Neighbourhoods are
Does it take a village? Investigations of neighbourhood influences on young children’s development in Australia Dr Ben Edwards, Research Fellow
Structure of the presentation
Does it take a village? (Edwards, 2005) Pathways of influence –community factors (Edwards & Bromfield, in press; in preparation)
Pathways of influence –family factors (Edwards, Sawyer & Pfeiffer, in preparation)
Ghosts and islands…
Bronfenbrenner’s ecological model of children’s development
Neighbourhood inequality is increasing…
Source: Hunter, B. (2003). Australian Economic History Review, 43, 22-44
Neighbourhood inequality is increasing…
Source: Hunter, B. (2003). Australian Economic History Review, 43, 22-44
Neighbourhood socioeconomic status and children’s development International studies (mainly US) suggest that neighbourhood SES is associated with: Social and emotional Cognitive and learning Antisocial behavior and chances of becoming a teenage parent Less is known about young children and very little research in Australia
Policy context Government funded area based initiatives, for example: Communities for Children (Federal) Best Start (Victoria) Better Futures (NSW) Community Renewal Unit (SA) Schools as Communities (ACT) Building Healthier Communities (NT)…
Does it take a village? Very little Australian research, at the time… Children with the lowest levels of emotional and social adjustment lived in disadvantaged neighbourhoods (Homel & Burns, 1989) Aspects of the physical environment of neighbourhoods have been associated with physical activity and obesity (Timperio et al. 2004;
Hume et al. 2005; Timperio et al. 2005).
Growing up in Australia, the Longitudinal Study of Australian Children -Study background
Initiated and funded by the Australian Government Department of Families, Community Services and Indigenous Affairs
Managed in partnership with the Australian Institute of Family Studies
Advice from a consortium of universities and other research agencies
Data collection by the Australian Bureau of Statistics (I-view in Wave 1)
LSAC -Broad study design
Four waves, 2-cohort cross sequential design
5,000 per cohort wave 1; 90% interviewed at wave 2
Nationally representative of Australian children in selected age ranges
Medicare sampling frame Wave 1 2004
Wave 2 2006
Wave 3 2008
Wave 4 2010
B (Infant)
0-1 years
2-3 years
4-5 years
6-7 years
K (Child)
4-5 years
6-7 years
8-9 years 10-11 years
Cohorts
Does it take a village? Variables in analysis Neighbourhood SES –Socio-Economic Indices for Area’s (SEIFA) Index of Advantage/Disadvantage (Trewin 2001) Controlling for selection bias -income, child’s gender and age, child is ATSI, single-parent household, in a household where no one worked, one or both parents were born overseas, and mother’s education And outcomes…
Outcome indexes Overall Mean score = 100 and SD = 10 Physical, learning and social/emotional Physical BMI, overall health, Physical health scale of PEDSQL, need more health care Social/emotional Strengths and Difficulties Questionnaire Learning Peabody Picture Vocabulary Test , parent & teacher ratings of reading skills & teacher ratings of writing & numeracy skills
Analyses Multilevel analyses1: Level 1 family and child factors including perceptions of the neighbourhood Level 2 neighbourhood level variables Population sampling weights 1Mixed
model, random coefficient model, hierarchical linear model, linear random effect regression
Neighbourhood socioeconomic status and children’s outcomes -overall
Source: Edwards, B. (2005). Family Matters, 72, 36-45.
Neighbourhood socioeconomic status and young children’s outcomes –social/emotional
Source: Edwards, B. (2005). Family Matters, 72, 36-45.
Neighbourhood socioeconomic status and young children’s outcomes –learning
Source: Edwards, B. (2005). Family Matters, 72, 36-45.
More disadvantaged neighbourhoods have a greater impact on boys than girls
Source: Edwards, B. (2005). Family Matters, 72, 36-45.
Pathways of influence- Community factors
What are the pathways of influence?
Traditional model of neighbourhood influences on children’s development Source: Roosa, Jones, Tein, & Cree. (2003). American Journal of Community Psychology, 31, 55-72.
What are the pathways of influence?
