number of multiracial cohabitors is too small to analyze separately. Instead, we apply race-bridging methods to predict the single race category that a person would most likely accept reported if (s)he could written report only ane race (Liebler and Halpern-Manners 2008). Our terminal measure identifies the most mutual combinations of partner races.

Socioeconomic Status

We examine the education level of each partner every bit well as school enrollment and employment condition. We as well consider family poverty levels, measured as the ratio of family income to needs. We base our estimates of poverty status on the federal poverty thresholds (U.Southward. Census Bureau 2006, 2007, 2008b). ⁸ Our measures of poverty differ from the official poverty measurements considering we treat cohabiting partners as members of the same family unit. Including cohabiting partner incomes in family poverty measurements more completely accounts for the economic resources available in cohabiting families and thus substantially reduces estimated poverty rates (Carlson and Danziger 1999; Iceland 2007; Manning and Dark-brown 2006).

We calculate income-to-needs poverty in two ways:

1. Couple poverty ratio: In this mensurate, we calculate total family unit income past using the cohabiting couple's income, plus any income contributed past their adult children. In add-on, only the couple and their children are used to calculate family unit size. The ratio of the couple's income to the poverty threshold for their family size is our first measure of family unit poverty. We construct a similar parent/child poverty measure for children in married, cohabiting, and single-parent families.

2. Total family poverty ratio: The second approach is more traditional and includes other relatives who alive in the household in calculations of family income and family size. The difference between this measure out and the couple poverty ratio helps identify the extent to which couples benefit financially from residing with extended family.

We then allocate each couple and each kid into one of four categories: in poverty (income is less than 100 % of the poverty threshold), 100 %–199 % of the poverty threshold, 200 %–299 % of the poverty threshold, and 300 % and more of the poverty threshold.

Analytic Approach

We begin past examining the touch of the directly cohabitation question on estimates of cohabitation. To evaluate whether newly identified cohabitors differ significantly from unmarried partners, we consider the living arrangements besides as the demographic and socioeconomic characteristics of unmarried partner unions, newly identified householder unions, and subfamily cohabiting unions. Multinomial logistic regression models produce similar results and are bachelor in Online Resource one; important differences are noted in the text. We conclude with an analysis of the living arrangements and poverty status of children younger than fifteen.

Variance estimates calculated from the ASEC must take into business relationship the complex sample design of the ASEC. The Demography Agency adult a set of 160 replicate weights that suit for clustering and stratification (Fay and Train 1995). We utilise Stata survey procedures to calculate variances, using the replicate weights. ⁹

We include multiple years of the ASEC to maximize our sample size. The CPS sampling strategy introduces complexities when pooling multiple years. The CPS identifies housing units, non individuals, for inclusion in the sample; information technology then conducts surveys from residents of selected households for iv months in a row, breaks for eight months, and and then collects data for an additional iv months. If respondents move, the CPS does not follow them, instead collecting information on new residents at the address. Because nosotros pool three years of data, the same individuals can appear twice. To avoid including duplicate individuals, our assay includes all respondents to the 2008 ASEC, year 2007 respondents in Months five–8 of the interview bike, and yr 2009 respondents in Months 1–4 of the interview cycle.

Our assay includes information collected 15 months after the recession began in December 2007. The prevalence of the different types of cohabiting unions is unchanged between the 2009 ASEC and before ASEC samples. Poverty rates increased past ane per centum point between the 2008 and 2009 samples (U.S. Demography Bureau 2009).

Results

Prevalence and Family Structure of Cohabiting Couples

Table 2 presents estimates of the proportion of all U.S. adults who were living in a cohabiting union broken down past age groups and type of cohabiting wedlock. ¹⁰ Overall, 6 % of U.S. adults (ages 15+) were cohabiting with a different-sex activity partner. Most cohabitors (82 %) selected "unmarried partner" in the human relationship to the householder question and would have been identified without the new cohabitation question. The remaining couples are newly identified: 11 % were in a union between the householder and a nonrelative; and 6 % were in a union betwixt two household members, neither of whom was the householder. Of the roughly 6.7 meg different-sex couples cohabiting in 2008, 5.4 meg were unmarried partners. Of the 1.25 million newly identified cohabitors, roughly 850,000 are householder unions (relationships including the householder), and 400,000 are subfamily cohabiting unions, residing in the households of parents, other relatives, or nonrelatives.

