Measurement of Home Environment: The Family Care Indicators Patrice Engle California Polytechnic State University Yuko Nonoyama-Tarumi UNICEF.

Measurement of Home Environment: The Family Care Indicators

Patrice EngleCalifornia Polytechnic State University

Yuko Nonoyama-TarumiUNICEF

Why Indicators for Family Care?

Caregiving Practices and

Resources

Quality of Interactions

with the Child

Child Development

Outcomes

Multiple Indicator Cluster Survey (MICS)

• Household Survey• Nationally representative sample• www.childinfo.org• MICS 3 (2005)

• 56 countries• Household module

• household characteristics, education, water and sanitation, nutrition, child labor, support HIV/AIDS orphans, etc.

• Women module• women’s characteristics, child mortality, maternal and newborn

health, marriage/union, HIV/AIDS knowledge, female genital mutilation, sexual behavior, etc.

• Children under five module• children’s characteristics, birth registration, early learning, breast

feeding, immunization, anthropometry, malaria, etc.

Development of Items

• Phase I: Item identification• Literature review• Meeting of global experts (Nov, 2002)

• Phase II: Item evaluation• Field tests in 7 countries (Spring, 2003)

• Qualitative analyses: Focus groups (Content validity)• Quantitative analyses: Frequency analyses (Discrimination)

• Phase III: Item selection• Meeting of global experts (Nov, 2003)

Domains selected

• Caregiving Practices• Quality of verbal

interaction• Learning/stimulating

activities• Limit setting and

discipline techniques• Responsiveness and

acceptance• Responsive feeding

• Caregiving resources• Caregiver stress• Caregiver physical

health• Caregiver knowledge• Alternate caregiver• Father’s involvement• Family cohesion• Social networks• Learning/stimulating

materials

Family Care Items in MICS3 (Core Early Learning Module: 52 countries)

(

Mother Father Other No one

Read books or look at picture books

Tell stories

Sing songs

Take outside the home, compound yard or enclosure

Play

Spend time naming, counting and/or drawing things

Learning/stimulating activities

Engage in any of the activities with the child (in the past 3 days) [multiple responses] (Asked to caretakers of children under 5 years old for each child)

Family Care Items in MICS3(Optional Child Development Module: 33 countries)

Learning/stimulating materials(Asked to caretakers of children under 5 years old once)

• Number of books• Number of children’s books• Play materials that child play with at home

• Household objects; Objects and materials found outside the living quarters; Homemade toys; Toys that come from a store; None

Alternate caregiver (in the last week)

• Number of times the child was left in the care of another child (younger than 10 years old)

• Number of times the child was left alone

Child Discipline Items in MICS3(Child Discipline Optional Module)

Setting Limits (Methods used in the past month)(Asked to caretakers of children 2-14 years old for a randomly selected child)• Non-violent

• Forbade something he/she liked• Explained why something was wrong• Gave him/her something else to do

• Psychological aggression• Shouted, yelled at or screamed at him/her• Called him/her dumb, lazy, etc

• Minor physical assault• Shook him/her• Spanked, hit or slapped him/her on the bottom with bare hand

• Severe physical assault• Hit him/her on the body with something a belt, stick, etc• Hit or slapped him/her on the face, head or ears• Hit or slapped him/her on the hand, arm, or leg• Beat him/her with an implement

• Do you believe that in order to bring up properly, you need to physically punish him/her

ECD Indicators in MICS3

Indicator Definition

Support for learning % of children aged 0-59 months living in households in which an adult has engaged in four or more activities to promote learning and school readiness in the past 3 day

Father’s support for learning

% of children aged 0-59 months whose father has engaged in one or more activities to promote learning and school readiness in the past 3 days

Support for learning: non-children’s books

% of households with three or more non-children’s books

Support for learning: children’s books

% of households with three or more children’s books

Support for learning: materials for play

% of households with three or more materials intended for play

Non-adult care % of children aged 0-59 months left alone or in the care of another child younger than 10 years of age in the past week

Pre-school attendance % of children aged 36-59 months that attend some form of early childhood education programme

School readiness %of children in first grade that attended some form of pre-school the previous year

Preliminary cross-national analyses

• To what extent do countries differ in their level of family care?

• To what extent is positive family care equally distributed within the country?

