23

Developments with ONS’ Small Area Population Estimates Project Andy Bates

  • View
    217

  • Download
    1

Embed Size (px)

Citation preview

Developments with ONS’ Small Area Population Estimates Project

Andy Bates

Outline of talk

● Current Position

● Ratio Change method

● Data Sources used in estimates

● User Consultation

● Geography Considerations

● Future publication and work plans

● Current Position

● Ratio Change method

● Data Sources used in estimates

● User Consultation

● Geography Considerations

● Future publication and work plans

Current Position

● Produced mid-2001 and mid-2002 CAS ward estimates– quinary age and sex

● Published as experimental statistics– April 2005

● User consultation– 12 weeks

● Available on Neighbourhood Statistics this week

Coverage Dataset 0-14 15-64 65+ Child Benefit Older Persons Dataset Patient Registers

Eg to derive ratios for 0-4 year olds by sex, we create dataset specific ratios:

Eg CB ratio = Year 2 dataset count

Year 1 dataset count

0-4 ratio = Child Benefit ratio + Patient Register ratio

2

Ratio Change method

Base population

Year 1SP x ratioSP

Year 2

Population

2nd period

Ratio Change method

Constrain to LA MYEs less SP

Data Sources - Child Benefit

● Now provided by Inland Revenue– previously provided by DWP

● Receive counts for 0-4’s, 5-9’s & 10-14’s by sex

● Data provided for Lower Layer SOAs

● Good coverage of children 0-14 – undercount (mid-2003) compared to MYEs 1.7%

● Data consistent over time – overall counts– geographically

Data Sources - Older Persons Dataset

● Provided by DWP

● Comprises a number of benefit databases– eg State Pension, Widows Benefit & Winter Fuel

Allowance

● Receive counts by quinary age group & sex– 65-69, 70-74, 75-79 & 80+

● Data provided for Lower Layer SOAs

● Good coverage of elderly population– undercount (mid-2003) compared to MYEs 1.3%

● Data consistent over time & geographically

Data Sources - Patient Registers

● Used by ONS to calculate internal migration

● Get postcode counts by single year of age & sex– can easily aggregate to different geographies

● Issue of list inflation– mid-2004 PR counts > 2004 MYEs by 2.9m (5.4%)

● Data not always consistent over time– problems in student areas– problems caused by list cleaning

Data Sources - Patient Registers

2

2.1

2.2

2.3

2.4

2.5

2.6

2.7

2.8

2.9

3

mid-1999 mid-2000 mid-2001 mid-2002 mid-2003 mid-2004

Mill

ions

Patient Register counts less MYEs

Data Sources - Patient Registers

2004 Patient Registers and 2004 MYEs for England & Wales

-2

0

2

4

6

8

10

12

14

16

0-4 5-9 10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-79 80-84 85+

Age Group

Males (7.5%)

Females (3.4%)

Pe

rce

nta

ge

Diff

ere

nce

(P

R-M

YE

s)

Data Sources - Patient Registers

Patient Register Counts

Mid-2000 8,400Mid-2001 8,512Mid-2002 12,039Mid-2003 13,152Mid-2004 12,926

Patient Register counts - 00BKGQ St James's, City of Westminster

0

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

4,500

0-4

5-9

10-1

4

15-1

9

20-2

4

25-2

9

30-3

4

35-3

9

40-4

4

45-4

9

50-5

4

55-5

9

60-6

4

65-6

9

70-7

4

75-7

9

80-8

485

+

Mid-2000 Mid-2001 Mid-2002 Mid-2003 Mid-2004

-1.7

9.2

41.4

1.3

-10 0 10 20 30 40 50

Year tomid-2004

Year tomid-2003

Year tomid-2002

Year tomid-2001

% Change

User Consultation

● 12-week period of consultation

● 28 responses received– Govt. depts/agencies 2

– Regional Health Observatories 1– Local Government

• County Councils 6• District Councils 3• Joint Strategy Units & GLA 3• London Boroughs/Unitaries 13

● 21 organisations completed the response form

User Consultation

● Requirement for alternative age groupings– 13-19, 16-19 & 18-19

● Quality of the estimates:– Excellent 1 6%– Good 13 72%– Fair 2 11%– Poor 2 11%

● Clear requirement for ward estimates– CAS Wards 3 15%– Standard Table Wards 0– Statistical Wards 17 85%

User Consultation

● Favourable reaction to the estimates

● Gives some early indication of Ratio Change suitability

However● Base population issue identified in wards with

large Armed Forces presence – methodological solution found

● Base population distribution - evidence in 2 LAs of ward overestimation and underestimation– further investigation required

Geography Considerations

Geography Number Minimum

population Average

population

Lower Layer SOAs 34,378 1,000 1,500

Middle Layer SOAs 7,193 5,000 7,400

Upper Layer SOAs ? 25,000? ?

CAS Wards 8,850 <100 6,000

Local Authorities 376 2,200 141,100

Geography Considerations – eg Dover, Kent

Middle Layer SOA Estimates

Statistical Ward Estimates

MSOA estimates less SP constrained to Patient Register postcode counts

Patient Register MSOA constrained counts aggregated to Statistical Wards

Statistical Wards overlaid onto Patient Register postcodes

Add back in special population

Remove special populationGeography Considerations – SOAs & Wards

Publication Plans

● Mid-2001, mid-2002 & mid-2003 SOA estimates– Lower Layers by broad age group & sex– Middle Layers by quinary age group & sex

– publication provisionally planned for early 2006

● Mid-2003 Statistical Ward estimates– only if suitable methodology can be found– problematic because base population may be needed– lack of time series when wards change– if recasting methodology suitable, publish in 2006

Work Plans

● Consider methodological enhancements for 3 shortlisted methods– Apportionment

– Cohort component

– Ratio Change

● Evaluate estimates for mid-2002, mid-2003 & mid-2004 from all 3 shortlisted methods

● Following evaluation identify a preferred method

● Consider transition from Experimental Statistics to National Statistics

Contact Information

Email the project team at:

[email protected]

Updated information on the NS website at:

www.statistics.gov.uk/SAPE

Any questions?