VALUES AND SUPPORT FOR IMMIGRATION: A CROSSCOUNTRY COMPARISON

1
VALUES AND SUPPORT FOR IMMIGRATION A
CROSS-COUNTRY COMPARISON

• Eldad Davidov (University of Mannheim and
  GESIS-ZA, University of Cologne)
• Bart Meuleman (CeSO, University of Leuven)
• Jaak Billiet (CeSO, University of Leuven)
• Peter Schmidt (Univeristy of Giesen)

2
Outline

• Theoretical background
• Research hypotheses
• Methods measurements
• Data
• Cross-cultural measurement equivalence
• Results

3
Theoretical background (1)

• Clear distinction between attitudes and human
  values
• Attitude an individuals disposition to react
  with a certain degree of favorableness or
  unfavorableness to an object, behavior, person,
  institution, or event or to any other
  discriminable aspect of the individuals world
  (Ajzen, 1993).
• E.g. attitudes towards a certain immigration
  policy
• There are as many possible attitudes as objects
  in the world

4
Theoretical background (2)

• Human values desirable transsituational goals,
  varying in importance, that serve as guiding
  principles in the life of a person or other
  social entity (Schwartz, 1994).
• Schwartz postulated a theory that describes 10
  basic types of human values that are
  distinguished by their motivational goals

5
Theoretical background (3)
6
Theoretical background (3)
7
Theoretical background (4)

• Human values can be seen as abstract principles
  that guide concrete attitudes
• Values will influence the attitude toward a
  certain object if this object has relevant
  consequences for the attainment of the
  motivational goals that are associated with these
  respective values.

8
Research hypotheses (1)

• How does a certain immigration policy relate to
  the motivational goals that are associated with
  certain human values?
• Immigration is an obstacle for conservation since
  immigration often brings along changing
  traditions and norms.
• Admitting immigrants offers possibilities to
  attain self-trancendence
• Far less consensus on the relation with openness
  to change and self-enhancement

9
Research hypotheses (2)

• This leads to the following research hypotheses
• Conservation values coincide with more negative
  attitudes towards a liberal immigration policy
• Self-transcendence values coincide with more
  positive attitudes towards a liberal immigration
  policy
• In the past, empirical evidence was found for
  these hypotheses (SagivSchwartz 1995 Adorno,
  Frenkel-Brunswick, Levinson, and Sanford, 1950
  Schwartz, 1994 ) .
• There is no theoretical consensus regarding the
  relation of the other value dimensions and
  attitudes to immigration policies.

10
Research hypotheses (2)

• Based on Schwartzs argument regarding the
  universality of values, and
• Since there was no other evidence in the
  literature for differences
• We do not have reasons to expect effect
  differences of the values across countries.

11
Methods measurements (1)

• Data ESS round 1 (2002-03)
• 19 countries (3 countries were omitted Israel,
  Luxemburg and Italy)

12
Methods measurements (2)

• Measurements of human values based on Portrait
  Values Questionnaire (Schwartz et al. 2001)
• 11 verbal portraits for conservaton and
  self-trancendance valuesSelf-transcendence
  Universalism (UN) and Benevolence (BE)
• Ipeqopt/un1 Important that people are treated
  equally and have equal opportunities
• Ipudrst/un2 Important to understand different
  people
• Impenv/un3 Important to care for nature and
  environment

13
Methods measurements (3)

• Benevolence (BE)
• Iphlppl/be1 Important to help people and care
  for others well-being
• Iplylfr/be2 Important to be loyal to friends and
  devote to close people
• Conservation-by tradition, conformity and
  security
• Tradition (TR)
• Ipmodst/tr1 Important to be humble and modest,
  not draw attention
• Imptrad/tr2 Important to follow traditions and
  customs

14
Methods measurements (4)

• Conformity (CO)
• Ipfrule/co1 Important to do what is told and
  follow rules
• Ipbhprp/co2 Important to behave properly
• Security (SEC)
• Impsafe/sec1 Important to live in secure and
  safe surroundings
• Ipstrgv/sec2 Important that government is strong
  and ensures safety
• How much is this person like you? Very much like
  me (1) Not like me at all (6)