Neighbourhood SES
Internalising problems
Xue, Leventhal, Brooks-Gunn & Earls. (2005). Archives of General Psychiatry, 62, 554-563.
What are the pathways of influence? Collective efficacy Resident involvement in local organisations
Neighbourhood SES
Internalising problems
Xue, Leventhal, Brooks-Gunn & Earls. (2005). Archives of General Psychiatry, 62, 554-563.
What are the pathways of influence? Roosa and colleagues alternative model Neighbourhood perceptions
Neighbourhood Characteristics
Neighbourhood social processes
Family factors
Child outcomes
Source: Roosa, Jones, Tein, & Cree. (2003). American Journal of Community Psychology, 31, 55-72.
Present study
Test Roosa and colleagues’ model
Perceptions of neighbourhood mediate neighbourhood SES Children’s behavioral and emotional outcomes are, in turn, mediated by neighbourhood social processes
Focus on problematic and positive aspects of children’s behaviour Test influence of neighbourhood remoteness (ARIA) and observational measure of physical disorder
Method Sample Wave 1 of 4-5 year old cohort from LSAC (n = 4,983) 330 postcodes as “neighbourhoods” Variables Outcomes - prosocial behavior, conduct problems, hyperactivity, emotional symptoms and peer problems using the SDQ Family demographics –family income, child’s gender and age, child ATSI, single parent household, household where no-one was employed, household where parents were born overseas, maternal education
Neighbourhood variables Parental perceptions of neighbourhood facilities, belonging, cleanliness and safety derived from multilevel CFA (Edwards, 2006) Neighbourhood level SES (SEIFA), residential instability, ethnic homogeneity, remoteness (ARIA) Interviewer-rated physical disorder of the condition of nearby residence
Analyses Multilevel analyses1: Level 1 family and child factors including perceptions of the neighbourhood Level 2 neighbourhood level variables Population sampling weights 1Mixed
model, random coefficient model, hierarchical linear model, linear random effect regression
Conduct problems Model 1: Neighbourhood quality
Neighbourhood quality Neighbourhood SES % moved in last 5 years % born in Australia Remoteness Physical disorder
χ2(4) = 17.32, p
Structure of the presentation
Does it take a village? (Edwards, 2005) Pathways of influence –community factors (Edwards & Bromfield, in press; in preparation)
Pathways of influence –family factors (Edwards, Sawyer & Pfeiffer, in preparation)
Ghosts and islands…
Bronfenbrenner’s ecological model of children’s development
Neighbourhood inequality is increasing…
Source: Hunter, B. (2003). Australian Economic History Review, 43, 22-44
Neighbourhood inequality is increasing…
Source: Hunter, B. (2003). Australian Economic History Review, 43, 22-44
Neighbourhood socioeconomic status and children’s development International studies (mainly US) suggest that neighbourhood SES is associated with: Social and emotional Cognitive and learning Antisocial behavior and chances of becoming a teenage parent Less is known about young children and very little research in Australia
Policy context Government funded area based initiatives, for example: Communities for Children (Federal) Best Start (Victoria) Better Futures (NSW) Community Renewal Unit (SA) Schools as Communities (ACT) Building Healthier Communities (NT)…
Does it take a village? Very little Australian research, at the time… Children with the lowest levels of emotional and social adjustment lived in disadvantaged neighbourhoods (Homel & Burns, 1989) Aspects of the physical environment of neighbourhoods have been associated with physical activity and obesity (Timperio et al. 2004;
Hume et al. 2005; Timperio et al. 2005).
