Implications of Ethnic Identity Exploration and Ethnic Identity Affirmation and Belonging for Intergroup Attitudes Among Adolescents
Kevin A. Whitehead, Andrew T. Ainsworth, Michele A. Wittig* and Brandy Gadino
California State University, Northridge
Ancillary Materials
Results of the Factor Analysis of the Two Scales of the Multigroup Ethnic Identity Measure
Multi-group Factor Analysis of Exploration and Affirmation and Belonging
Confirmatory factor
analyses of the ethnic identity exploration items (at time 1) and the ethnic identity affirmation and belonging items (at time 2) were performed on each of the three ethnic groups
separately. These constructs were tested separately (prior to testing the
hypothesized and alternative models) in order to establish that the factor
structure obtained from previous exploratory methods (e.g. Phinney et al.,
1997) not only fit the data for each of the three sub-samples but that the
three groups were relatively equivalent in the structure of these two
constructs. These CFA's use assessments that are keyed to the relationships
specified in our hypothesized model. The factor structure consisted of the five
exploration items loading together on an exploration factor, while the four affirmation and belonging items loaded
together on a separate affirmation and
belonging factor. For all three groups the factors were allowed to correlate.
Each of the three groups required a minimum number of correlated errors to be
added to the model in order for the model fit to reach acceptable levels. After
model fit was reached for each individual group, the groups were than compared
to each other in three complete invariance multi-group models. Complete
invariance means that the groups are compared on every aspect of the factor
structures, loadings and variances (e.g. error variances, factor variances).
For simplicity, the three groups were compared pair-wise, and in each paired
multi-sample analysis the loadings and variances were constrained to be equal
between the two groups. When using this strategy, the two groups can be
considered equal if, after the models are constrained to be equal, the multi-group
model still fits. This procedure was followed three times, in order to compare
each group to each of the other groups. The results of this procedure are
presented in the sections that follow.
Confirmatory factor analysis in the
Euro-American sub-sample. The structure of the two ethnic identity factors,
exploration and affirmation and
belonging, fit well for the Euro-American sub-sample, with c2(26) = 35.09, p = .110, CFI = .982 and RMSEA = .051. Even though the hypothesized
factor model fit the data the Lagrange multiplier test indicated two large
correlated errors. The paths appear superfluous in this factor but serve to
dramatically improve the fit of the full hypothesized model (shown in Figure 4
below). The first correlated error was between two of the exploration items
(items 2 and 3) and the addition of this path significantly improved the
chi-square, c2(1) =
9.881, p < .05. The addition of a
correlation between errors associated with the second and third items of the affirmation and belonging subscale was
also a significant improvement, c2(1)
= 6.713, p < .05. The final model
fit was nearly perfect for the Euro-American sub-sample, with c2(24) = 19.108, p = .746, CFI = 1.000 and RMSEA < .001.
For the Euro-American sub-sample, all of the loadings were significant for both factors and the correlation between the two factors was significant and moderately large (r = .352). The exploration factor accounted for a range of 12% to 69% of the variability in the exploration items and the affirmation and belonging factor accounted for a range of 68% to 72% of the variability in the affirmation and belonging items.
Confirmatory factor analysis in the
Asian American sub-sample. As hypothesized, the exploration and affirmation and belonging items only marginally
fit the data for the Asian American sub-sample, with c2(26)
= 75.138, p < .001, CFI = .908 and
RMSEA = .115. The Lagrange multiplier test suggested the inclusion of a
correlated error between items one and three of the affirmation and belonging subscale and this was a significant
improvement, c2(1) =
13.625, p < .05. A correlation
between the error terms for items one and five of the exploration subscale was
also included, c2(1) =
9.560, p < .05. In addition, error
terms for items three and five of the exploration subscale were also
correlated, c2(1) =
9.286, p < .05. Finally, the
addition of a correlation between the error terms associated with items two and
four of the affirmation and belonging
subscale significantly improved the model, c2(1)
= 8.479, p < .05. The final model
for the Asian American sub-sample fit well, c2(22)
= 34.188, p = .047, CFI = .977 and
RMSEA = .062.