Tabular array 2

Percentage of adults in a different-sexual practice cohabiting union by historic period and spousal relationship type

			Newly Identified Unions (%)
	All Cohabiting Unions (%)	Unmarried Partner Unionsb (%)	Newly Identified Householder Unionsc	Subfamily Cohabiting Unionsd	% of Unions Newly Identified
Agea
xv–19	ane.78	i.xiv	0.15	0.49	36
20–29	12.38	10.00	ane.29	1.09	19
30–39	vii.65	6.52	0.79	0.34	15
forty–49	5.15	4.40	0.58	0.17	15
50+	2.50	2.05	0.41	0.04	eighteen
Total	five.52	4.54	0.63	0.34	18
Unweighted n (adults)	313,188	313,188	313,188	313,188	17,588
Unweighted n (cohabitors)	17,588	14,668	i,872	1,048	17,588

The new cohabitation mensurate substantially improves estimates of cohabitation prevalence at all ages. Teenage cohabitation is rare but is the well-nigh likely to be missed by unmarried partner measures; of the two % of teens who are currently cohabiting, one-third are newly identified, primarily in subfamily unions. Cohabitation peaks among adults in their twenties, at 12 % of all individuals ages 20–29. At these ages, the single partner measure misses nearly 1-fifth of consensual unions, and nearly ane-half of newly identified couples reside in a subfamily. The new direct cohabitation question improves CPS estimates of the prevalence of cohabitation by 15 % for cohabitors between 30 and 50, and by nearly 20 % at older ages. For older cohabitors, subfamily unions are extremely rare.

Who are these newly identified cohabitors? Effigy one presents information on the relationship between the householder and the cohabiting couple. When one partner is the householder ("newly identified householder unions"), we written report the relationship of the nonhouseholder partner. Usually, the partner is identified as a housemate or roommate (65 %). The remaining partners are identified equally an "other nonrelative" of the householder (31 %) or a boarder (five %). When unions occur betwixt two individuals who are non the householder ("subfamily cohabiting unions"), we report the relationship of the partner most closely related to the householder. Typically, the couple resides with relatives. Near 60 % of subfamily cohabiting unions involve a kid of the householder, and an additional 25 % involve a sibling or other related person. Just 15 % of subfamily unions involve two persons unrelated to the householder, who are identified as roommates, nonrelatives, or boarders.

Relationship of newly identified cohabiting partner to the householder. For subfamily cohabitors, nosotros report the relationship of the partner nearly closely related to the householder. Newly identified unions in the left panel are those between the householder and a person identi-fied as a roommate or other nonrelative. Newly identified unions on the correct panel are those between 2 persons who are non householders. Source: Annual Social and Economic Supplement to the Current Population Survey, 2007–2009

Previous qualitative research suggests that the proportion of cohabitors who ever live with roommates or developed relatives might exceed one-3rd (Manning and Smock 2005). In this cross-sectional snapshot of full-fourth dimension cohabiting couples, we find that but 17 % of cohabitors reside with adults other than their ain children, and only 6 % reside with unrelated roommates.

The new CPS variables are specially valuable for researchers studying circuitous household and family relationships. We demonstrate the rich information available on coresidence in Table iii, which describes the living arrangements of cohabiting couples. The first panel describes the distribution of cohabiting couples across three mutually exclusive household compositions: couple-but, couple and their children but, and couple living with other relatives or nonrelatives (regardless of whether children are present). The 2nd panel examines the presence of children and their human relationship to the couple (shared or stepchild), using the mutually exclusive categories described earlier. The final panel reports the percentage of cohabiting couples residing with parents, other relatives, or nonrelatives; in this department, couples can fall into multiple categories. The CPS household roster allows us to identify all persons related to the householder. For persons unrelated to the householder, just relatives who are linked through spouse, cohabiting partner, or parent-kid pointers can be identified. Consequently, estimates in Table 3 overstate the number of nonrelatives residing with the cohabiting couple.

Table 3

Living arrangements of cohabiting couples

	Unmarried Partner Unionsa (%)	Newly Identified Householder Unionsb(%)	Subfamily Cohabiting Unionsc (%)	Total (%)
A. Household Composition
Couple-only household	52	54	due north/a	49
Couple and children only	36	32*	v*	34
Couple lives with other persons	11	13	95*	17
Total	100	100	100	100
B. Residence With Children
No children	58	62	70*	60
Shared biological/adopted children
Shared children only	16	9*	10*	15
Shared and stepchildren	6	3*	iii*	5
Stepchildren but
Female person partner's only	fourteen	21*	12	fifteen
Male partner's merely	4	4	iii	4
Stepchildren of both partners	ii	i	1	2
Total	100	100	100	100
C. Residence With Other Family/Nonfamilyd
Parents	2	4*	64*	vi
Other relatives	5	v	52*	8
Nonrelatives/unknown	5	5	nineteen*	6
Unweighted due north (couples)	7,334	936	524	8,794