15
Methods measurements (5)

• Attitudes towards immigration policy are
  operationalised by two scales
• Willingness to allow immigrants into the
  countryTo what extent do you think country
  should allow people of a different race to come
  and live here? (4-point scale many 1, some
  2, a few 3, none 4)immigrants from poorer
  countries in Europeimmigrants from richer
  countries outside Europe immigrants from poorer
  countries outside Europe

16
Methods measurements (6)

• Support for conditions for immigrationHow
  important you think each of these things should
  be in deciding whether someone born, brought up
  and living outside country should be able to
  come and live here?
• good educational qualifications
• work skills needed in country
• 11-point scale 0 extremely unimportant 10
  extremely important

17
Self-transcendence

Allow

-
No Conditions
Conservation
-
18
Methods measurements (7)

• Before cross-cultural comparisons can be made,
  measurement equivalence has to be guaranteed
• whether or not, under different conditions of
  observing and studying phenomena, measurement
  operations yield measures of the same attribute
  (Horn McArdle 1992)
• If it is guaranteed, then cross-cultural
  comparison is allowed

19
Measurement Invariance
Group A
Group B
dB11
Item a
dA11
Item a
lB111
lA111
fB11 k B1
fA11 k A1
?B1
?A1
lB21
lA21
dB22
dA22
Item b
Item b
lB31
lA31
Item c
Item c
dB33
dA33
fB21
fA21
dB44
dA44
Item d
Item d
lB421
lA421
?B2
?A2
lB52
lA52
Item e
Item e
dB55
dA55
lB62
lA62
fB22 k B2
fA22 k A2
dB66
dA66
Item f
Item f
20
Steps in testing for Measurement Invariance

• Configural Invariance
• Metric Invariance
• Scalar Invariance
• Invariance of Factor Variances
• Invariance of Factor Covariances
• Invariance of latent Means
• Invariance of Unique Variances

21
Steps in testing for Measurement Invariance

• Configural Invariance
• Metric Invariance
• Equal factor loadings
• Same scale units in both groups
• Presumption for the comparison of latent means
• Scalar Invariance
• Invariance of Factor Variances
• Invariance of Factor Covariances
• Invariance of latent Means
• Invariance of Unique Variances

22
Full vs. Partial Invariance

• Concept of partial invariance introduced by
  Byrne, Shavelson Muthén (1989)
• Procedure
• Constrain complete matrix
• Use modification indices to find non-invariant
  parameters and then relax the constraint
• Compare with the unrestricted model
• Steenkamp Baumgartner (1998) Two indicators
  with invariant loadings and intercepts are
  sufficient for mean comparisons
• One of them can be the marker one further
  invariant item

23
Methods measurements (8)

• To sum up to be able to compare regression
  coefficients, at least metric equivalence is
  needed.
• Metric equivalence can be tested by means of
  multigroup structural equation modeling.

24
Methods measurements (7)

• Metric equivalence equality of factor loadings
• with

25
Methods measurements (8)

• For some items, the equivalence constraints had
  to be relaxed.
• Yet, for the attitude scales and the two
  higher-order value scales, at least partial
  metric equivalence was established
• ? comparison is allowed! (cfr. Steenkamp
  Baumgartner 1998)

26
Results (1)

• Following multigroup structural model was tested
  (expected effects between brackets)

Conservation
ALLOW
(-)
(-)
()
Self-transcendence
NO CONDITIONS
()
VALUES
ATTITUDES
27
Results (2)

• In all countries, the effects of conservation and
  self-transcendence on ALLOW were found to be
  significant and in the expected direction.
• However, there are differences in strength across
  countries
• Conservation strongest in CZ, PL, SI, FR, GB,
  GR weakest in SE, DK, AT, CH
• Self-transcendence strongest in PL, GB, FI, CZ
  weakest in SE, PT, DK

28
Clusters

• A test of clusters of countries was conducted.
  For this, unstandardized regression coeficients
  were investigated by a sequence of chi-square
  difference tests.