For the Asian American sub-sample, all of the loadings were significant for both factors and the correlation between the two factors was significant and large (r = .698). The exploration factor accounted for a range of 15% to 56% of the variability in the exploration items and the affirmation and belonging factor accounted for a range of 55% to 71% of the variability in the affirmation and belonging items.
Confirmatory factor analysis in the
Latino sub-sample. For the Latino sub-sample, the hypothesized model fit
the data moderately well, with c2(26)
= 39.852, p = .040, CFI = .925 and
RMSEA = .071. The addition of a correlated error between the second and third
items on the exploration subscale significantly improved the model, c2(1) = 8.196, p < .05, and the resultant model fit well, with c2(25) = 31.656, p = .168, CFI = .964 and RMSEA = .050.
For the Latino American sub-sample, all of the loadings were significant for the affirmation and belonging items, however two of the loadings on the exploration factor were not significant, namely the second and third items on the exploration subscale. The correlation between the two factors was moderate (r = .367) and significant. The exploration factor accounted for a range of 4% to 40% of the variability in the exploration items and the affirmation and belonging factor accounted for a range of 32% to 66% of the variability in the affirmation and belonging items.
Complete invariance multi-group confirmatory factor analysis: Euro-Americans and Asian Americans. When the Euro-American and Asian American sub-samples were constrained in a complete invariance multi-group analysis, the model fit the data for the groups moderately well, with c2(65) = 113.027, p < .001, CFI = .955 and RMSEA = .051. The Lagrange multiplier test suggested that the two groups should be allowed to differ in terms of the correlation between the two factors and releasing this constraint significantly improved the model fit, c2(1) = 13.105, p < .05. After accounting for this difference between the groups, the multi-group model fit the data well, with c2(64) = 99.922, p = .002, CFI = .966 and RMSEA = .045.
Complete invariance multi-group confirmatory factor analysis: Asian
Americans and Latinos. The multi-group comparison of the Asian American and
Latino sub-samples showed that the two groups were very similar and the model
fit was very good, with c2(66)
= 95.511, p = .010, CFI = .959 and
RMSEA = .042. This degree of model fit shows that, even though the factor
loadings of two of the ethnic identity exploration items in the Latino
sub-sample were not significant, those loadings were not significantly
different from the loadings found in the Asian American sample.
Complete invariance multi-group confirmatory factor analysis: Latinos and Euro-Americans. When the Latino and Euro-American sub-samples were compared in a complete invariance multi-group analysis, there was an additional similarity between the groups, namely the correlation between the error terms of the second and third items of the exploration factor was significant for both groups. Since the two groups shared this extra parameter, it was also constrained between the groups. The resultant multi-group model fit the groups fairly well, with c2(69) = 106.101, p = .003, CFI = .947 and RMSEA = .047. The Lagrange multiplier test suggested that the two groups should be allowed to differ in terms of the error term associated with the second affirmation and belonging item, and releasing this constraint significantly improved the model fit, c2(1) = 21.177, p < .05, After accounting for this difference between the groups, the multi-group model fit the data well, c2(68) = 84.924, p = .080, CFI = .976 and RMSEA = .032. As in the case of the multi-group model for Asian Americans and Latinos, this shows that even though the factor loadings of two of the ethnic identity exploration items in the Latino sub-sample were not significant, these loadings were not significantly different from the loadings of the same items in the Euro-American sub-sample.
