FEMALE AND MALE MIGRATION PATTERNS INTO THE URBAN SLUMS OF NAIROBI, 1996 - 2006: EVIDENCE OF FEMINISATION OF MIGRATION? Ligaya Batten PhD Student Centre.

FEMALE AND MALE MIGRATION PATTERNS INTO THE URBAN SLUMS OF NAIROBI, 1996 - 2006: EVIDENCE OF FEMINISATION OF MIGRATION?

Ligaya BattenPhD StudentCentre for Population StudiesLondon School of Hygiene and Tropical Medicine

GENERAL BACKGROUND• Population growth and urbanisation in sub-

Saharan Africa• Mainly due to Rural to Urban Migration and

Natural Increase• Negative outcomes related to urbanisation in

SSA:– Population pressure on services in ill-equipped cities (such

as housing, health and education) and economic opportunities often leads to:• Slum formation – poor quality housing, lack of sanitation,

lack of access to clean water and health services.• Unemployment and growth in the informal labour market

– poverty, precarious livelihoods

GENERAL BACKGROUND• Phenomenon of female autonomous migration

emerging from previously male dominated process• Evidence of autonomous female migration in South-

East Asia and Latin America, West Africa, South Africa

• Causes of feminisation of migration– Household poverty, fragile ecosystems– Less marriage, better female education– Increase in family and refugee migration

• Consequences of feminisation of migration– Change of gender roles in the family and labour market– Potential knock on effect of reducing fertility

• But no evidence on trends, causes and consequences of sex composition of migration in African slums yet

STUDY SETTING• High Rural-Urban

migration (esp. Nairobi)

• Over half urban population living in slums

• Rel. high education• Informal Sector• Poverty

STUDY SETTING (cont.)

Source: APHRC 2002

STUDY SITE APHRC (African

Population and Health Research Centre)

Two urban slums – Viwandani and Korogocho

Population ≈60,000 Area ≈ 1km2 Employment Fertility Highly mobile

population

DATA• Nairobi Urban Health Demographic Surveillance

Site (NUHDSS)– Who?

• No sampling – ALL residents– When?

• Initial Census in August 2002• Every 4 month• I will use data from 01 January 2003 – 31

December 2007– What is collected in the main DSS?

• Demographic data (births, deaths, in and out migration)

• Socio-Economic data (marriage, education, employment, assets)

• Health Data (morbidity, vaccinations, verbal autopsy)

DATA• Nairobi Urban Health Demographic Surveillance Site

(NUHDSS)• Nested surveys:

– Migration history• Who?

– >= 12 years old– 14000 sampled 11487 responses

• When?– September 2006 - April 2007

• What is collected?– 11 year migration history calendar (every month)– Detailed cross-sectional questionnaire

– Birth histories and marital histories collected periodically

Timeline of Available Data  1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

NUHDSS

Data

N=112003                        

Birth History*

N=17532                        

Migration

History

N=12634                        

Employment

History^

N=12634                        

*Birth histories collected retrospectively as part of the main NUHDSS  

^ Time period covered (in retrospect)    Year during which data collection occurred  Time period covered in retrospect  

Aims

1. Define migrant typologies and assess differences between female and male migrant types.

2. Assess whether or not there has been a trend of feminisation of migration between 1996 and 2006.

METHODS• Basic descriptive analysisAim 1• Sequence Analysis

– Descriptive Analysis of Sequences– Compare sub-groups– Create typologies

• Logistic Regression• Multinomial logistic regressionAim 2• Mantel-Haenzel test for trend

– sex ratio of migrants over time– sex ratio of autonomous migrants over time– sex ratio of economic migrants over time

Definition of Variables• Outcomes:

– Migrant (Long term, recent, serial, circular)

– Autonomous/Associational– Economic/Non-economic

• Explanatory variables:– Sex– Study site, age, education level,

ethnicity, marital status, socio-economic status, relationship to household head

RESULTS

i. Descriptive Results

ii. Migrant typologies

iii.Feminization of migration?