Householder couples—that is, unions involving the householder and their partner—reside primarily in nuclear families; just over one-half reside in couple-but households, and an additional one-3rd reside with their children and no other persons. Of these couples, unmarried partners are significantly more likely to reside with their children and no other persons than newly identified householder partners (36 % vs. 32 %). In total, 40 % of householder couples alive with the children of one or both partners. Unmarried partners, however, are well-nigh twice as probable to have shared biological or adopted children as couples in newly identified householder unions (22 % vs. 12 %). Unmarried partners are also less probable to be raising but those children related to a unmarried partner (twenty % vs. 26 %). These differences are robust to controls for age and marital status. Less than five % of couples in householder unions reside with their parents; five % reside with other relatives, such as siblings; and five % reside with someone unrelated to both partners. With the exception of parent-child relationships, differences betwixt single partner unions and newly identified householder unions are pocket-size.

Subfamily cohabitors, by definition, cannot live alone. Only 5 % live with only their children, and in these households, an adult child is identified as the householder. The vast majority (more than 80 %) live simply with relatives. Of these, two-thirds reside with parents; one-third reside with their ain children; and half reside with other relatives—virtually commonly, siblings. Less than 20 % live with nonrelatives. There is considerable overlap between these family situations: xvi % of subfamily cohabitors live with both parents and children, 34 % alive with parents and other relatives, and 9 % live with both related and unrelated persons. These living arrangements differ significantly from householder unions.

The living arrangements of subfamily partners raising children differ significantly from childless subfamily partners (p < .05). In general, subfamily cohabitors are more likely to reside with the male partner's parents than the female partner's parents (37 % vs. 26 %). Couples raising children, withal, are more than probable to reside with the female partner'south family: 35 % compared with xx % with male person partner'southward family. The breakup for childless couples was 25 % with her family unit and 43 % with his family. Family coresidence patterns appear to follow the gendered nature of caregiving, suggesting that assist with child care may be an important reason couples reside with the female partner'due south family.

Demographic Characteristics of Cohabiting Couples

Nosotros present data on the demographic characteristics of unmarried partners and newly identified cohabitors in Tabular array 4. Subfamily cohabiting unions differ substantially from householder unions more than generally, and the differences between householder unions (unmarried partner unions and newly identified householder unions) are smaller.

Table iv

Demographic feature of cohabitors by blazon of cohabiting union

	Unmarried Partner Unionsa (%)	Newly Identified Householder Unionsb(%)	Subfamily Unionsc (%)	Total(%)
Couple Age Distribution
Age of younger partner
15–19 years	3.8	three.8	nineteen.0*	iv.8
20–29	45.6	40.0*	56.9*	45.vii
30–39	21.5	xix.5	12.ix*	twenty.seven
40–49	sixteen.viii	xviii.seven	8.five*	16.5
l+ years	12.3	18.0*	two.7*	12.iii
Female is older	27.ii	32.3*	24.0	27.vi
Marital Status
Both never married	48.9	42.3*	69.four*	49.5
One partner always-married	23.0	24.8	nineteen.4	23.0
Both ever-married	28.1	32.9*	xi.2*	27.5
Race, Ethnicity, and Nativity
Hispanic origin
Neither Hispanic	81.4	83.three	76.3*	81.3
One partner Hispanic	8.2	ix.6	vi.vi	8.2
Both Hispanic	ten.v	seven.0*	17.ane*	x.five
Couple race
Both American Indian	0.9	0.3*	1.ix	0.9
American Indian and White	0.ix	ane.iii	2.3*	1.0
Both Asian	ane.half dozen	1.6	ii.2	1.half-dozen
Asian and White	1.8	1.6	1.5	1.8
Both Black	10.eight	xi.9	six.4*	10.6
Black and White	iii.7	3.2	2.6	three.6
Both White	79.9	79.5	82.0	eighty.0
Other	0.3	0.6	1.1	0.4
Either partner is foreign born	14.2	12.4	xvi.ane	14.1
Additional Characteristics
Residential history
Neither moved	68.6	65.7	67.half-dozen	68.2
Both moved	25.seven	26.vii	18.2*	25.three
Different history	v.8	seven.half-dozen	14.2*	half-dozen.5
Metropolitan area	82.iv	84.9	79.2	82.5
Region
New England	4.nine	5.1	5.8	five.0
Mid-Atlantic	12.five	11.4	ix.1*	12.1
E N Central	15.9	15.2	18.5	xvi.0
West North Central	7.7	nine.6	5.7*	7.8
South Atlantic	19.1	21.two	14.five*	19.0
Due east South Central	5.4	vi.iii	5.1	5.5
West S Central	9.0	9.6	x.nine	9.2
Mount	8.0	8.5	vii.i	8.0
Pacific	17.vi	13.1*	23.four*	17.5
Survey twelvemonth
2007	24.six	23.9	22.two	24.4
2008	50.4	52.6	48.iv	50.5
2009	25.0	23.5	29.4	25.1
Unweighted n (couples)	7334	936	524	eight,794