Correlation
Tables for Each Ethnic Group
|
|
|
Exploration |
Affirmation &
Belonging |
Warmth |
|||||||||
|
|
|
Item 1 |
Item 2 |
Item 3 |
Item 4 |
Item 5 |
Item 1 |
Item 2 |
Item 3 |
Item 4 |
Asian |
Latinos |
Euro |
|
Exploration |
Item 1 |
1.000 |
|
|
|
|
|
|
|
|
|
|
|
|
Item 2 |
0.252 |
1.000 |
|
|
|
|
|
|
|
|
|
|
|
|
Item 3 |
0.201 |
0.352 |
1.000 |
|
|
|
|
|
|
|
|
|
|
|
Item 4 |
0.543 |
0.251 |
0.269 |
1.000 |
|
|
|
|
|
|
|
|
|
|
Item 5 |
0.350 |
0.294 |
0.198 |
0.518 |
1.000 |
|
|
|
|
|
|
|
|
|
Affirmation & Belonging |
Item 1 |
0.224 |
0.069 |
0.272 |
0.244 |
0.155 |
1.000 |
|
|
|
|
|
|
|
Item 2 |
0.200 |
0.074 |
0.201 |
0.249 |
0.247 |
0.771 |
1.000 |
|
|
|
|
|
|
|
Item 3 |
0.185 |
0.050 |
0.190 |
0.222 |
0.225 |
0.727 |
0.651 |
1.000 |
|
|
|
|
|
|
Item 4 |
0.209 |
0.076 |
0.172 |
0.267 |
0.251 |
0.737 |
0.787 |
0.691 |
1.000 |
|
|
|
|
|
Warmth |
Asian |
0.211 |
-0.203 |
-0.116 |
0.274 |
0.102 |
0.183 |
0.144 |
0.154 |
0.096 |
1.000 |
|
|
|
Latinos |
0.009 |
-0.225 |
-0.121 |
0.132 |
-0.005 |
0.100 |
0.117 |
0.202 |
0.091 |
0.679 |
1.000 |
|
|
|
Euro |
0.062 |
-0.107 |
0.035 |
0.165 |
0.017 |
0.326 |
0.297 |
0.379 |
0.272 |
0.632 |
0.640 |
1.000 |
|
|
Standard Dev. |
1.750 |
1.852 |
1.623 |
1.602 |
1.864 |
1.390 |
1.473 |
1.376 |
1.460 |
1.444 |
1.375 |
1.172 |
|
|
Mean |
4.336 |
3.374 |
3.464 |
4.245 |
4.659 |
5.491 |
5.194 |
5.536 |
5.145 |
5.472 |
5.490 |
5.855 |
|
* Ingroup warmth in boldface
|
|
|
Exploration |
Affirmation &
Belonging |
Warmth |
|||||||||
|
|
|
Item 1 |
Item 2 |
Item 3 |
Item 4 |
Item 5 |
Item 1 |
Item 2 |
Item 3 |
Item 4 |
Asian |
Latinos |
Euro |
|
Exploration |
Item 1 |
1.000 |
|
|
|
|
|
|
|
|
|
|
|
|
Item 2 |
0.299 |
1.000 |
|
|
|
|
|
|
|
|
|
|
|
|
Item 3 |
0.267 |
0.183 |
1.000 |
|
|
|
|
|
|
|
|
|
|
|
Item 4 |
0.458 |
0.191 |
0.333 |
1.000 |
|
|
|
|
|
|
|
|
|
|
Item 5 |
0.263 |
0.316 |
0.053 |
0.382 |
1.000 |
|
|
|
|
|
|
|
|
|
Affirmation & Belonging |
Item 1 |
0.408 |
0.303 |
0.143 |
0.247 |
0.470 |
1.000 |
|
|
|
|
|
|
|
Item 2 |
0.388 |
0.450 |
0.205 |
0.228 |
0.399 |
0.661 |
1.000 |
|
|
|
|
|
|
|
Item 3 |
0.339 |
0.280 |
0.092 |
0.255 |
0.427 |
0.821 |
0.589 |
1.000 |
|
|
|
|
|
|
Item 4 |
0.469 |
0.427 |
0.149 |
0.299 |
0.420 |
0.688 |
0.628 |
0.717 |
1.000 |
|
|
|
|
|
Warmth |
Asian |
0.173 |
0.306 |
0.111 |
0.252 |
0.273 |
