Nola du Toit
Jennifer Brown
Cathy Haggerty
Who Really Lives here and does it Matter? Household Structure Trajectories for Children Living with Other Adults in the Home
2
Background
• Household structure is important for child well-being
(Manning and Brown 2006, McClanahan and Sandefur 1994)
• Transitions and changes in family structure can be detrimental to child wellbeing
(Brown 2006, Magnuson and Berger 2009)
• Most examine household structure based on relationships of parents
•Single, two-parent, married, cohabiting, etc.
3
Background
• What about other adults in the home besides parents?
• Current trends shows a resurgence of people living with extended families
(Glick et al 1997, Goldscheider and Bures 2003)
4
Background
• Previous findings
(du Toit, Bachtell, and Haggerty 2011)
• Children in low-income neighborhoods•12% live with extended family adults•22% live with grandparents•6% live with non-related adults•10% have no parent present
5
Overall Research Questions
What does household structure look like when include other adults?•Household structure “through the eyes of the child”•Not focused on relationship status of parents•Who is actually in the home with the child?
How are children affected by other people living in the home over time?•Considering new household structure, what are childhood trajectories?
6
Data
• Making Connections Survey
• Funded by Annie E. Casey Foundation
• Households in low-income neighborhoods
• 10 US cities
• 3 waves of data (6-7 years)• Baseline 2002-2004• Wave 2 2005-2007• Wave 3 2008-2011, 7 cities
7
Data
• Economic hardship, neighborhood involvement, services and amenities, employment history, etc.
• Data on children• Activities, schooling, health, etc.• Randomly selected focal child • All children in household (W23)
• Relationship of adults to focal child• Grandparent• Extended (aunts, uncles, adults cousins)• Non-related adult (roommate, boarder, other)
8
Data
• Analytic subset of cases• W123 panel households with children• Same focal child in all waves• Valid on relationship variables
• N=672 households
9
Household Structure
Relationship data to identify five household structures
Structure # Parents # Adults Example
Single parent only SO 1 parent 1 adult Single mom
Single parent plus SP 1 parent 1+ adults Single mom + gran
Two parent only 2O 2 parents 2 adults Mom, dad
Two parent plus 2P 2 parents 2+ adults Mom, dad, + uncle
Non parent NP 0 parents 1+ adults Gran, or gran + aunt
10
Household Structure at W1
W1 Household StructureUnweighted
NWeighted
%
Single parent only SO 220 28%
Two parent only 2O 272 45%
Single parent plus SP 88 12%
Two parent plus 2P 50 9%
Non parent NP 42 5%
N 672 100%
11
Household Structure Trajectories
• Household structures change over time• Example for W1 single-parent-only trajectories
W1 W2 What happened?
SO
SO Same in both waves or stable
SP Gained other adults, e.g. gran moved in
2O Gained 2nd parent, e.g. dad moved in
2P Gained 2nd parent, e.g. dad and gran
NP Lost single parent, gained non-parent
12
Household Structure Trajectories
• Across W123 waves = 125 possible trajectories
• Five stable trajectories (no change)• SOSOSO (single parent only)• SPSPSP (single parent plus)• 2O2O2O (two parent only)• 2P2P2P (two parent plus)• NPNPNP (no parent)
• Many instable trajectories (change in at least one wave)
13
Household Structure Trajectories
78/125 trajectories recognized (not all shown)
45% stable across 3 waves
More than half experienced changed
14
Types of Trajectories
How many children experience trajectories that include …
…change at all waves?
…other adults?
…non parent households?
15
Types of Trajectories
16
Odds of Instability
• We examine the odds of instability across waves for each Wave 1 household type
• Instability in at least one wave• For example, if child starts in SO, what are odds
of experiencing instability compared to other household structures?
• Controlling for race/ethnicity, education, number of adults, number of children, income
17
Odds of Instability
Odds Ratios for W1 Structure Experiencing Instable Trajectory: Any Changes (N=672)
SO (ref) SP (ref) 2O (ref) 2P (ref) NP (ref)
Single Parent Only -- 0.284 *** 1.433 *** 0.463 *** 0.102 ***
Single Parent Plus 3.518 *** -- 5.040 *** 1.629 *** 0.358 ***
Two Parent Only 0.698 *** 0.198 *** -- 0.323 *** 0.071 ***
Two Parent Plus 2.160 *** 0.614 *** 3.094 *** -- 0.220 ***
Non Parent 9.833 *** 2.795 *** 14.086 *** 4.552 *** --
*p<0.05, **p<0.01, ***p<0.001. Note: Controlling for education, race/ethnicity, number of adults/children, HH income
Odds of experiencing instability is significantly greater for HH with other adults than single-parent-only and two-parent-only
18
Do Trajectories Matter?
Do stable or instable trajectories matter?
So what if there is change in household structure?
•W1W3 • Odds of decrease in household income• Odds of increase in economic hardship
19
Do Trajectories Matter?
•Household income • Wave 1 (categories) v. Wave 3 (continuous)• $5k increments
• $0 $5k• $5 $10k, etc…
•Decrease in household income 0 = same or higher income category in W31 = lower income category in W3
20
Do Trajectories Matter?
•Economic hardship is a scale of 5 items• No money for phone, prescriptions, bills,
utilities, food
•Increase in economic hardship 0 = same/less economic hardship in W31 = greater economic hardship in W3
•Controlling for race/ethnicity, education, number of adults/children, income (economic hardship)
21
Do Trajectories Matter?
Effect of Instable Trajectory: Odds Ratios (N=672)
Odds of Decrease
in IncomeOdds of Increase in Economic Hardship
Intercept 0.191 *** 0.639 ***
Instable Trajectory (ref stable) 1.688 *** 1.741 ***
Less than HS (ref HS) 2.120 *** 1.093 *
BA degree+ (ref HS) 1.002 0.516 ***
NH Black (ref NH White) 0.807 *** 0.873 *
NH Other (ref NH White) 1.155 * 1.020
Hispanic (ref NH White) 0.655 *** 1.088
Number of adults 1.295 *** 0.809 ***
Number of children 0.806 *** 0.969 *
Income -- 1.000
-2 Log Likelihood 18666.92 23168.322
df 8 8
22
Overall Findings
• Nearly half of children will spend time in households with other adults
• 14% of children will live with no parents
• Other adults increase odds of children experiencing instability in household structure
• Instability increases odds that children experience decrease in household income and increase in economic hardship
23
Limitations
• Sample size
• Low income neighborhoods
• Grandparents ~= roommates?
• Two parents married or cohabiting?
• Ignores change in household children
24
Conclusions
• Presence of other adults in the home matters
• More research should examine how children are affected by these other adults
• Economic• Non-economic• Over time
• Future research• Look at all household children, not just focal child
Thank You!
Nola du Toit: [email protected]
Jennifer Brown: [email protected]
Cathy Haggerty: [email protected]