The largest differences between the three types of cohabiting unions are plant in age and marital status. The newly identified householder partners are, on average, older than unmarried partners: 18 % are 50 or older, compared with simply 12 % of unmarried partners. ¹¹ In contrast, couples in subfamily couples are younger than unmarried partners. Nearly 20 % of subfamily cohabiting unions include a partner younger than 20; and in an boosted 57 % of couples, the younger partner is in his or her 20s. For unmarried partners, these numbers are just four % and 46 %, respectively. Consistent with these age differences, newly identified householder partners are more likely to be ever-married than single partners, only subfamily cohabitors are less probable.

Few other notable demographic differences exist. A higher proportion of subfamily cohabitors are both Hispanic compared with unmarried partners and newly identified householder partners, only these differences are not robust to multivariate analysis and likely reflect age differences in ethnicity. Nearly couples (about 2-thirds) written report living at the same address for at least one year. Couples in subfamily unions are more probable to study different residential histories, presumably because ane partner is already living with family when the union begins.

Socioeconomic Characteristics of Cohabiting Couples

An important advantage of the new CPS family unit variables is that they ameliorate the accuracy and detail with which the income and poverty status of cohabiting couples can be measured. In particular, when couples reside with extended family, we can examine the ability of a couple to support themselves and their children in a higher place the poverty level to the actual poverty levels they experience. Notation that poverty estimates are based on income earned in the years 2006–2008 and largely predate the increment in poverty observed during the Great Recession, which began in December 2007.

In most respects—including income, education, and employment levels—unmarried partner unions and newly identified householder unions are extremely similar (see Table 5). Unmarried partner unions and newly identified householder unions differ significantly in but two instances: unmarried partners are less likely to both have a bachelor's degree and are more likely to written report near-poverty income (between 100 % and 200 % of poverty threshold) than higher incomes. The magnitude of these differences is small. Single partner unions are highly representative of the socioeconomic characteristics of all unions involving the householder.

Tabular array 5

Socioeconomic characteristics of cohabitors by type of cohabiting spousal relationship

	Unmarried Partner Unionsa	Newly Identified Householder Unionsb	Subfamily Unionsc	Total
	(%)	(%)	(%)	(%)
A. Education and Employment
Educational attainment
Both < loftier school	half dozen.eight	half dozen.iv	14.four*	vii.3
One < high schoolhouse, no college graduate	14.6	14.v	21.3*	fifteen.0
Both loftier school graduate or some college	l.5	48.5	53.2	50.five
One higher graduate (iv yrs.)	17.4	xvi.5	7.ane*	sixteen.six
Both college graduate (4 yrs.)	10.6	14.1*	4.0*	10.6
Female person more educated	28.six	29.2	26.five	28.5
Enrolled in schoolhouse (either)	8.1	7.4	12.7*	8.3
Female works full-fourth dimension	55.2	53.6	42.four*	54.1
Male works full-fourth dimension	68.9	65.two	62.1*	68.0
B. Poverty Condition of Couples
Couple poverty level: 2006–2008
0–99 % of poverty threshold	nine.9	10.8	22.0*	10.9
100–199 %	18.iii	14.6*	29.3*	18.5
200–299 %	xviii.0	22.three*	18.3	18.5
300%+	53.viii	52.3	thirty.4*	52.1
Total family poverty level: 2006–2008
0–99 % of poverty threshold	nine.6	10.two	8.ix	nine.6
100–199 %	eighteen.6	15.five*	xviii.7	eighteen.iii
200–299 %	17.9	22.4*	21.5	xviii.seven
300 %+	53.8	51.9	50.9	53.4
C. Financial Assistance
Couple receives regular fiscal assistance  from outside household	2.iv	1.5	0.eight*	2.ii
Unweighted n (couples)	7334	936	524	8,794

In contrast, subfamily cohabiting unions have significantly lower socioeconomic status (SES) than both types of householder unions. Subfamily cohabitors are less probable to have always attended higher or even to accept finished high school. They are too less likely to exist employed. These differences persist in multivariate models that control for age.