0.497 |
0.404 |
0.430 |
0.430 |
1.000 |
|
|
|
Latinos |
0.073 |
0.017 |
0.030 |
0.244 |
0.210 |
0.328 |
0.183 |
0.244 |
0.155 |
0.539 |
1.000 |
|
|
|
Euro |
0.169 |
0.053 |
0.037 |
0.182 |
0.197 |
0.237 |
0.158 |
0.194 |
0.124 |
0.529 |
0.601 |
1.000 |
|
|
Standard Dev. |
1.487 |
1.598 |
1.427 |
1.660 |
1.510 |
1.298 |
1.433 |
1.469 |
1.439 |
1.173 |
1.406 |
1.296 |
|
|
Mean |
4.660 |
4.009 |
3.860 |
4.293 |
5.371 |
5.578 |
5.208 |
5.391 |
5.231 |
6.101 |
5.390 |
5.530 |
|
* Ingroup warmth in boldface
|
|
|
Exploration |
Affirmation &
Belonging |
Warmth |
|||||||||
|
|
|
Item 1 |
Item 2 |
Item 3 |
Item 4 |
Item 5 |
Item 1 |
Item 2 |
Item 3 |
Item 4 |
Asian |
Latinos |
Euro |
|
Exploration |
Item 1 |
1.000 |
|
|
|
|
|
|
|
|
|
|
|
|
Item 2 |
0.071 |
1.000 |
|
|
|
|
|
|
|
|
|
|
|
|
Item 3 |
0.081 |
0.315 |
1.000 |
|
|
|
|
|
|
|
|
|
|
|
Item 4 |
0.385 |
0.175 |
0.197 |
1.000 |
|
|
|
|
|
|
|
|
|
|
Item 5 |
0.341 |
0.172 |
0.170 |
0.233 |
1.000 |
|
|
|
|
|
|
|
|
|
Affirmation & Belonging |
Item 1 |
0.218 |
0.085 |
0.047 |
0.134 |
0.304 |
1.000 |
|
|
|
|
|
|
|
Item 2 |
0.223 |
-0.031 |
0.143 |
0.020 |
0.145 |
0.358 |
1.000 |
|
|
|
|
|
|
|
Item 3 |
0.094 |
-0.073 |
0.063 |
-0.051 |
0.249 |
0.561 |
0.496 |
1.000 |
|
|
|
|
|
|
Item 4 |
0.203 |
-0.090 |
-0.005 |
0.046 |
0.312 |
0.545 |
0.396 |
0.621 |
1.000 |
|
|
|
|
|
Warmth |
Asian |
0.165 |
-0.267 |
-0.117 |
0.016 |
0.104 |
0.325 |
0.097 |
0.185 |
0.331 |
1.000 |
|
|
|
Latinos |
0.208 |
-0.056 |
-0.014 |
-0.021 |
0.168 |
0.596 |
0.280 |
0.340 |
0.497 |
0.404 |
1.000 |
|
|
|
Euro |
0.103 |
-0.245 |
-0.092 |
-0.012 |
-0.041 |
0.238 |
0.025 |
0.104 |
0.186 |
0.623 |
0.512 |
1.000 |
|
|
Standard Dev. |
1.513 |
1.541 |
1.461 |
1.491 |
1.545 |
1.202 |
1.347 |
1.261 |
1.350 |
1.350 |
1.090 |
1.216 |
|
|
Mean |
4.361 |
3.343 |
3.815 |
4.298 |
5.460 |
5.889 |
5.358 |
5.917 |
5.526 |
5.250 |
6.091 |
5.470 |
|
* Ingroup warmth in boldface
|
|
|
Exploration |
Affirmation &
Belonging |
Warmth |
|||||||||
|
|
|
Item 1 |
Item 2 |
Item 3 |
Item 4 |
Item 5 |
Item 1 |
Item 2 |
Item 3 |
Item 4 |
Asian |
Latinos |
Euro |
|
Exploration |
Item 1 |
1.000 |
|
|
|
|
|
|
|
|
|
|
|
|
Item 2 |
0.443 |
1.000 |
|
|
|
|
|
|
|
|
|
|
|
|
Item 3 |
0.426 |
0.515 |
1.000 |
|
|
|
|
|
|
|
|
|
|
|
Item 4 |
0.603 |
0.246 |
0.395 |
1.000 |
|
|
|
|
|
|
|
|
|
|
Item 5 |
