GEOG 2003: Methods in Human Geography Census Analysis

1
GEOG 2003 Methods in Human GeographyCensus
Analysis

• Jürgen Essletzbichler
• (j.essletzbichler_at_geog.ucl.ac.uk)
• http//www.geog.ucl.ac.uk/jessletzbichler/geog200
  3

2
Overview

• Continue with course work topics
• Continue thinking about research questions
• Use CASWEB to uncover more complex relationships
  and ask more sophisticated research questions

3
General goals

• How to proceed?
• what is the issue define your research
  questions?
• what is current status using Census data?
• describe current geographical patterns (e.g.
  spatial concentration, unevenness, etc)
• how are things changing? what is being done?
• what are strengths and weakness? how could this
  be explored with other methods

4
Useful London ideas sources

• present the official style of analysis on
  current social problems through broadly
  quantitative approaches

http//www.statistics.gov.uk/downloads/ theme_comp
endia/FOL2003/00Prelims.pdf
http//www.london.gov.uk/mayor/strategies/sds/lond
on_plan/lon_plan_all.pdf
5
Employment topic

• Focus on London, 2003, The service sector
  dominates Londons economy. Key services among
  the sector are finance and business services, the
  public sector, tourism and hospitality and the
  creative and cultural industries. Finance and
  business services alone accounted for 35 per cent
  of Londons Gross Value Added (GVA) in 1998.
• Focus on London 2003, chapter 6 labour markets
• A key concern is changing patterns employment
• Social-spatial patterns in unemployment,
  especially long-term unemployment.
• Ethnic and gender differences in occupation,
  unemployment, etc.?

6
(No Transcript)
7

• http//www.london.gov.uk/gla/publications/factsand
  figures/DMAG-briefing202003-26.pdf

8
Unemployment rates

• Pronounced spatial variation
• Ward unemployment rates in London vary widely
  from 2.2 per cent in Nonsuch ward (Sutton) up to
  17.5 per cent in Bromley-by-Bow ward (Tower
  Hamlets).
• The level of variation in rates within some
  boroughs is considerable. For example,within
  Kensington and Chelsea, unemployment rates in
  Golborne ward (16.5 per cent) are four times
  higher than in Campden ward (3.8 per cent).
• In some boroughs such as Newham and Hackney,
  unemployment levels across all wards are
  consistently high. In these two boroughs, all
  wards have unemployment rates lying between 9.4
  and 16.1 per cent.
• Almost one in five (18 per cent) of Londons
  wards have unemployment rates of ten per cent or
  more most of these are in Inner London where
  38 per cent of wards have unemployment rates of
  ten per cent or more.
• Nearly two thirds (62 per cent) of Londons wards
  have rates above the national average of 5.2 per
  cent. The majority of wards in Inner London (89
  per cent of wards) have rates above the national
  average.
• While unemployment rates are low in Outer London
  relative to Inner London, it still has pockets of
  high unemployment four per cent of wards in
  Outer London had rates of 10 per cent. 45 per
  cent of all Outer London wards had unemployment
  rates above the national average of 5.2 per cent.

9

• A GLA research paper provides overview of
  complexities of measuring unemployment
• Can get very different answers depending on the
  measurement used
• Number of Londoners who are unemployed can range
  from 150,000 up to 580,000 depending on how
  narrowly or widely you define unemployment

http//www.london.gov.uk/gla/publications/factsand
figures/DMAG-Briefing202004-9.pdf
10
(No Transcript)
11
Ethnicity and Identity theme

• Focus on London 2003 Over 300 languages are
  spoken and the 2001 Census shows that 29 per cent
  of Londons population belonged to a minority
  ethnic group. London has an unrivalled cultural
  and artistic heritage, helping to confirm its
  place as one of the most vibrant and dynamic
  cities in the world.
• London is unique. The non-White population of
  the UK is concentrated in the large urban
  centres. Nearly half (45 per cent) lived in the
  London region in 2001, where they comprised 29
  per cent of all residents.
• Ethnicity is often used as explanatory for in
  social disadvantage.
• But ethnicity is not the cause

12
(No Transcript)
13
(No Transcript)
14

• what is the geography of religion in London?

15
Education

• What are implications for London?
• Focus on London 2003, chapter 7 education and
  training

16
Spatial patterns of educational attainment
Percent of people with no qualification

• Percent of people with no/some/lots of
  qualifications
• Ethnic / gender differences
• Relationship to unemployment / health?

