A Dynamic and Multidimensional Model of Integration

1

• A Dynamic and Multidimensional Model of
  IntegrationBuilding on Longitudinal Quantitative
  and Qualitative Data on Migrants in Vienna
• Rossalina Latcheva Institute for Advanced
  Studies (IHS)
• Vienna
• with
• Barbara Herzog-Punzenberger
• Austrian Academy of Science
• International Conference on Migration and
  Life-Course Research
• 5/6 December, 2008, Conference Center Teerhof,
  University of Bremen

Institute for Advanced Studies Stumpergasse
56 A-1060 Wien Tel [phone]
www.ihs.ac.at www.equi.at
2
Point of departure (1) Theoretical challenges

• Migration and integration a battlefield of
  political interests ? blurring boundaries between
  categories of analysis and categories of practice
  
• Migrants, ethno-national groups and receiving
  societies treated as homogenous entities
• Integration as a
• contested
• one-dimensional
• dichotomous
• static
• unidirectional linear
• decontextualized
• non-relational and non-interactive concept, often
  without explicit longitudinal or life-course
  perspective

3
Point of departure (2) Methodological challenges

• For decades (at least for Austria)
• almost no data on naturalized persons
• almost no longitudinal data on migrants and their
  descendants
• small samples
• a missing life-course perspective within
  migration and integration research
• no systematic integration of quantitative and
  qualitative approaches in research design and
  analysis

4
Implications for research and policy-making (1)

• primarily descriptive, non-longitudinal and
  one-sided research
• due to small sample sizes impossibility to model
  complexity and heterogeneity and control for
  intervening variables such as gender, age,
  generation, year of arrival, country of origin,
  (year of) naturalization etc. at the same time
• causal inferences derived mainly from theory and
  rarely based on empirical evidence
• which
• often leads to invalid and not sufficiently
  differentiated comparisons between and within
  social categories
• imposes homogeneity on both, dependent and
  independent variables
• leads to uncertain and spurious causal
  inferences, gtgt that often legitimise political
  interventions

5
Implications for research and policy-making (2)

• The recognition that individual and social change
  are mutually dependent generates the credo that
  the relationship between individuals and their
  settings is bi-directional (Colby 1998 viii) and
  that this should be explicitly approached in
  theory, study design and techniques of analysis
  and of course in designing policy measures
• Studying the integration of migrants means
  conceptualising the dynamic links between human
  agency and structure, between events in one life
  domain and changes in another, between past
  events and current statuses and thus adopting
  study design and research methods which take all
  this into account (Giele and Elder 1998).

6
An attempt to meet the challenges integrating
research approaches and data gathered within the
LIMITS and SiM projects

• Combining quantitative and qualitative approaches
  using method triangulation with connected samples
• Integrating objective and subjective perspective
  in the assessment of migration and integration
  processes analysing the intersection between
  migrants own perceptions and interpretations on
  the one side and the objective traces of their
  biographies on the other
• Conducting a variety of multivariate and content
  analysis (event history, optimal matching,
  multidimensional scaling, cluster analysis,
  content analysis across cases) and a systematic
  triangulation of findings
• Developing a dynamic multidimensional model for
  explaining the interdependency of time and
  context within the individual migration
  biography, which allows for flexibility by
  grasping complexity

7
The LIMITS-projectImmigrants and Ethnic
Minorities in European Cities Life Courses and
Quality of Life in a World of Limitations

• LIMITS (2003-2006) funded by the EC within the
  FP5
• Major objectives
• collection and analysis of longitudinal data on
  life courses of migrants in six European cities
• to catch the dynamics of migration and
  integration processes in different national
  contexts
• 6 cities in 5 countries
• Vienna (Austria),
• Bielefeld (Germany),
• Amsterdam Rotterdam (Netherlands),
• Lisbon (Portugal),
• Stockholm (Sweden)
• 2 groups in each city 300 interviews per group
  ( 600 respondents per city)
• database of approximately 3,300 cases

8
The LIMITS-sampleCriteria for inclusion in the
sample
at least 15 years of stay in the receiving
country, first generation migrants (proxy at
least 35 years of age), heterogeneous criteria of
group definition, but independent of current
nationality
9
Data