0.530 |
0.408 |
0.298 |
0.404 |
1.000 |
|
|
|
|
|
|
|
|
|
Affirmation & Belonging |
Item 1 |
0.199 |
0.347 |
0.334 |
0.210 |
0.285 |
1.000 |
|
|
|
|
|
|
|
Item 2 |
0.182 |
0.264 |
0.340 |
0.224 |
0.242 |
0.670 |
1.000 |
|
|
|
|
|
|
|
Item 3 |
0.242 |
0.235 |
0.305 |
0.430 |
0.440 |
0.438 |
0.319 |
1.000 |
|
|
|
|
|
|
Item 4 |
0.292 |
0.292 |
0.302 |
0.355 |
0.375 |
0.464 |
0.471 |
0.565 |
1.000 |
|
|
|
|
|
Warmth |
Asian |
0.345 |
0.141 |
0.147 |
0.142 |
0.261 |
-0.033 |
-0.030 |
0.137 |
0.128 |
1.000 |
|
|
|
Latinos |
0.309 |
0.027 |
0.084 |
0.198 |
0.119 |
-0.005 |
-0.054 |
0.179 |
0.109 |
0.679 |
1.000 |
|
|
|
Euro |
0.263 |
0.139 |
0.145 |
0.158 |
0.222 |
0.132 |
0.079 |
0.246 |
0.248 |
0.632 |
0.640 |
1.000 |
|
|
Standard Dev. |
1.674 |
1.741 |
1.592 |
1.642 |
1.673 |
1.502 |
1.566 |
1.278 |
1.477 |
1.444 |
1.375 |
1.172 |
|
|
Mean |
4.648 |
3.912 |
3.840 |
4.656 |
4.873 |
5.380 |
4.808 |
5.482 |
4.887 |
5.472 |
5.490 |
5.855 |
|
* Ingroup warmth in boldface
|
|
|
Exploration |
Affirmation &
Belonging |
Warmth |
|||||||||
|
|
|
Item 1 |
Item 2 |
Item 3 |
Item 4 |
Item 5 |
Item 1 |
Item 2 |
Item 3 |
Item 4 |
Asian |
Latinos |
Euro |
|
Exploration |
Item 1 |
1.000 |
|
|
|
|
|
|
|
|
|
|
|
|
Item 2 |
0.428 |
1.000 |
|
|
|
|
|
|
|
|
|
|
|
|
Item 3 |
0.406 |
0.505 |
1.000 |
|
|
|
|
|
|
|
|
|
|
|
Item 4 |
0.596 |
0.260 |
0.371 |
1.000 |
|
|
|
|
|
|
|
|
|
|
Item 5 |
0.484 |
0.469 |
0.322 |
0.353 |
1.000 |
|
|
|
|
|
|
|
|
|
Affirmation & Belonging |
Item 1 |
0.277 |
0.285 |
0.160 |
0.278 |
0.361 |
1.000 |
|
|
|
|
|
|
|
Item 2 |
0.342 |
0.279 |
0.221 |
0.318 |
0.251 |
0.609 |
1.000 |
|
|
|
|
|
|
|
Item 3 |
0.389 |
0.378 |
0.367 |
0.366 |
0.462 |
0.573 |
0.465 |
1.000 |
|
|
|
|
|
|
Item 4 |
0.313 |
0.373 |
0.302 |
0.427 |
0.374 |
0.536 |
0.608 |
0.646 |
1.000 |
|
|
|
|
|
Warmth |
Asian |
0.242 |
0.261 |
0.266 |
0.261 |
0.249 |
0.289 |
0.276 |
0.295 |
0.453 |
1.000 |
|
|
|
Latinos |
0.128 |
-0.052 |
0.155 |
0.230 |
0.045 |
0.128 |
0.063 |
0.137 |
0.202 |
0.539 |
1.000 |
|
|
|
Euro |
0.196 |
0.062 |
0.141 |
0.278 |
0.169 |
0.158 |
0.116 |
0.061 |
0.256 |
0.529 |
0.601 |
1.000 |
|
|
Standard Dev. |
1.435 |
1.669 |
1.520 |
1.598 |
1.413 |
1.408 |
1.525 |
1.472 |
1.621 |
1.173 |
1.406 |
1.296 |
|
|
Mean |
4.938 |
4.271 |
4.194 |
4.688 |
5.467 |
5.502 |
5.021 |
5.382 |
5.048 |
