FEMALE AND MALE MIGRATION PATTERNS INTO THE URBAN SLUMS OF NAIROBI, 1996-2006: EVIDENCE OF FEMINISATION OF MIGRATION? Ligaya Batten PhD Student Centre for Population Studies London 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 NUHDSS Data N=112003 Birth History* N=17532 Migration History N=12634 Employment History^ 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 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 analysis Aim 1 Sequence Analysis Descriptive Analysis of Sequences Compare sub-groups Create typologies Logistic Regression Multinomial logistic regression Aim 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 stay Kaplan-Meier survival estimates 0.25.5.75 1 0 1 2 3 4 5 Duration 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
AIM 1: CREATING MIGRANT TYPOLOGIES
LB-LSHTM2 Num ber of Sequences Migration History Indexplot for Whole Sample 0 3000 6000 9000 Within DSA Nairobi Slum Nairobi Non-Slum Other Urban Rural Outside Kenya 12000 0 2 4 6 8 10 11 Years
Slide 23 LB-LSHTM2 insert graphs comparing migrant types insert economic related graphs as well for IUSSP Ligaya, 08/09/2009
Migration History Indexplot for Males in Korogocho Migration History Indexplot for Females in Korogocho N u m b e r o f S e q u e n c e s 0 1000 2000 Within DSA Nairobi Slum Nairobi Non-Slum Other Urban Rural Outside Kenya N u m b e r o f S e q u e n c e s 0 1000 2000 Within DSA Nairobi Slum Nairobi Non-Slum Other Urban Rural Outside Kenya 3000 0 2 4 6 8 10 11 Years 3000 0 2 4 6 8 10 11 Years
Migration History Indexplot for Males in Viwandani Migration History Indexplot for Females in Viwandani N u m b e r o f S e q u e n c e s 0 1000 2000 3000 Within DSA Nairobi Slum Nairobi Non-Slum Other Urban Rural Outside Kenya N u m b e r o f S e q u e n c e s 0 1000 2000 3000 Within DSA Nairobi Slum Nairobi Non-Slum Other Urban Rural Outside Kenya 4000 0 2 4 6 8 10 11 Years 4000 0 2 4 6 8 10 11 Years
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 Regression Independent 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 Long Term Migrants - Male Long Term Migrants - Female 0 Within DSA 0 Within DSA Nairobi Slum Nairobi Slum N um ber of Sequences 200 400 600 800 1000 Nairobi Non-Slum Other Urban Rural Outside Kenya N um ber of Sequences 200 400 600 Nairobi Non-Slum Other Urban Rural Outside Kenya 1200 800 1400 1000 0 1 2 3 4 5 6 7 8 9 10 11 Years 0 1 2 3 4 5 6 7 8 9 10 11 Years
Index plots comparing migration typologies: Recent migrants Recent Migrants - Male Recent Migrants - Female 0 Within DSA 0 Within DSA Nairobi Slum Nairobi Slum Nairobi Non-Slum Nairobi Non-Slum N u m ber of Sequences 250 500 750 Other Urban Rural Outside Kenya N u m ber of Sequences 250 500 750 Other Urban Rural Outside Kenya 1000 1000 0 1 2 3 4 5 6 7 8 9 10 11 Years 0 1 2 3 4 5 6 7 8 9 10 11 Years
Index plots comparing migration typologies: Serial migrants Serial Migrants - Male Serial Migrants - Female 0 Within DSA 0 Within DSA Nairobi Slum Nairobi Slum N u m ber of Sequences 100 200 300 400 500 Nairobi Non-Slum Other Urban Rural Outside Kenya N u m ber of Sequences 100 200 300 Nairobi Non-Slum Other Urban Rural Outside Kenya 600 400 700 0 1 2 3 4 5 6 7 8 9 10 11 Years 500 0 1 2 3 4 5 6 7 8 9 10 11 Years