29
Results (3)
Self-transcendence on ALLOW
Conservation on ALLOW
30
Results (3)

• Conservation has a negative effect on NO
  CONDITIONS in 17 countries.
• Self-transcendence has a positive significant
  effect in 16 countries, and a negative one in 1
  country (Greece)
• Here, even more cross-national differences in
  effect size.

31
Results (3)
Self-transcendence on NO CONDITIONS
Conservation on CONDITION
32

• In total, 69 out of 76 relations were according
  to our hypotheses.

33
The robustness of the findings in the presence of
control variables

• age, gender, level of education, household
  income, religiosity, attendance of religious
  services and left-right orientation.
• Education (Jackman and Muha, 1984 Coenders and
  Scheepers, 2003 Kunovich, 2004)
• Age, economic situation, political orientation
  and religiosity (see for example Billiet, 1995
  Fetzer, 2000 Semyonov, Raijman and Gorodzeisky,
  2006)

34

• Additionally, we controlled for four contextual
  variables, namely GDP per capita, GDP growth, the
  inflow of immigrants and the stock of foreign
  born population.

35

• Substantive results in a multi-level analysis did
  not change.
• the values conservation and self-transcendence
  remained the strongest predictors for attitudes
  toward immigration.
• These explanations seem to be not exclusive but
  rather complementary.

36
Explaining effect size differences across nations

• Effects were robust but there was some variation
  across countries.
• Giving a sound explanation is difficult because
  of the relatively small number of units
  (countries).
• Probably, the differences in effect sizes are the
  result of complex processes, in which certain
  characteristics of the population, historical
  path dependence and national characteristics
  interact. The fact that various of these factors
  are extremely hard to measure makes it harder to
  model them.

37

• An attempt two national context variables,
  namely GDP per capita and the gross immigration
  inflow (per 1000 inhabitants).

38
GDP

• Allow
• The effects of self-transcendence and
  conservation on allow appear to be more
  pronounced in countries with a low GDP per
  capita.
• human values give a stronger prediction of the
  willingness to allow immigrants for poorer
  countries than for richer ones.
• No conditions
• For the effects on no conditions, the opposite
  pattern is found human values tend to provide a
  stronger explanation in richer countries.

39
Possible explanations

• It may be the case that in poorer countries, the
  immigration debate is focused on the quantity of
  immigration, as inhabitants of these countries
  may fear that immigration will constitute a
  competition with their already limited economic
  resources.
• Values may be related to such fears
  (universalistic individuals may fear less than
  conservative individuals) and as a consequence be
  better predictors for allow in poorer
  countries.

40

• The question whether migration should be
  conditioned on certain criteria on the other
  hand, may occupy a more central role in the
  debate in wealthier countries.
• Further research is needed to verify these
  explanations

41
Inflow of immigrants

• the inflow of immigrants is connected to the
  effect sizes for only one of the concepts, namely
  allow.
• In countries with a high level of immigration,
  human values tend to affect allow only to a
  lesser extent.

42
Possible explanation

• This may be due to the fact that inhabitants of
  these immigration countries have a larger amount
  of information obtained via personal
  experiences or through the mass media - at their
  disposal in the attitude formation process.
  Therefore, these persons might fall back on
  general human values only to a lesser extent than
  inhabitants without this information.
• Even in these countries values have a substantial
  effect.

43
Conclusion (1)

• The effects of human values on attitudes towards
  immigration policy is quite robust (69 out of 76
  hypotheses confirmed)
• These effects remain significant when we control
  for age, gender, education and religiosity ?
  additional explanatory power
• Human values deserve more attention than they
  have received so far

44

• Thank you very much for your attention!!!!

45
(No Transcript)
46
Table 6. Correlations between effect sizes and
national context variables