Source Census 2001, Qualifications and Students
17
(No Transcript)
18
Wealth, poverty inequality

• The BIG issue one in some senses MORE DETAIL in
  lecture 8 slides
• Research to understand patterns of social
  difference and trying to explain the processes
  driving inequalities
• Much of government intervention across all manner
  of policy areas (health, taxation, education,
  criminal justice) is about dealing with impacts
  of inequalities and trying to reduce poverty in
  the future
• Ideological issues on the inevitability of
  inequality in society.
• Is is necessary, even beneficial, in a capitalist
  system to differentiate people? Isnt it right
  that more talented, hardworking people are
  entitled to be richer? Is it the case that more
  hardworking people are richer (think about the
  illegal immigrants working 14 hour days)? Should
  society let less able people fail?
• But how far is it really a meritocracy?
• Hot political issue
• Related to all other kinds of issues (health,
  education, ethnicity, housing, )

19
Extreme levels of inequality in London
20
Wealth poverty in London

• Focus on London 2003, Whilst London stands out
  as one of the most successful regions in the UK,
  it is also a city divided between the extremes of
  wealth creation and success, and deprivation and
  social exclusion. This is reflected in the high
  proportion of wards (20 per cent) which are in
  the 10 per cent most deprived wards in England.
• What are the patterns of deprivation across
  London?
• Is inequality increasing? Greater social
  polarisation? Is this reflected spatially? What
  are the implications? What can be done?
• How does London compare to the rest of the UK? Is
  it a special case? Focus on London 2003,
  Unemployment in London is higher than the
  national average. The Inner London rate is twice
  the national average and rates are particularly
  high among Londons minority ethnic groups.
• But large variation among ethnic groups,
  inner-outer London, WHY?

21

• London Divided Income Inequality and Poverty in
  the Capital (GLA, Nov. 2003)
• http//www.london.gov.uk/mayor/economy/docs/london
  _divided_all.pdf
• Does not use the Census as a data source. Why?
• Largely benefit data (routine administration)
• ONS sample national surveys

• The issue Once housing costs are taken into
  account, 41 per cent of children in London are
  living in income poverty, the highest rate in
  Great Britain. In Inner London, this rises to one
  in two children. Poverty rates for working age
  adults and pensioners are also far higher in
  Inner London than in any region of Great Britain.
  Shockingly, 36 per cent of Inner London
  pensioners are living in poverty, compared to
• 25 per cent nationally.

22
Broad policy agenda as set out in The London Plan
(p. 310) to solve inequalities
23
Measuring wealth and poverty

• Many practical difficulties in accurately
  measuring the wealth / poverty of individual,
  households and areas
• Material poverty and wealth in terms of income is
  known for most people (tax records, benefit
  claimant records)
• Can collapse the other 5 topics into broader
  deprivation explanation. for example, health
  inequalities are often believed to be significant
  element in overall deprivation
• Background reading
• Focus on London, 2003, chapter 8 living in
  London
• Lee P, (1999) "Where are the deprived? Measuring
  deprivation in cities and regions" in Dorling D,
  Simpson S, (eds) Statistics in Society The
  Arithmetic of Politics (Arnold, London), pp.
  172-180. Senate House Library, SOCIOLOGY 6th
  Flr, K45 Sta
• MORE ON THAT NEXT TUESDAY

24

• The Census has no direct question on poverty or
  income
• Several possible proxies - car ownership, house
  size (no rooms per person), occupation (social
  class), unemployment, deprivation indices

25
(No Transcript)
26
Reading Official Statistical reports

• Useful to read Focus on London and GLA reports
  for their style, format and modes of analysis and
  data presentation
• These are professionally produced as a means to
  describe social situations and drive official
  action
• They are key element of discourses of social
  geography
• The quality of brief descriptions and clear
  analysis and presentation are what you should be
  aiming for in your dissertations
• Blending together text, charts and small tables
  to build a convincing arguments. statistics are
  tactically deployed to demonstrate understanding
  and give authority (numbers never lie)
• Full supporting tables in the appendices

27
Case study Ethnicity Employment

• Ethnic minority unemployment and spatial
  mismatch the case of London, Urban Studies, 1999
  
• author Edward Fieldhouse, University of
  Manchester
• Research Questions context of the validity of
  spatial mismatch theory to explain the high
  unemployment level of ethnic minorities
• investigate the effect on unemployment of the
  geographical distribution of ethnic minorities
  within Greater London
• Data Sources 1991 Census, London wards.
  acknowledges limits due to undercount, but argues
  Census remains the most complete and reliable
  data source on minority ethnic populations in
  Britain

