Missing%20income%20data%20in%20the%20millennium%20cohort%20study:

1

• Missing income data in the millennium cohort
  study
• Evidence from the first two sweeps
• Authors Denise Hawkes and Ian Plewis
• Discussant Nicholas Biddle
• nicholas.biddle_at_anu.edu.au

2
Introduction and overview

• Data Millennium Cohort Study
• Research questions What are the factors
  associated with non-response? More specifically
• Are there within household and individual
  correlations for missing income data?
• Is the sex of the interviewer an important
  explanatory variable?
• How is missing data in sweep one related to
  missing data in sweep two?
• Is attrition at sweep two related to the level of
  household income or the failure to provide data
  in sweep one?
• Method
• Descriptive analysis
• Binary and Multinomial Logit models with
  non-response as dependent variable
• Binary Logit with attrition between sweep one and
  sweep two as dependent variable

3
Data

• Millennium Cohort Study
• First sweep 18,819 babies born in the UK from
  1st September 2000 (from 18,552 families).
  Interviewed when baby was 9 months old
• Second Sweep 14,898 families from original
  sample and 692 new families. Interviewed when
  children around 3 years old.
• Information from main respondent (usually mother)
  and partner of respondent (usually father)
• Incomplete information on income through
• Unit non-response (response rate 72 in first
  sweep)
• Partner non-response (88 of families with
  partners responded)
• Item non-response for income (6 of main
  respondents and partners did not provide income
  data)
• Attrition between sweeps (79 of eligible
  families responded in sweep two)
• Income information
• Collected from those currently doing paid work,
  those who have a paid job but are on leave, those
  who have worked in the past but have no current
  job.
• For employees total take home pay and gross pay
• For self employed amount you personally took
  out of the business after all taxes and costs

4
Data

• Millennium Cohort Study
• First sweep 18,819 babies born in the UK from
  1st September 2000 (from 18,552 families).
  Interviewed when baby was 9 months old
• Second Sweep 14,898 families from original
  sample and 692 new families. Interviewed when
  children around 3 years old.
• Information from main respondent (usually mother)
  and partner of respondent (usually father)
• Incomplete information on income through
• Unit non-response (response rate 72 in first
  sweep)
• Partner non-response (88 of families with
  partners responded)
• Item non-response for income (6 of main
  respondents and partners did not provide income
  data)
• Attrition between sweeps (79 of eligible
  families responded in sweep two)
• Income information
• Collected from those currently doing paid work,
  those who have a paid job but are on leave, those
  who have worked in the past but have no current
  job.
• For employees total take home pay and gross pay
• For self employed amount you personally took
  out of the business after all taxes and costs

5
Data

• Millennium Cohort Study
• First sweep 18,819 babies born in the UK from
  1st September 2000 (from 18,552 families).
  Interviewed when baby was 9 months old
• Second Sweep 14,898 families from original
  sample and 692 new families. Interviewed when
  children around 3 years old.
• Information from main respondent (usually mother)
  and partner of respondent (usually father)
• Incomplete information on income through
• Unit non-response (response rate 72 in first
  sweep)
• Partner non-response (88 of families with
  partners responded)
• Item non-response for income (6 of main
  respondents and partners did not provide income
  data)
• Attrition between sweeps (79 of eligible
  families responded in sweep two)
• Income information
• Collected from those currently doing paid work,
  those who have a paid job but are on leave, those
  who have worked in the past but have no current
  job.
• For employees total take home pay and gross pay
• For self employed amount you personally took
  out of the business after all taxes and costs

6
Patterns of income response

• Original sample (paper has information on new
  families and proxies)

Sweep one Sweep one Sweep two Sweep two Sweep two
Main Partner Partner Main Partner Partner Partner
Income response 45.9 64.7 64.7 50.6 62.9 62.9 62.9
Dont know 1.8 2.1 2.1
Refusal 0.9 2.1 2.1
Total non-response 2.7 4.3 4.3 4.4 8.7 8.7 8.7
Not applicable 51.5 31.0 31.0 45.1 28.4 28.4 28.4
Sample 18,552 18,552 14,898 14,898
7
Patterns of income response