Based on couple incomes, we gauge that more than 20 % of subfamily cohabitors would live below the poverty level, compared with x % of single partners and newly identified householder partners. Couples in subfamily unions are also significantly more probable to report near-poverty incomes. These differences in economic well-being remain significant even afterward we command for the younger age and lower educational attainment of subfamily cohabitors. Because subfamily couples ordinarily reside with extended family, actual family poverty levels are comparable with those of householder unions.

The financial benefits of living with extended family are substantial. Median income in subfamily unions increases from $30,000 to $lxx,000 when contributions from all family members are considered. Although couples living in independent households could receive financial assistance from family and friends who live outside the household, just 2 % of couples in householder unions written report regular financial assistance. Thus, the vast majority of cohabitors receive little or no regular fiscal assistance from families and friends, outside of the benefits of coresidence.

Children's Living Arrangements and Economical Well-being

In Table 6, we demonstrate the value of these new variables for describing children's family construction and assessing economic well-being. ¹² The majority (68 %) of children ages 0–14 live with two married parents. A small percentage (6 %) live with cohabiting parents; and of these, 12 % were newly identified using the direct question on cohabitation. An additional 23 % of children alive with a single parent. Besides, a very pocket-size percent (3 %) alive in a household without their parents, with grandparents, other relatives, or in a foster family unit; these children are not included in the Tabular array 6.

Tabular array half-dozen

Living arrangements and poverty status of children ages 0–14 living with parents

	Married parents	Cohabiting parents	Single parent	Full
	(%)	(%)	(%)	(%)
A. Lives With Two Biological/Adoptive Parents	93.0	47.nine*	n/a	due north/a
B. Parent-Child Relationships in Two-Parent Families
Family has shared biological/adopted children
Shared children just	87.0	35.8*	northward/a	n/a
Shared and stepchildren	9.v	22.four*	n/a	due north/a
Family has stepchildren only
Female partner's only	2.ii	26.7*	n/a	due north/a
Male partner'due south only	0.four	vii.one*	north/a	n/a
Stepchildren of both partners	0.ix	8.0*	n/a	north/a
Full	100.0	100.0		n/a
C. Household Composition
Parent(s) and children only	91.2	87.0*	71.5*	86.3
Lives with other relatives or nonrelatives	8.8	xiii.0*	28.5*	thirteen.7
Grandparentsa	4.i	iv.2	18.9*	7.half-dozen
Other relativesa	4.seven	6.7*	sixteen.2*	7.half dozen
Nonrelatives/Unknowna	ane.2	4.3*	iii.0*	1.8
D. Poverty Condition: 2006–2008
Full family poverty	9.2	19.7*	39.2*	17.0
Parent-kid povertyb	nine.vii	21.3*	47.9*	nineteen.vi
Unweighted n (children)	68,621	five,990	21,195	95,806

The parent-child relationship variables are especially useful for studying trends in family complexity, including stepfamily and extended-family unit residence, and highlight the much higher prevalence of stepfamilies and other complex families among cohabiting-couple households. The vast majority of children with two married parents (93 %) are the shared biological or adopted children of both parents, and only 13 % live in stepfamilies (in which some or all of the children are stepchildren). In contrast, just less than one-half of all children in cohabiting families alive with two biological or adoptive parents, and well-nigh 2-thirds live in stepfamilies. Amid children in married stepparent families, nearly 3-quarters live in a family that is also raising at to the lowest degree one shared child, simply this is true of only one-third of children in cohabiting stepfamilies. Cohabiting families more often involve children from more 1 partnership: 22 % of children with cohabiting parents live in a family with both shared and stepchildren, and another eight % live in a family with children of both partners but no shared children. In married couple families, ten % of children alive in as circuitous families. As researchers explore the implications of stepfamilies and multiple-partner fertility for child well-beingness, information technology will be important to utilise the CPS data to runway the prevalence of these living arrangements (Carlson and Furstenberg, Jr. 2006; Guzzo and Furstenberg, Jr. 2007; Halpern-Meekin and Tach 2008).

The new CPS variables also demonstrate the complexity of living arrangements in single-parent families: xix % live with a grandparent, and 16 % live with other relatives, with overlap between these categories. In total, 29 % of children in unmarried-parent families reside with persons other than their parents and siblings, compared with 9 % of children with married parents and xiii % of children with cohabiting parents.