Growing up in Australia, the Longitudinal Study of Australian Children -Study background
Initiated and funded by the Australian Government Department of Families, Community Services and Indigenous Affairs
Managed in partnership with the Australian Institute of Family Studies
Advice from a consortium of universities and other research agencies
Data collection by the Australian Bureau of Statistics (I-view in Wave 1)
LSAC -Broad study design
Four waves, 2-cohort cross sequential design
5,000 per cohort wave 1; 90% interviewed at wave 2
Nationally representative of Australian children in selected age ranges
Medicare sampling frame Wave 1 2004
Wave 2 2006
Wave 3 2008
Wave 4 2010
B (Infant)
0-1 years
2-3 years
4-5 years
6-7 years
K (Child)
4-5 years
6-7 years
8-9 years 10-11 years
Cohorts
Does it take a village? Variables in analysis Neighbourhood SES –Socio-Economic Indices for Area’s (SEIFA) Index of Advantage/Disadvantage (Trewin 2001) Controlling for selection bias -income, child’s gender and age, child is ATSI, single-parent household, in a household where no one worked, one or both parents were born overseas, and mother’s education And outcomes…
Outcome indexes Overall Mean score = 100 and SD = 10 Physical, learning and social/emotional Physical BMI, overall health, Physical health scale of PEDSQL, need more health care Social/emotional Strengths and Difficulties Questionnaire Learning Peabody Picture Vocabulary Test , parent & teacher ratings of reading skills & teacher ratings of writing & numeracy skills
Analyses Multilevel analyses1: Level 1 family and child factors including perceptions of the neighbourhood Level 2 neighbourhood level variables Population sampling weights 1Mixed
model, random coefficient model, hierarchical linear model, linear random effect regression
Neighbourhood socioeconomic status and children’s outcomes -overall
Source: Edwards, B. (2005). Family Matters, 72, 36-45.
Neighbourhood socioeconomic status and young children’s outcomes –social/emotional
Source: Edwards, B. (2005). Family Matters, 72, 36-45.
Neighbourhood socioeconomic status and young children’s outcomes –learning
Source: Edwards, B. (2005). Family Matters, 72, 36-45.
More disadvantaged neighbourhoods have a greater impact on boys than girls
Source: Edwards, B. (2005). Family Matters, 72, 36-45.
Pathways of influence- Community factors
What are the pathways of influence?
Traditional model of neighbourhood influences on children’s development Source: Roosa, Jones, Tein, & Cree. (2003). American Journal of Community Psychology, 31, 55-72.
What are the pathways of influence?
Neighbourhood SES
Internalising problems
Xue, Leventhal, Brooks-Gunn & Earls. (2005). Archives of General Psychiatry, 62, 554-563.
What are the pathways of influence? Collective efficacy Resident involvement in local organisations
Neighbourhood SES
Internalising problems
Xue, Leventhal, Brooks-Gunn & Earls. (2005). Archives of General Psychiatry, 62, 554-563.
What are the pathways of influence? Roosa and colleagues alternative model Neighbourhood perceptions
Neighbourhood Characteristics
Neighbourhood social processes
Family factors
Child outcomes
Source: Roosa, Jones, Tein, & Cree. (2003). American Journal of Community Psychology, 31, 55-72.
Present study
Test Roosa and colleagues’ model
Perceptions of neighbourhood mediate neighbourhood SES Children’s behavioral and emotional outcomes are, in turn, mediated by neighbourhood social processes
Focus on problematic and positive aspects of children’s behaviour Test influence of neighbourhood remoteness (ARIA) and observational measure of physical disorder
Method Sample Wave 1 of 4-5 year old cohort from LSAC (n = 4,983) 330 postcodes as “neighbourhoods” Variables Outcomes - prosocial behavior, conduct problems, hyperactivity, emotional symptoms and peer problems using the SDQ Family demographics –family income, child’s gender and age, child ATSI, single parent household, household where no-one was employed, household where parents were born overseas, maternal education
Neighbourhood variables Parental perceptions of neighbourhood facilities, belonging, cleanliness and safety derived from multilevel CFA (Edwards, 2006) Neighbourhood level SES (SEIFA), residential instability, ethnic homogeneity, remoteness (ARIA) Interviewer-rated physical disorder of the condition of nearby residence
Analyses Multilevel analyses1: Level 1 family and child factors including perceptions of the neighbourhood Level 2 neighbourhood level variables Population sampling weights 1Mixed
model, random coefficient model, hierarchical linear model, linear random effect regression
Conduct problems Model 1: Neighbourhood quality
Neighbourhood quality Neighbourhood SES % moved in last 5 years % born in Australia Remoteness Physical disorder
χ2(4) = 17.32, p