Descriptive Results

Age and Gender Structure of Viwandani & Korogocho in Dec 2006, by in-migrant

status

Viwandani Korogocho

Proportions of in-migrants

Origin of In-Migrants

Form (In-Migrants)

Motivations for In-Migration

Duration of stay0

.25

.5.7

51

0 1 2 3 4 5Duration of stay in the DSA (Years)

95% CI

95% CI

95% CI

95% CI

slumid = VIWANDANI/ sex = Male

slumid = VIWANDANI/ sex = Female

slumid = KOROGOCHO/ sex = Male

slumid = KOROGOCHO/ sex = Female

Kaplan-Meier survival estimates

Aim 1:Creating Migrant

Typologies

0

3000

6000

9000

12000

Numb

er of

Seq

uenc

es

0 2 4 6 8 10 11Years

Within DSA

Nairobi Slum

Nairobi Non-Slum

Other Urban

Rural

Outside Kenya

Migration History Indexplot for Whole Sample

0

1000

2000

3000

Numb

er of S

equenc

es

0 2 4 6 8 10 11Years

Within DSA

Nairobi Slum

Nairobi Non-Slum

Other Urban

Rural

Outside Kenya

Migration History Indexplot for Males in Korogocho0

1000

2000

3000

Numb

er of S

equenc

es

0 2 4 6 8 10 11Years

Within DSA

Nairobi Slum

Nairobi Non-Slum

Other Urban

Rural

Outside Kenya

Migration History Indexplot for Females in Korogocho

0

1000

2000

3000

4000

Numb

er of

Sequen

ces

0 2 4 6 8 10 11Years

Within DSA

Nairobi Slum

Nairobi Non-Slum

Other Urban

Rural

Outside Kenya

Migration History Indexplot for Males in Viwandani0

1000

2000

3000

4000

Numb

er of

Sequen

ces

0 2 4 6 8 10 11Years

Within DSA

Nairobi Slum

Nairobi Non-Slum

Other Urban

Rural

Outside Kenya

Migration History Indexplot for Females in Viwandani

Descriptive Analysis of Sequences

Sex Both Sites Korogocho Viwandani

Mean length of stay (months) [Freq]

Male 97.35 [6561] 111.09 [2703] 87.72 [3858]

Female 93.14 [4926] 108.14 [2420] 78.67 [2506]

Total 95.55 [11487] 109.70 [5123] 84.15 [6364]

Mean number of places lived [Freq]

Male 1.63 [6561] 1.37 [2703] 1.82 [3858]

Female 1.65 [4926] 1.40 [2420] 1.90 [2506]

Total 1.64 [11487] 1.38 [5123] 1.85 [6364]

Mean number of residence episodes [Freq]

Male 1.67 [6561] 1.39 [2703] 1.86 [3858]

Female 1.69 [4926] 1.43 [2420] 1.95 [2506]

Total 1.68 [11487] 1.41 [5123] 1.90 [6364]

Logistic RegressionIndependent Variables Odds Ratio (95% Conf. - Interval)

Sex

Male (ref.) 1.00 -

Female 1.41** (1.27 – 1.58)

Study site

Viwandani (ref.) 1.00 -

Korogocho 0.28** (0.25 – 0.31)

Age group (at time of migration for migrants, 1996 for non-migrants)

0-4 0.01** (0.01 – 0.02)

5-9 0.06** (0.05 – 0.07)

10-14 0.17** (0.14 – 0.21)

15-19 0.77* (0.66 – 0.91)

20-24 (ref.) 1.00 -

25-29 0.56** (0.47 – 0.67)

30-34 0.32** (0.27 – 0.40)

35-39 0.19** (0.15 – 0.25)

40-44 0.19** (0.14 – 0.26)

45-49 0.17** (0.11 – 0.26)