• Original sample (paper has information on new
  families and proxies)

Sweep one Sweep one Sweep two Sweep two Sweep two
Main Partner Partner Main Partner Partner Partner
Income response 45.9 64.7 64.7 50.6 62.9 62.9 62.9
Dont know 1.8 2.1 2.1
Refusal 0.9 2.1 2.1
Total non-response 2.7 4.3 4.3 4.4 8.7 8.7 8.7
Not applicable 51.5 31.0 31.0 45.1 28.4 28.4 28.4
Sample 18,552 18,552 14,898 14,898
8
Partner and main respondent income response
Sweep one
Partner respondent Partner respondent Partner respondent
Dont know/ refusal Dont know/ refusal Not applicable Income response Income response
Dont know/refusal (464) 26.6 26.6 27.4 45.9 45.9
Main respondent Not applicable (10,264) 3.9 3.9 42.7 53.4 53.4
Income response (7,824) 3.5 3.5 18.0 78.5 78.5
9
Partner and main respondent income response
Sweep two
Partner respondent Partner respondent Partner respondent
Dont know/ refusal Dont know/ refusal Not applicable Income response Income response
Dont know/refusal (614) 26.7 26.7 29.0 44.3 44.3
Main respondent Not applicable (7,190) 9.6 9.6 36.5 54.0 54.0
Income response (7,094) 6.5 6.5 21.0 72.5 72.5
10
Sweep one and sweep two income response Main
respondent
Sweep two Sweep two Sweep two
Dont know/ refusal Dont know/ refusal Not applicable Income response Income response
Dont know/refusal (357) 17.9 17.9 26.7 55.4 55.4
Sweep one Not applicable (7,733) 2.9 2.9 74.4 22.8 22.8
Income response (6,504) 5.3 5.3 14.8 79.9 79.9
11
Sweep one and sweep two income response Partner
Sweep two Sweep two Sweep two
Dont know/ refusal Dont know/ refusal Not applicable Income response Income response
Dont know/refusal (501) 35.2 35.2 0.4 64.4 64.4
Sweep one Not applicable (1,778) 22.9 22.9 2.4 74.7 74.7
Income response (7,433) 8.7 8.7 0.1 91.2 91.2
12
Modelling non-response Main respondent
Sweep one Sweep two
Spec. (I) Spec. (II) Spec. (III)
Self employed 6.4 6.8 6.6 6.7
Has a partner 0.58 0.57 0.56
Social class Intermediate 1.6
- Reference managerial Small employers and self employment 1.8
and professional Lower supervisors and technical
Semi routine and routine
Ethnicity Mixed
- Reference white Indian 2.4 2.3 2.3
Pakistani and Bangladeshi
Black or Black British 1.6
Other ethnic group 2.3
Country Wales
- Reference England Scotland
Northern Ireland 1.7 1.5
Respondent did not respond in sweep one - - 3.0 3.0
Respondent same in sweep one and two - - - 5.3
Sample Size 8,190 5,800 5,800 5,800
13
Modelling non-response Main respondent
Sweep one Sweep two
Spec. (I) Spec. (II) Spec. (III)
Self employed 6.4 6.8 6.6 6.7
Has a partner 0.58 0.57 0.56
Social class Intermediate 1.6
- Reference managerial Small employers and self employment 1.8
and professional Lower supervisors and technical
Semi routine and routine
Ethnicity Mixed
- Reference white Indian 2.4 2.3 2.3
Pakistani and Bangladeshi
Black or Black British 1.6
Other ethnic group 2.3
Country Wales
- Reference England Scotland
Northern Ireland 1.7 1.5
Respondent did not respond in sweep one - - 3.0 3.0
Respondent same in sweep one and two - - - 5.3
Sample Size 8,190 5,800 5,800 5,800
14
Modelling non-response Partner (I)
Sweep one Sweep two
Spec. (I) Spec. (II) Spec. (III)
Self employed 1.7 3.6 3.6 3.6
Social class Intermediate
- Reference managerial Small employers and self employment 3.0
and professional Lower supervisors and technical 0.68
Semi routine and routine 0.66
NVQ Level 1
NVQ Levels NVQ Level 2 0.63
- Reference none NVQ Level 3 0.59
NVQ Level 4 0.47
NVQ Level 5 0.34
Other/overseas qual only
Ethnicity Mixed 2.3 2.4 2.5
- Reference white Indian 1.8 2.5 2.3 2.3
Pakistani and Bangladeshi 2.2 2.4 2.2 2.2
Black or Black British
Other ethnic group 2.0
Owner occupier 0.76 0.76 0.77
15
Modelling non-response Partner (I)
Sweep one Sweep two
Spec. (I) Spec. (II) Spec. (III)
Self employed 1.7 3.6 3.6 3.6
Social class Intermediate
- Reference managerial Small employers and self employment 3.0
and professional Lower supervisors and technical 0.68
Semi routine and routine 0.66
NVQ Level 1
NVQ Levels NVQ Level 2 0.63
- Reference none NVQ Level 3 0.59
NVQ Level 4 0.47
NVQ Level 5 0.34
Other/overseas qual only
Ethnicity Mixed 2.3 2.4 2.5
- Reference white Indian 1.8 2.5 2.3 2.3
Pakistani and Bangladeshi 2.2 2.4 2.2 2.2
Black or Black British
Other ethnic group 2.0
Owner occupier 0.76 0.76 0.77
16
Modelling non-response Partner (II)
Sweep one Sweep two
Spec. (I) Spec. (II) Spec. (III)
Country Wales
- Reference England Scotland
Northern Ireland 1.9 1.5 1.6 1.6
Respondent did not respond in sweep one - - 4.6 4.5
Respondent same in sweep one and two - - - 0.39
Sample Size 10,754 7,893 7,893 7,893
17
Other modeling Multinomial Logit and attrition