The final panel of Table 6 displays child poverty rates in 2006–2008 past family structure, using the ii measures described before: total family and parent-child poverty. Living with grandparents and other relatives provides a substantial boost to the economic well-beingness of children in unmarried-parent families; without the income contributed by extended family, 48 % would live in poverty. Overall, 17 % of children live under the poverty line using the total family measure. Equally found in previous research, poverty in cohabiting families falls between married-parent and single-parent families; 9 % of children in married families, 20 % in cohabiting families, and nigh 40 % in unmarried-parent families live below the poverty threshold. These estimates are similar to 1998 poverty rates (Manning and Brown 2006), and the bear on of the December 2007 recession was non yet apparent in these poverty calculations. The new CPS variables will be particularly useful for tracking child poverty rates through business organization cycles.

Give-and-take

This article uses newly developed measures of cohabitation and family unit relationships in the CPS to provide an upward-to-date portrait of U.S. cohabitation and family structure. Using these new measures, nosotros gauge that vi % of U.S. adults were cohabiting with a dissimilar-sex partner, representing about 6.7 million couples in March 2008. Virtually adult cohabiting unions (82 %) are identified through the relationship to householder question (unmarried partners), but a substantial minority (18 %) are identifiable only with the addition of a directly question on cohabitation. Surveys without this question, like the ACS and the decennial demography, volition essentially underestimate the prevalence of cohabitation. The unmarried partners identified in these surveys may not be representative of all cohabiting couples. Addressing this issue required examining two distinct groups of newly identified cohabitors.

One group of newly identified cohabitors—the group that we label "subfamily cohabitors"—cannot be captured by the unmarried partner measure because neither partner is identified as a householder. These couples correspond six % of all cohabitors and 33 % of the newly identified cohabitors. The other group—"newly identified householder cohabitors"—has a household structure similar to unmarried partners (a householder and cohabiting partner) and represents a larger share of cohabitors: 11 % of the total and 66 % of newly identified cohabitors.

Our analysis reveals some limitations in using "single partners" equally a proxy for all householder cohabiting partners. These newly identified householder couples are slightly older than those in unmarried partner unions. They are too essentially less likely to be raising shared biological or adopted children simply more than likely to be raising stepchildren. In about all other respects, across a multifariousness of socioeconomic and demographic characteristics, the couples who identify themselves as unmarried partners are remarkably similar to couples who exercise non employ this term. We detect no evidence that researchers should distinguish analytically between newly identified householder unions and unmarried partners. However, until cohabitation measurement is improved in the ACS and other surveys that capture only unmarried partners, researchers should proceed cautiously when using these data sources to study cohabiting families.

Our assay also demonstrates that subfamily cohabitors differ substantially from unmarried partners. These couples are significantly younger, are less educated, and accept lower incomes than single partners. More than than 80 % reside with other family unit members—primarily parents, which offers a living arrangement that essentially improves their standard of living. A clear do good of the new CPS measures is the power to identify this small but important subset of cohabiting couples.

The new CPS family human relationship variables are essential for studying cohabiting families, which is proving to be an important and growing family form. They correct for a substantial underestimate of cohabitation levels in the United States, permit us to portray the complexity of cohabiting family life, and make possible regular and accurate assessment of the economic well-being of cohabiting couples. Although we focus primarily on cohabitation, our analysis demonstrates the broader value of these new variables. They enable researchers to measure the variety of married and cohabiting families, distinguishing between families with just shared biological children, only stepchildren, and both shared and stepchildren; these differences may have implications for child well-being (Halpern-Meekin and Tach 2008). In addition, they improve the identification of single-parent families, multigenerational families, and the living arrangements of young adults during transition to adulthood. The new CPS measures represent an important development in the availability of data to accurately measure trends in the living arrangements and economic well-being of Americans over the life course.

Acknowledgments

An before version of this commodity was presented at the 2009 Population Clan of America meetings. We are grateful to Jason Fields, Steve Ruggles, Carolyn Liebler, and anonymous reviewers for helpful comments. Funding was provided by the Minnesota Population Centre and by grants from the NSF (SES-0617560) and the NICHD (R01-HD-054643).

Footnotes

¹A householder is the person in whose proper name the household unit is owned or rented. When multiple household members meet this requirement, the survey respondent selects i person every bit the householder.

ⁱⁱA usual residence is the place where a person usually lives and sleeps, and can return to at whatever moment.

³Kreider (2008) as well compared cohabitation estimates produced by the new CPS measures, the Survey of Income and Program Participation (SIPP), and the American Customs Survey (ACS).