Learning/stimulating activities (four or more) by wealth

0

10

20

30

40

50

60

70

80

90

100

Gu

inia

Bis

sau

Bu

rkin

a F

aso

Bu

run

di

Mau

rita

nia

To

go

Co

te d

'Ivo

ire

Gh

an

a

Cam

ero

un

Iraq

Gam

bia

Ban

gla

de

sh

CA

R

Syr

ia

Mo

ng

olia

Vie

tnam

Ta

jikis

tan

Alg

eri

a

Pal

est

inia

ns

in

Nig

eri

a

Sie

rra

Leo

ne

So

mal

ia

Pal

est

inia

ns

in S

yri

a

Alb

an

ia

Su

rin

ame

Kyr

gys

tan

Uzb

eki

sta

n

Bo

sn

ia&

Her

zeg

ov

ina

Th

aila

nd

Kas

akh

stan

Gu

yan

a

Geo

rgia

Bel

aru

s

Ser

bia

Mac

edo

nia

Bel

ize

Jam

aic

a

Mo

nte

neg

ro

Tri

nid

ad

an

d T

ob

ag

o

Countries

Per

cen

tag

e

Poorest (%) Richest (%) Average

Average=61%Median= 64.6%Average Ratio (poorest/richest)= 0.76Correlation (prevalence, inequality)=.46

Non-children's books (three or more) by wealth

0

10

20

30

40

50

60

70

80

90

100

Pal

est

inia

ns

inL

eb

ano

n CA

R

To

go

Sie

rra

Leo

ne

Co

te d

'Ivo

ire

*

Nig

eri

a

Kyr

gys

tan

Gh

an

a

Ta

jikis

tan

Cam

ero

un

Mac

edo

nia

Mo

ng

olia

Gu

yan

a

Alb

an

ia

Pal

est

inia

ns

in S

yri

a

Su

rin

ame

Vie

tnam

Alg

eri

a

Syr

ia

Th

aila

nd

Bel

ize

Bo

snia

&H

erze

go

vin

a

Ser

bia

Uzb

eki

sta

n

Mo

nte

neg

ro

Geo

rgia

Jam

aic

a

Kas

akh

sta

n

Tri

nid

ad

an

d T

ob

ag

o

Ukr

ain

e

Country

Per

cen

tag

e

Poorest (%) Richest (%) Average

Average=58.1%Median=59.6%Average Ratio (poorest/richest)=0.54Correlation (prevalence, inequality)=.69

Children's books (three or more) by wealth

0

10

20

30

40

50

60

70

80

90

100

Co

te d

'Ivo

ire

To

go

CA

R

Cam

ero

un

Sie

rra

Leo

ne

Gh

ana

Nig

eria

Taj

ikis

tan

Pal

esti

nia

ns

inL

eban

on

Vie

tnam

Mo

ng

olia

Alg

eria

Syr

ia

Alb

ania

Pal

esti

nia

ns

in S

yria

Uzb

ekis

tan

Th

aila

nd

Su

rin

ame

Mac

edo

nia

Gu

yan

a

Jam

aica

Bel

ize

Kas

akh

stan

Bo

snia

&H

erze

go

vin

a

Geo

rgia

Kyr

gys

tan

Mo

nte

neg

ro

Ser

bia

Tri

nid

ad a

nd

To

bag

o

Ukr

ain

e

Country

Per

cen

tag

e

Poorest (%) Richest (%) Average

Average=41.6%Median=42.3%Average Ratio (poorest/richest)=0.36Correlation (prevalence, inequality)=.83

Inadequate care (left in the care of another child or left alone) by wealth

0

10

20

30

40

50

60

70

Tri

nid

ad

an

d T

ob

ag

o

Jam

aic

a

Bel

ize

Uzb

eki

sta

n

Mo

nte

neg

ro

Bo

sn

ia&

Her

zeg

ov

ina

Su

rin

ame

Geo

rgia

Alg

eri

a

Ser

bia

Mac

edo

nia

Kas

akh

stan

Ukr

ain

e

Kyr

gys

tan

Gu

yan

a

Ta

jikis

tan

Alb

an

ia

Mo

ng

olia

Th

aila

nd

Pal

est

inia

ns

in S

yri

a

Syr

ia

Gam

bia

Vie

tnam

Pal

est

inia

ns

inL

eb

ano

nS

ierr

a L

eon

e

Gh

an

a

To

go

Cam

ero

un

Nig

eri

a

CA

R

Co

te d

'Ivo

ire

Country

Per

enta

ge

Poorest (%) Richest (%) Average

Average=16.1%Median=12.6%Average Ratio (Poorest/Richest)=2.2Correlation (prevalence, inequality)=-.18

How well do these scales work?