28
Ethnicity and employment (Fieldhouse)

• Analytical Methods Descriptive statistics on
  social patterns of unemployment in ethnic groups
  and by social characteristics (especially housing
  tenure),
• logistic regression modelling and summary tables
• numerous thematic maps of spatial patterns
• Results Overall, although the geography of
  unemployment is not co-terminous with the
  geography of ethnicity, there is a very high
  degree of similarity, in particular the
  inner-outer pattern, and the coincidence of
  unemployment and minority ethnic populations to
  the north-east and the south of the City of
  London.
• whilst this research has demonstrated a clear
  relationship between minority ethnic unemployment
  and the characteristics of areas in which
  minorities live, there is no evidence that this
  is a product of the spatial mismatch of people
  and jobs.
• most plausible that this results from a
  combination of the residential segregation of the
  most vulnerable groups

29
Combining poverty, ethnicity and education

• You want to establish causal relationships
• Example
• Why are certain ethnic groups, on average, poorer
  than others?
• Hypothesis 1 They are unable to obtain jobs
• Hypothesis 2 They cannot obtain those jobs
  because they are less educated
• Hypothesis 3 Some ethnic minorities are not able
  to obtain access to education
• Hypothesis 4 Ethnic minorities in different
  areas have different educational attainment
  (there is a social as well as spatial component
  to educational attainment)
• Try to answer/explain hypotheses

30
Ethnicity and unemployment
31
Percent of people with no qualifications and
percent Bengali
no qualifications Bengali
32
Correlation is positive, but relationship looks
rather weak Why is that???
33
(No Transcript)
34
(No Transcript)
35
Use Casweb for cross-tabulations
http//census.ac.uk/casweb/
36
(No Transcript)
37
(No Transcript)
38
(No Transcript)
39
(No Transcript)
40
(No Transcript)
41
(No Transcript)
42
(No Transcript)
43
2
1
44
(No Transcript)
45
(No Transcript)
46
Variable code in Excel Table matches code in
CASWEB table tt0130681 means Casweb table tt013
(ethnicity) and cell 0681 (all people in London
with no qualifications)
47
Then create the percentages you are interested in
48
CASWEB options

• Lots of cross-tabulations possible
• From London to CASWARD level (if you want to
  compare individual wards, etc.)
• Will help you to establish causal relations

49
Case study A divided nation?

• A nation still dividing the British census and
  social polarisation 1971-2001, Environment and
  Planning A, 2003
• authors Danny Dorling, Phil Rees, University of
  Leeds
• Research Questions calculate the degree to which
  British society has geographically polarised over
  time by various social characteristics (status,
  occupation, tenure, etc)
• some of the first analysis using newly released
  2001 Census data
• ideological position that socio-spatial
  polarisation is bad as it implies segregation
  and, to some degree, discrimination
• Data Source all census data. access to all the
  digital censuses - 1971, 81, 91, 2001. key
  statistic tables for 2001 Census
• scale is limited to local authorities (406).
  quite crude, especially compared to our ward
  analysis (only 33 LA in London, versus 650 wards)
• important methodological developments in linking
  through time

50
A divided nation? Dorling et al. 2003

• Analytical methods develop their own metric of
  segregation which they call index of
  dissimilarity, the proportion of people or
  households who would have to move home, across
  local authority boundaries, were that group to be
  equally represented everywhere.
• empirically driven, largely descriptive narrative
  with data tables
• Results runs through large range of social
  characteristics commenting on the trends in
  polarisation
• white flight, all we can say here is that
  between local authorities the segregation between
  White and all others is growing. All other ethnic
  groups have become more geographically mixed
• tenure, the only tenure to have polarised
  geographically is council housing.
• car ownership. Only no car HHs increased,
  ..again there are signs that another group of
  households in Britain, are almost literally,
  being left behind.
• unfortunately, there is no evidence on income
  polarisation as this is missing from the census

51
Conclusion of study

• it does turn out that, in the end, Margaret
  Thatcher retains her status as being the prime
  minister who presided over a country pulling
  itself apart faster than before or after. What
  matter most now is that it kept pulling itself
  apart almost as quickly in the decade that follow
  her.
• New Labour is not making much different
• underlying drivers of increased mobility for
  larger sections of the population. choosing to
  segregate
• heading towards an American model - the land of
  cars and social division
• What we have report here simply widens the
  social gaps between different people living in
  different places. Fundamentally the human
  geography of Britain may be little altered, just
  made yet more stark.