50-54 0.16** (0.10 – 0.27)

55-59 0.19** (0.09 – 0.38)

60+ 0.14** (0.07 – 0.28)

Highest education level reached

No education (ref.) 1.00 -

Primary 2.62** (1.94 – 3.54)

Secondary 2.32** (1.70 – 3.16)

Higher 3.32** (1.70 – 6.48)

** p<0.001 * p=0.002

Index plots comparing migration typologies: Long term migrants

0

200

400

600

800

1000

1200

1400

Numb

er of

Seq

uenc

es

0 1 2 3 4 5 6 7 8 9 10 11Years

Within DSA

Nairobi Slum

Nairobi Non-Slum

Other Urban

Rural

Outside Kenya

Long Term Migrants - Male0

200

400

600

800

1000

Numb

er of

Seq

uenc

es

0 1 2 3 4 5 6 7 8 9 10 11Years

Within DSA

Nairobi Slum

Nairobi Non-Slum

Other Urban

Rural

Outside Kenya

Long Term Migrants - Female

Index plots comparing migration typologies: Recent migrants

0

250

500

750

1000

Numb

er of

Sequ

ence

s

0 1 2 3 4 5 6 7 8 9 10 11Years

Within DSA

Nairobi Slum

Nairobi Non-Slum

Other Urban

Rural

Outside Kenya

Recent Migrants - Male0

250

500

750

1000

Numb

er of

Sequ

ence

s

0 1 2 3 4 5 6 7 8 9 10 11Years

Within DSA

Nairobi Slum

Nairobi Non-Slum

Other Urban

Rural

Outside Kenya

Recent Migrants - Female

Index plots comparing migration typologies: Serial migrants

0

100

200

300

400

500

600

700

Numb

er of

Sequ

ence

s

0 1 2 3 4 5 6 7 8 9 10 11Years

Within DSA

Nairobi Slum

Nairobi Non-Slum

Other Urban

Rural

Outside Kenya

Serial Migrants - Male0

100

200

300

400

500

Numb

er of

Sequ

ence

s

0 1 2 3 4 5 6 7 8 9 10 11Years

Within DSA

Nairobi Slum

Nairobi Non-Slum

Other Urban

Rural

Outside Kenya

Serial Migrants - Female

Index plots comparing migration typologies: Circular migrants

0

25

50

75

100

125

150

175

Numb

er of

Sequ

ence

s

0 1 2 3 4 5 6 7 8 9 10 11Years

Within DSA

Nairobi Slum

Nairobi Non-Slum

Other Urban

Rural

Outside Kenya

Circular Migrants - Male0

25

50

75

100

125

Numb

er of

Sequ

ence

s

0 1 2 3 4 5 6 7 8 9 10 11Years

Within DSA

Nairobi Slum

Nairobi Non-Slum

Other Urban

Rural

Outside Kenya

Circular Migrants - Female

Index plots comparing migration typologies: Rural (to slum) migrants

0

300

600

900

1200

1500

1800

Numb

er of

Sequ

ence

s

0 1 2 3 4 5 6 7 8 9 10 11Years

Within DSA

Rural

Rural Migrants - Male0

300

600

900

1200

1500

1800

Numb

er of

Sequ

ence

s

0 1 2 3 4 5 6 7 8 9 10 11Years

Within DSA

Rural

Rural Migrants - Female

Index plots comparing migration typologies: Urban (to slum) migrants

0

200

400

600

800

1000

1200

1400

Numb

er of

Seq

uenc

es

0 1 2 3 4 5 6 7 8 9 10 11Years

Within DSA

Nairobi Slum

Nairobi Non-Slum

Other Urban

Rural

Urban Migrants - Male0

200

400

600

800

1000

Numb

er of

Seq

uenc

es

0 1 2 3 4 5 6 7 8 9 10 11Years

Within DSA

Nairobi Slum

Nairobi Non-Slum

Other Urban

Rural

Urban Migrants - Female

Multinomial Logistic Regression

  Recent Migrant Serial Migrant Circular Migrant Independant Variables RRR RRR RRR Sex      Male (ref.) Ref. Ref. Ref.Female + ns nsStudy site      Viwandani (ref.) Ref. Ref. Ref.Korogocho - --- nsAge group      15-19 --- --- --20-24 (ref.) Ref. Ref. Ref.25-29 ns +++ +++30-34 ++ ns +++35-39 ns ns +++40-44 +++ ns ns45-49 ns ns ns50-54 ns ns ns55-59 ns Ns ++60+ ns Ns nsEthnic Group      Kikuyu (ref.) Ref. Ref. Ref.Luhya +++ +++ ++Luo ++ +++ +Kamba ns +++ nsKisii ++ ns ++Other ns ns ns

