1. The Presence of
Non-Parent Adults and Economic Realities for Children in Low-income Neighborhoods
Kate Bachtell, Ph.D.
Nola du Toit, Ph.D.
Catherine Haggerty .

3. Background
Twenty-five years of extensive and rigorous research has shown that children raised in stable, secure families have a better chance to flourish. Family structure is an important factor in reducing poverty, too: children raised in single-parent families are nearly five times as likely to be poor as those in married-couple families. In part, this is the result of simple math: two parents, on average, have far greater resources to devote to raising children than does one parent attempting to raise children alone.
_______________________________________________________________________
Original citations:
57 See Sara McLanahan, Laura Tach, and Daniel Schneider, “The causal effects of father absence,” Annual Review of Sociology 39 (2013): 399-
427; and Waldfogel, Craigie, and Brooks-Gunn (2010) for a survey of the literature.
58 U.S. Census Bureau, Detailed Poverty Tables, POV03: People in families with related children under 18 by family structure, age, and sex, Iterated
by Income-to-Poverty Ratio and Race: 2013,
Aber, L., Butler, S., Danziger, S., Doar, R., Ellwood, D.T., Gueron, J., Haidt, J., Haskins, R., Hymowitz, R., Mincy, R., Reeves, R., Strain, M.R., & Waldfogel, J. (2015). Opportunity, responsibility, and security: A consensus plan for reducing poverty and restoring the American dream. AEI/Brookings Working Group on Poverty and Opportunity. Washington, D.C.: American Enterprise Institute for Public Policy Research and the Brookings Institution.

4. The Problem(s)
For many children, especially the most economically vulnerable, their primary social setting—the household—includes adults other than their parents
Prevailing framing of family “instability” elicits a fictional dichotomy of secure versus insecure families among the poor
Insert Presentation Title and Any Confidentiality Information

5. Our Contribution
Define household structure based on the relationship of each adult to a focal child
Include non-parent adults such as grandparents, aunts, uncles, unrelated others, etc.
Define instability based on any change in the adult composition of the household

6. Data Title Making Connections Survey Funding agency
Annie E. Casey Foundation
Summary
Longitudinal study of families in low-income neighborhoods in ten U.S. cities
Access
Restricted use within NORC's Data Enclave
Sample type
AP and list of focal children
Key advantages
Representation of very poor and racial/ethnic minority households
Detailed household roster information (relationship of each person to the R and FC) and linked personal identifiers

7. Current Objectives
Quantify the prominence of grandparents, aunts, uncles, etc. in low-income households
Determine whether the presence of these non- parent adults is associated with any benefit or detriment to economic wellbeing
Examine the impact of losing and/or gaining non-parent adults on economic wellbeing

8. Analytic Sample
Panel of 1,619 households with children that participated in waves 2 and 3 of the Making Connections Survey
Inclusion criteria:
Same focal child selected in both waves
Valid relationship codes
n = 1,212

9. Dependent Variables
Square root of household income per capita: continuous measure of the household's total income from all sources in the year prior to the wave 3 interview
Economic hardship: 6-point continuous measure indicating how many of the following the household experienced in the past year: (1) did not fill or postponed filling a prescription for drugs; were not able to pay the (2) mortgage, (3) phone, or (4) utility bills; or (5) were without enough money to buy food.

10. Independent Variables
Focus variables:
Ever had a non-parent present in the household: Dummy variable indicating whether, across two waves, the focal child EVER lived with a non-parent adult (e.g. grandparent, aunt or uncle, other extended relative, unrelated adult, etc.) in the home
Ever had a change in non-parent adult in the household: three dummy variables indicating whether the household experienced a numeric change in the number of non-parent adults living in the home between waves (gained, lost, no change)

11. Independent Variables
Family Disruption variables:
Ever a change in parents
Moved since prior wave
Controls:
Respondent characteristics: foreign-born, race/ethnicity, sex, age, highest level of education, married, cohabiting
Household characteristics: number of people, number of parents, relationships to the focal child, employment status across all adults

12. Analysis Descriptive statistics OLS regression models
Weighted to represent households with children in MC neighborhoods as of wave 2 ( )