Index plots comparing migration typologies: Circular migrants Circular Migrants - Male Circular Migrants - Female 0 Within DSA 0 Within DSA Nairobi Slum Nairobi Slum N u m b e r o f S e q u e n c e s 25 50 75 100 125 Nairobi Non-Slum Other Urban Rural Outside Kenya N u m b e r o f S e q u e n c e s 25 50 75 100 Nairobi Non-Slum Other Urban Rural Outside Kenya 150 175 0 1 2 3 4 5 6 7 8 9 10 11 Years 125 0 1 2 3 4 5 6 7 8 9 10 11 Years
Index plots comparing migration typologies: Rural (to slum) migrants Rural Migrants - Male Rural Migrants - Female 0 Within DSA 0 Within DSA Rural Rural 300 300 N u m ber of Sequences 600 900 1200 1500 N u m ber of Sequences 600 900 1200 1800 1500 0 1 2 3 4 5 6 7 8 9 10 11 Years 1800 0 1 2 3 4 5 6 7 8 9 10 11 Years
Index plots comparing migration typologies: Urban (to slum) migrants Urban Migrants - Male Urban Migrants - Female 0 Within DSA 0 Within DSA Nairobi Slum Nairobi Slum N um ber of Sequences 200 400 600 800 1000 Nairobi Non-Slum Other Urban Rural N um ber of Sequences 200 400 600 Nairobi Non-Slum Other Urban Rural 1200 800 1400 0 1 2 3 4 5 6 7 8 9 10 11 Years 1000 0 1 2 3 4 5 6 7 8 9 10 11 Years
Multinomial Logistic Regression Recent Migrant Serial Migrant Circular Migrant Independant Variables RRR RRR RRR Sex Male (ref.) Ref. Ref. Ref. Female + ns ns Study site Viwandani (ref.) Ref. Ref. Ref. Korogocho - --- ns Age group 15-19 --- --- -- 20-24 (ref.) Ref. Ref. Ref. 25-29 ns +++ +++ 30-34 ++ ns +++ 35-39 ns ns +++ 40-44 +++ ns ns 45-49 ns ns ns 50-54 ns ns ns 55-59 ns Ns ++ 60+ ns Ns ns Ethnic Group Kikuyu (ref.) Ref. Ref. Ref. Luhya +++ +++ ++ Luo ++ +++ + Kamba ns +++ ns Kisii ++ 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 ns Ever Married Status Never Married (ref.) Ref. Ref. Ref. Ever Married --- --- --- Socio-economic status (1-10) Poorest [1] (ref.) Ref. Ref. Ref. Less poor - - Ns Relationship to Household Head Household Head (ref.) Ref. Ref. Ref. Spouse +++ ns ns Child ++ ns +++ Other relative ++ ns ns Unrelated --- --- --- 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
Odds ratios comparing female migration compared to male migration, by cohort of migration Year Group Odds Ratio Confidence Interval 1996-99 0.85 [0.79 0.93] 2000-02 1.06 [0.97 1.15] 2003-05 1.21 [1.11 1.31]
Numbers of male and female autonomous migrants, and sex ratios, 1996-2005
Odds ratios for a one year increase, comparing autonomous and association migrants, by sex. Sex Form Odds Ratio [95% Conf. Interval] Male Autonomous 0.98 [0.97 0.99] Male Associational 1.14 [1.12 1.16] Female Autonomous 1.07 [1.04 1.09] Female Associational 1.10 [1.08 1.11]
Numbers of male and female economic migrants, and sex ratios, 1996-2005
Odds ratios for a one year increase, comparing economic and non- economic migrants, by sex. Sex Reason Odds Ratio [95% Conf. Interval] Male Non-economic 1.03 [1.01 1.05] Male Economic 1.04 [1.02 1.05] Female Non-economic 1.09 [1.07 1.10] Female Economic 1.07 [1.04 1.10]
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, socioeconomic 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) Data African Population and Health Research Center (APHRC) Funding Economic & Social Research Council (ESRC). Thank you for listening!