Multinomial Logistic Regression (cont.)

  Recent Migrant Serial Migrant Circular Migrant Independant Variables RRR RRR RRR Highest education level reached      No education (ref.) Ref. Ref. Ref.Higher education level - ns nsEver Married Status      Never Married (ref.) Ref. Ref. Ref.Ever Married --- --- ---Socio-economic status (1-10)      Poorest [1] (ref.) Ref. Ref. Ref.Less poor - - NsRelationship to Household Head      Household Head (ref.) Ref. Ref. Ref.Spouse +++ ns nsChild ++ ns +++Other relative ++ ns nsUnrelated --- --- ---Economic reason for moving to the DSA?      No (ref.) Ref. Ref. Ref.Yes ns --- ---Associational migrant?      No (ref.) Ref. Ref. Ref.Yes +++ +++ +++

Aim 2:Is there a trend of

feminization of migration?

Numbers of male and female migrants, and sex ratios, 1996-

2005

Numbers of male and female autonomous migrants, and sex

ratios, 1996-2005

Numbers of male and female economic migrants, and sex ratios,

1996-2005

Conclusions and discussion

Conclusions (i)• Female migrants more mobile than

male• Strong differences between study sites• Migrant types:

• Females – recent migrants• Korogocho – serial migrants• Economic migrants – serial and circular

migrants• Associational migrants – recent, serial and

circular migrants

Conclusions (ii)• Trend of feminisation of migration

found:• Decrease in the sex ratio of migration

into the study site from 1996 - 2006• Decrease in the sex ratio of autonomous

migration into the study site from 1996 - 2006

• Decrease in the sex ratio of economic migration into the study site from 1996 - 2006

Limitations• Under-sampling of migrants in the

migration history survey• Recall bias• Time varying data lacking for certain

important characteristics• E.g. Marital status, education level, socio-

economic status

• Definition of economic and autonomous migration open to interpretation

Implications

• Feminisation of migration may have both social and demographic consequences:• Change in women’s roles, increase in

women’s empowerment• May lead to a number of positive

consequences – gender equality in the labour market, improvements in child health and education

• Urban “modernised” lifestyles - potential for fertility decline and therefore reduction in future population growth

Planned Future Work• Use cluster analysis to group sequences

according to characteristics other than the place of origin, such as motivation, ethnicity, education level, and perhaps other demographic characteristics

• Use migration typologies as explanatory variables for exploring the following:• Employment

• Identify which migrant types have the best chances of employment in the study site, by sex (controlling for employment status in the place of origin).

• Establish the extent to which unemployment increases the likelihood of out-migration from the study site.

• Fertility• Describe the trends in family building patterns of

migrants on non-migrants over the last eleven years.

Acknowledgements• Supervisor Angela Baschieri

(LSHTM)• Advisors Eliya Zulu (APHRC)

Jane Falkingham (Soton)John Cleland (LSHTM)

• DataAfrican Population and Health Research Center (APHRC)

• Funding Economic & Social Research Council (ESRC).

• Thank you for listening!