6.101 |
5.390 |
5.530 |
|
* Ingroup warmth in boldface
|
|
|
Exploration |
Affirmation &
Belonging |
Warmth |
|||||||||
|
|
|
Item 1 |
Item 2 |
Item 3 |
Item 4 |
Item 5 |
Item 1 |
Item 2 |
Item 3 |
Item 4 |
Asian |
Latinos |
Euro |
|
Exploration |
Item 1 |
1.000 |
|
|
|
|
|
|
|
|
|
|
|
|
Item 2 |
0.281 |
1.000 |
|
|
|
|
|
|
|
|
|
|
|
|
Item 3 |
0.407 |
0.367 |
1.000 |
|
|
|
|
|
|
|
|
|
|
|
Item 4 |
0.359 |
0.240 |
0.301 |
1.000 |
|
|
|
|
|
|
|
|
|
|
Item 5 |
0.291 |
0.291 |
0.186 |
0.360 |
1.000 |
|
|
|
|
|
|
|
|
|
Affirmation & Belonging |
Item 1 |
0.118 |
0.072 |
0.223 |
0.225 |
0.337 |
1.000 |
|
|
|
|
|
|
|
Item 2 |
0.037 |
0.218 |
0.369 |
0.117 |
0.192 |
0.383 |
1.000 |
|
|
|
|
|
|
|
Item 3 |
0.309 |
0.144 |
0.041 |
0.117 |
0.405 |
0.272 |
0.232 |
1.000 |
|
|
|
|
|
|
Item 4 |
0.280 |
0.213 |
0.258 |
0.245 |
0.411 |
0.220 |
0.328 |
0.646 |
1.000 |
|
|
|
|
|
Warmth |
Asian |
0.291 |
0.038 |
0.146 |
0.182 |
0.254 |
0.099 |
-0.0003 |
0.210 |
0.174 |
1.000 |
|
|
|
Latinos |
0.261 |
0.019 |
0.041 |
0.140 |
0.439 |
0.277 |
-0.040 |
0.200 |
0.234 |
0.404 |
1.000 |
|
|
|
Euro |
0.270 |
-0.113 |
0.090 |
0.123 |
0.174 |
0.107 |
-0.231 |
0.026 |
0.047 |
0.623 |
0.512 |
1.000 |
|
|
Standard Dev. |
1.389 |
1.483 |
1.400 |
1.403 |
1.435 |
1.426 |
1.378 |
1.091 |
1.192 |
1.350 |
1.090 |
1.216 |
|
|
Mean |
4.657 |
4.023 |
4.180 |
4.778 |
5.294 |
5.774 |
5.005 |
6.028 |
5.681 |
5.250 |
6.091 |
5.470 |
|
* Ingroup warmth in boldface
Items from the
Multigroup Ethnic Identity Measure
Ethnic Identity Exploration Items
1.
I have spent time trying to find out more about my
ethnic group, such as its history, traditions, and customs
2.
I am active in organizations or social groups that
include mostly members of my own ethnic group
3.
I think a lot about how my life will be
affected by my ethnic group membership
4.
In order to learn more about my ethnic background, I
have often talked to other people about my ethnic group
5.
I participate in cultural practices of my own group,
such as special food, music, or customs
Ethnic Identity Affirmation and Belonging Items
1. I am happy
that I am a member of the ethnic group I belong to
2.
I have a strong sense of belonging to my own ethnic
group
3.
I have a lot of pride in my ethnic group
4.
I feel a strong attachment towards my own ethnic group