⁴Nonresident children cannot be identified.

⁵The cohabiting partner locator variable is called PECOHAB and includes unmarried partners and newly identified cohabiting couples. Spouses are identified by A-SPOUSE. PELNMOM and PELNDAD identify mother and father line numbers, and PEMOMTYP and PEDADTYP distinguish betwixt biological, pace-, and adoptive parents.

⁶Prior to 2010, the CPS produced a meaning underestimate of the prevalence of same-sex activity unions (Kreider 2008). A change in editing procedures implemented in Jan 2010 placed CPS estimates in line with those produced using the ACS (Kreider 2010). See Gates (2010) for an evaluation of the measurement of aforementioned-sexual activity couples in census data.

⁷We have excluded 11 couples in our sample where there appear to exist errors in relationship to householder, the cohabitation pointer, or the parent pointers.

^viiiNote that the ASEC collects income data for the previous year. Thus, our estimates of poverty condition cover the years 2006–2008.

^nineDocumentation is available online (http://cps.ipums.org/cps/repwt.shtml).

¹⁰Cohabitation levels past sex activity only reveal the overall earlier union formation of women compared with men.

¹¹We mensurate couple historic period every bit the historic period of the younger partner. Using the ages of male and female partner yielded similar results.

¹²Our analysis excludes one,057 children residing in a household where in that location announced to be errors in relationship to householder or pointers.

Contributor Data

S. Kennedy, Minnesota Population Center, Academy of Minnesota, 50 Willey Hall, 225 19th Ave S., Minneapolis, MN, 55455, United states ude.nmu@305ennek.