• Item comparison across countries

• Validation on the HOME and Bayley Scales

• Validation within country data

• Recommendations for next steps

Descriptive data on activities by country

• Selected three countries with publicly available data from different parts of the world• Kyrgyzstan (n=2987) Bangladesh (n=34710)

and Sierra Leone (n=5904)

• Examined activities separately to see which have reasonable variability and if they vary as expected

Activities anyone did: Percent of households

0

10

20

30

40

50

60

70

80

90

100

Readbooks

Stories Songs Outside Play Naming

Kyrgyzstan Sierra Leone Bangladesh

Sources of toys: percent of households

0

10

20

30

40

50

60

70

80

90

household outside handmade purchased none

Kyrgyzstan Sierra Leone

Conclusions based on descriptive data

• Differences by country are reasonable

• All families do something

• Some questions have little variability (e.g., taking child outside, play with child).

Validity study: Bangladesh

• 800 children at 18 months• HOME • Bayley MDI and PDI• Language Comprehension and Expression

• 129 of them also measured at 12 months on same measures

• 40 given 7-14 week test-retest on Activities and Toys

Grantham-McGregor, Hamadani, and Engle, 2008

Measures

• 6 activity items• Play – “play with toys” rather than “play”

• Sources of toys

• Variety of toys

• Books

• Childcare situation

Reliability

Measure 1 week (18 months)

12 to 18 months

Family Activity Score

.64** .57**

Number of sources of toys

.91** .09

N 40 129

Associations of Activity Index with Outcome Measures (n=798)

FCI FCI regr+

Source of toys

Source regr

HOME .72** ** .39** ns

MDI .29** * ns ns

PDI .19** ns .09* ns

Language .44** ** .23** ns

+Controlling for maternal education, wealth, family size, birthweight, gestational age, paternal education, income, age, gender, other family care measures

Means of MDI by Number of Family Activities

controlling for age (N=800; 18 months)

6668707274767880828486

1 2 3 4 5 6

Mean MDI

ANOVA significant at p<.001

Conclusion

• Family Activity Index appears to be reliable and valid

• Increases with MDI in a linear fashion – no clear cut-off

• Sources of toys is not so strong

• Variety of toys much stronger (not reported here)

Validity assessment with MICS data: Bangladesh (N=34,710)

• Internal consistency• Association of items with age• Associations with maternal education, household

wealth, gender• Associations with two parent report measures:

Do you do anything to prepare your child for school (3 and 4 only); and do you do things to develop your child’s intelligence

• Value of individual activity questions

Which items are related to age? Bangladesh, N= about 34,000

Total activities .34**

Read books .47**

Tell stories .24**

Sing Songs .01

Take outside .08*

Play .02

Naming, counting, drawing .40**

Do something to develop intelligence .17**

Prepare for school .15**

Internal consistency of Index: Alpha = .734

Measure Item-total correlation

Read books .54

Tell stories .60

Sing Songs .48

Take Outside .29

Play .38

Names, counts, draw .52

Correlations of items with SES measures controlling for age

M Ed Wealth Develop IQ

Prepare school

Total .34 .25 .31 .29

Read Books .35 .23 .24 .35

Tell stories .25 .23 .19 .16

Sing songs .21 .17 .19 .12

Outside .07 .05 .12 .06

Play .15 .13 .15 .09

Name, count. .25 .16 .24 .02

Conclusions

• Family Activities Scale works quite well

• Sources of toys functions less well

• Need more work to define a cut-off point – four or more activities may not be the best

Recommendations

• Analyze role of fathers separately• Make a separate code for some activities

such as “read books” • May revise wording on some questions

• Could replace “take outside”• Might use “play with toys” rather than “play”

• Complete analyses with the rest of the countries

• Apply and use for Advocacy