• Multinomial Logit Response vs. dont know vs.
  refuse
• Main respondent
• Self employed only significantly more likely to
  be dont know not refusal
• Same with social class variables
• Black or Black British as well as Northern
  Ireland more likely to refuse
• Partner respondent
• Self employed significantly more likely to refuse
  and not know
• NVQ levels and ethnicity both associated with
  refusal
• Attrition at sweep two
• Higher income in sweep one associated with lower
  odds of attrition between sweep one and sweep two
• Main income and partner income non-response in
  sweep one associated with higher odds of
  attrition between sweep one and sweep two

18
Other modeling Multinomial Logit and attrition

• Multinomial Logit Response vs. dont know vs.
  refuse
• Main respondent
• Self employed only significantly more likely to
  be dont know not refusal
• Same with social class variables
• Black or Black British as well as Northern
  Ireland more likely to refuse
• Partner respondent
• Self employed significantly more likely to refuse
  and not know
• NVQ levels and ethnicity both associated with
  refusal
• Attrition at sweep two
• Higher income in sweep one associated with lower
  odds of attrition between sweep one and sweep two
• Main income and partner income non-response in
  sweep one associated with higher odds of
  attrition between sweep one and sweep two

19
Summary

• Household and individual correlations for missing
  income data
• Self employment, some ethnic groups (though not
  consistent), Northern Ireland
• The sex of the interviewer is not an important
  explanatory variable in explaining income
  non-response
• Some variables only associated with dont know
  or refusal only
• Missing data in sweep one associated with higher
  odds of missing data in sweep two
• Especially amongst partner respondents
• Higher household income in sweep one associated
  with lower attrition in sweep two
• Missing data in sweep one associated with higher
  attrition in sweep two

20
Suggested further work and information

• Models for non-response
• More diagnostic information (e.g. tests of group
  significance)
• Information on the child?
• Interviewer bias
• Multilevel model?
• Interactions or other information on the
  interviewer
• Implications for survey design
• Difference between dont know and refusal

21
(No Transcript)