• Event histories (using calendar tool) of
• migration,
• education and employment,
• family formation,
• housing,
• social networks
• language acquisition (attained courses and
  self-assessment)
• Cross-sectional information on
• income, pre-migration and citizenship acquisition

10
SiM-project

• SiM (2006) project funded by the Austrian
  Ministry for Science and
• Research within the framework of the NODE
  research program
• Build on the LIMITS-Data for Austria (Vienna)
  quantitative information
• gathered on long-term residents in Vienna from
  Turkey and Serbia
• (n601)
• Qualitative information from 30 in-depth
  interviews -
• sampled on the basis of the produced trajectory
  clusters from
• the optimal matching analysis
• Research objectives
• to discover similarities in migrants
  trajectories through re-analysis of the
  LIMITS-data using optimal matching
• to assess the individual migration project from a
  subjective perspective
• to develop a dynamic and multidimensional model
  of integration

11
Methods of Analysis

• Optimal matching
• analysis of labour market participation over time
  (analysing the differences (distances) between
  pairs of individual trajectories)
• All possible pairs to be compared (601
  (601-1) / 2) 180.300 distances.
• Multidimensional scaling cluster analysis
• obtaining 5 different types (clusters) of labour
  market participation over time (1. fully
  employed, 2.difficult beginning then full
  employment, 3. discontinuous employment
  biography, 4.transition from education to full
  employment, 5. outside the labour market)
• Selection criteria for in-depth interviews
• cluster, ethnicity, gender, proficiency in German

12
Cluster 1 Fully employed all along the migration trajectory Cluster 2 Difficulties at the beginning Cluster 3 Discontinuous labour market trajectory Cluster 4 Transition from education to stably employment career Cluster 5 Outside the labour market Total
n 214 230 65 64 28 601
Percent 35,6 38,3 10,8 10,6 4,7 100
Gender (in ) Gender (in ) Gender (in ) Gender (in ) Gender (in ) Gender (in ) Gender (in )
male 77,6 58,3 33,8 35,9 0,0 57,4
female 22,4 41,7 66,2 64,1 100,0 42,6
Ethnicity (in ) Ethnicity (in ) Ethnicity (in ) Ethnicity (in ) Ethnicity (in ) Ethnicity (in ) Ethnicity (in )
Turkey 33,2 54,3 52,3 75,0 82,1 50,1
Serbia 66,8 45,7 47,7 25,0 17,9 49,9
Education (in ) Education (in ) Education (in ) Education (in ) Education (in ) Education (in ) Education (in )
none 3,4 6,8 3,4 6,3 4,0 5,1
elementary 46,6 45,5 50,8 37,5 64,0 46,3
secondary 36,3 38,6 32,2 31,3 12,0 35,1
university entrance certificate 13,7 9,1 13,6 25,0 20,0 13,5
Age (in ) Age (in ) Age (in ) Age (in ) Age (in ) Age (in ) Age (in )
35 - 44 22,9 43,9 73,8 70,3 57,1 43,1
45 - 54 35,5 38,3 24,6 25,0 32,1 34,1
55 - 64 31,3 13,0 1,5 1,6 10,7 17,0
65 and older 10,3 4,8 0,0 3,1 0,0 5,8
13
(No Transcript)
14
Discussion

• The dynamic of the model results from
• the interdependency of the different dimensions
  (within and between phases)
• the different composition of the relevant
  dimensions within phases and life cycles
• Next steps validation of the model
• other groups of origin,
• other types of migrants (refugees, transnational
  migrants,)
• other national contexts

15
Does the model meet the challenges?

• homogeneity ?
• unidimensional ?
• dichotomous ?
• static ?
• decontextualized ?

heterogeneity between and within groups
depending on time of arrival, gender,
generation multidimensional with at least 4
dimensions grades and shades dynamic
interdependency of phases and dimensions
(longitudinal data) framed by changing
conditions of the receiving society
The model allows to grasp complexity, it is
derived by a systematic integration of
quantitative and qualitative approaches in study
design and analysis
16
Thank you for your attention and critical remarks!

• Rossalina Latcheva
• Institute for Advanced Studies
• latcheva_at_ihs.ac.at
• www.equi.at
• Stumpergasse 56
• A-1060 Vienna
• Tel [phone]
• Fax [phone]