C. Fitch, Minnesota Population Center, University of Minnesota, United states.

References

• Carlson Marcia J., Furstenberg Frank F., Jr. The prevalence and correlates of multipartnered fertility amongst urban United states parents. Journal of Matrimony and Family. 2006;68:718–732. [Google Scholar]
• Carlson Marcia, Sheldon Danziger. Cohabitation and the Measurement of Child Poverty. Review of Income and Wealth. 1999;45:179–191. [Google Scholar]
• Casper Lynne Thousand., Cohen Philip North. How does POSSLQ measure upward? Historical estimates of cohabitation. Demography. 2000;37:237–245. [PubMed] [Google Scholar]
• Casper Lynne 1000., Hofferth Sandra Fifty. Playing Catch-up: Improving Data and Measures for Family Research. In: Hofferth Sandra L., Casper Lynne Yard., editors. Handbook of Measurement Problems in Family Research. Lawrence Erlbaum Associates; Mahwah, NJ: 2007. pp. 35–51. [Google Scholar]
• Fay Robert, Train George. Proceedings of the Section on Government Statistics. American Statistical Clan; Alexandria, VA: 1995. Aspects of Survey and Model Based Postcensal Estimation of Income and Poverty Characteristics for States and Counties; pp. 154–159. [Google Scholar]
• Fitch Catherine, Goeken Ron, Ruggles Steven. The Rise of Cohabitation in the U.s.a.: New Historical Estimates. Minnesota Population Heart; Minneapolis: 2005. (Working Paper Series: #2005-03) [Google Scholar]
• Gates Gary. Same sex couples in Usa Census Bureau Information: Who gets counted and why. Williams Institute, UCLA School of Law; Los Angeles: 2010. Retrieved from http://www.constabulary.ucla.edu/williamsinstitute/pdf/WhoGetsCounted_FORMATTED1.pdf. [Google Scholar]
• Guzzo Karen Benjamin, Furstenberg Frank F., Jr Multipartnered Fertility Among Young Women with a Nonmarital First Birth: Prevalence and Hazard Factors. Perspectives on Sexual and Reproductive Health. 2007;39:29–38. [PubMed] [Google Scholar]
• Halpern-Meekin Sarah, Tach Laura. Heterogeneity in ii-parent families and adolescent well-being. Periodical of Marriage and Family. 2008;lxx:435–451. [Google Scholar]
• Hayford Sarah R., Morgan South. Philip. The quality of retrospective data on cohabitation. Demography. 2008;45:129–141. [PMC free article] [PubMed] [Google Scholar]
• Iceland John. Measuring Poverty with Unlike Units of Analysis. In: Hofferth Sandra L., Casper Lynne M., editors. Handbook of Measurement Issues in Family Inquiry. Lawrence Erlbaum Associates; Mahwah, NJ: 2007. pp. 221–233. [Google Scholar]
• Kennedy Sheela, Bumpass Larry L. Cohabitation and children'southward living arrangements: New estimates from the United States. Demographic Research. 2008;19:1663–1692. article 47. doi:10.4054/DemRes.2008.19.47. [PMC free article] [PubMed] [Google Scholar]
• King Miriam, et al. Integrated Public Utilise Microdata Series, Current Population Survey. Version 3.0. University of Minnesota; Minneapolis: 2010. [Automobile-readable database] [Google Scholar]
• Knab Jean Tansey, McLanahan Sara. Measuring Cohabitation: Does How, When, Who You Ask Affair? In: Hofferth Sandra L., Casper Lynne Chiliad., editors. Handbook of Measurement Issues in Family Inquiry. Lawrence Erlbaum Associates; Mahwah, NJ: 2007. pp. 19–33. [Google Scholar]
• Kreider Rose Thou. Improvements to Demographic Household Data in the Current Population Survey: 2007. U.S. Census Agency; Washington, DC: 2008. (Housing and Household Economic Statistics Division Working Paper) ( http://www.census.gov/population/www/documentation/twps08/twps08.pdf) [Google Scholar]
• Kreider Rose M. Increase in Opposite-Sex Cohabiting Couples from 2009 to 2010 in the Annual Social and Economic Supplement (ASEC) to the Current Population Survey (CPS) U.S. Demography Agency, Housing and Household Economic Statistics Division; Washington, DC: 2010. Retrieved from http://world wide web.census.gov/population/www/socdemo/Inc-Opp-sexual activity-2009-to-2010.pdf. [Google Scholar]
• Liebler Carolyn A., Halpern-Manners Andrew. A Practical Arroyo to Using Multiple-Race Response Data: A Bridging Method for Public-Use Microdata. Demography. 2008;45:143–155. [PMC free article] [PubMed] [Google Scholar]
• Manning Wendy D., Brown Susan L. Children's Economic Well-Being in Married and Cohabiting Parent Families. Periodical of Marriage and Family. 2006;68:345–362. [Google Scholar]
• Manning Wendy D., Smock Pamela J. Measuring and Modeling Cohabitation: New Perspectives from Qualitative Data. Journal of Marriage and Family. 2005.;67:989–1002. [Google Scholar]
• Pollard Michael Southward., Harris Kathleen Mullan. Measuring Cohabitation in Add Health. In: Hofferth Sandra L., Casper Lynne Chiliad., editors. Handbook of Measurement Issues in Family Research. Lawrence Erlbaum Associates; Mahwah, NJ: 2007. pp. 35–51. [Google Scholar]
• Teitler Julien O., Reichman Nancy East., Koball Heather. Contemporaneous Versus Retrospective Reports of Cohabitation in the Fragile Families Survey. Journal of Marriage and the Family. 2006;68:469–477. [Google Scholar]
• U.South. Census Bureau . Poverty Thresholds for 2006 by Size of Family and Number of Related Children Under xviii Years. Housing and Household Economic Statistics Division; Washington, DC: 2006. Retrieved from http://world wide web.census.gov/hhes/www/poverty/information/threshld/thresh06.html. [Google Scholar]
• U.S. Census Bureau . Poverty Thresholds for 2007 by Size of Family and Number of Related Children Under 18 Years. Housing and Household Economic Statistics Division; Washington, DC: 2007. Retrieved from http://www.census.gov/hhes/www/poverty/data/threshld/thresh07.html. [Google Scholar]
• U.Southward. Census Bureau . Current Population Survey Interviewing Manual: January 2007. 2008a. Retrieved from http://www.census.gov/apsd/techdoc/cps/CPS_Interviewing_Manual_July2008rv.pdf. [Google Scholar]
• U.S. Census Agency . Poverty Thresholds for 2008 by Size of Family and Number of Related Children Under eighteen Years. Housing and Household Economic Statistics Division; Washington, DC: 2008b. Retrieved from http://www.census.gov/hhes/www/poverty/information/threshld/thresh08.html. [Google Scholar]
• U.S. Census Bureau . Poverty Status of Families, by Type of Family unit, Presence of Related Children, Race, and Hispanic Origin: 1959 to 2008. 2009. Table 4. Retrieved from http://www.census.gov/population/socdemo/hh-fam/tabMS-2.xls. [Google Scholar]
• U.S. Census Bureau . America's Families and Living Arrangements: 2009. 2010. Retrieved from http://www.census.gov/population/www/socdemo/hh-fam/cps2009.html. [Google Scholar]

juliandeparose.blogspot.com

Source: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3496021/

Share this post