13. Current Study: RQ1
R1: What are the characteristics of families that include one or more non- parent adults?

14. Never included non-parent adult Ever included non-parent adult
Table 1. Household and Respondent Characteristics: Select Controls (unweighted)
Never included non-parent adult
Ever included non-parent adult
(NNP)
(ENP)
Figure SD
Household Characteristics
Number of people in the HH (mean)
3.87
1.32
4.17
1.47
Number of parents (%)
None
0.00
16.09
One
49.60
51.50
Two
50.40
32.40
Relationships to focal child (adults, %)
Aunt/uncle
18.03
Grandparent
39.70
Other relative
5.36
Boarder/roommate/unrelated
12.66
Employment status (adults, mean)
Employed
1.11
0.82
1.20
0.96
Disabled
0.08
0.29
0.23
0.49
Retired
0.01
0.17
0.45
Unemployed
0.20
0.44
0.25
0.53
Family Disruption
Moved since prior wave (%)
48.53
55.58
Ever a change in parents (%)
15.15
31.76

15. Current Study: RQ2
R2: What impact, if any, does the presence of one or more non-parent adult have on household economic wellbeing?

16. Household income per capita (sq root)
Table 2. Regression Models Predicting Economic Outcomes, Isolating Presence of Non-Parent Adult
Household income per capita (sq root)
Economic hardship
Estimate
Std Error
Intercept
64.02
***
6.92
2.12
0.31
Focus Variables
Ever with non-parent adult in the home
-3.49
2.08
-0.08
0.09
Household Characteristics
Number of parents in household
3.15
1.81
-0.12
0.08
Number of employed adults
5.80
1.28
-0.10
0.06
Number of disabled adults
-7.82 **
2.60
0.25 *
0.11
Number of retired adults
-4.89
3.60
-0.25
0.16
Number of unemployed adults
-8.83
0.20
-4.39
0.65
Square root of income per capita
-0.01
0.00
Family Disruption
Moved since prior wave
-6.14
1.88
0.03 N
1,212
R-Squared
0.34
0.10 F
41.55
9.24
Degrees of Freedom 15
*p<0.05, **p<0.01, ***p<0.001

17. Current Study: RQ3
R3: What impact, if any, does change in the presence of one or more non-parent adult have on household economic wellbeing?

18. Household income per capita (sq root)
Table 3. Regression Models Predicting Economic Outcomes, Isolating Change in Presence of Non-Parent Adult
Household income per capita (sq root)
Economic hardship
Estimate
Std Error
Intercept
67.55
***
11.05
1.84
0.45
Focus Variables
Added a non-parent adult
6.76
4.20
0.33
0.17
Lost a non-parent adult
9.25 *
3.94
0.21
0.16
Household Characteristics
Number of parents in household
-0.26
2.81
-0.02
0.11
Number of employed adults
5.28
2.07
-0.10
0.08
Number of disabled adults
-6.90
3.51
0.42 **
0.14
Number of retired adults
-6.16
4.32
-0.22
Number of unemployed adults
-9.29
3.30
-0.09
0.13
-4.95
1.15
Square root of income per capita
-0.01
0.00
Family Disruption
Change in number of parents
0.63
3.23
-0.42
Moved since prior wave
-8.02
3.55
-0.07 N
466
R-Squared F
12.69
4.27
Degrees of Freedom 17
*p<0.05, **p<0.01, ***p<0.001

19. Recap of Objectives Economic hardship: results are less conclusive
Quantify the prominence of grandparents, aunts, uncles, etc. in low-income households
 38% at one or both waves
Determine whether the presence of these non-parent adults is associated with any benefit or detriment to economic wellbeing
 No significant impact for income nor hardship
Examine the impact of losing and/or gaining non-parent adults on economic wellbeing
Adding a non-parent=no impact
Losing a non-parent=more HH income per capita
No impact on economic hardship
Economic hardship: results are less conclusive

20. Next Steps Quantitative:
Use personal identifiers to capture substitutions of adults
Try fixed effects regression?
Qualitative:
Gather additional data?
Insert Presentation Title and Any Confidentiality Information

21. Contact me: bachtell-kate@norc.org
Download the Working Paper: /working-paper-series.aspx