Berry’s (1997) acculturation framework

endorses bicultural integration as the best

predictor of adjustment. The assessment of

integration used in this research has, however,

been criticized on psychometric grounds

(Rudmin, 2005). Ryder et al. (2000) directly

assessed the heritage and mainstream dimensions

of acculturation and found that only mainstream

acculturation predicted psychological adjustment.

Given that the effects of integration may be

underestimated without direct assessment (Berry

& Sam, 2003), various methods to compute

integration from separate heritage and mainstream

scores have been proposed.

We tested five of these methods using the

Vancouver Index of Acculturation (VIA),

Acculturation Index (AI; Ward & Rana-Deuba,

1999), and J. W. Berry’s four-fold acculturation

measure (personal communication, February 10,

1998) in order to (a) evaluate their psychometric

properties and (b) test the hypothesis that

integration predicts adjustment.

Only one integration method (“interaction

multiculturalism”) was independent of the

heritage and mainstream dimensions; this method

had no relation with social adjustment. None of

the other integration methods were related to

social anxiety after controlling for mainstream

and heritage dimensions. There was no evidence

in this sample that integration offers unique

benefits above and beyond mainstream



The acculturation framework of John Berry

(1996) proposes that a bicultural integration

strategy – identification with the heritage culture

combined with participation in the mainstream

culture – leads to the best outcomes for mental

health. The integration of heritage and

mainstream cultural identities remains a

compelling hypothesis. Increasing numbers of

researchers have criticized Berry’s four-fold

measurement approach on psychometric and

theoretical grounds. The most common proposed

solution has been to measure separate dimensions

of heritage and mainstream acculturation.

However, these researchers often fail to consider

how the two dimensions interact with one

another; thus, the role of integration remains


Berry and Sam (2003) argued that one cannot

assess the effects of integration unless it is directly

measured. Rather than using the problematic four-

fold method, Rudmin (2005) has proposed that

integration can be measured by combining results

from heritage and mainstream subscales. To this

end, Rudmin (2005) has suggested several

different methods of manipulating subscale scores

from standard bidimensional acculturation

instruments in order to yield indices of

integration. The current study is designed to test

these manipulations in a sample of Chinese-

Canadian undergraduate students in the context of

social adjustment.


Data for this study were drawn from a larger

cross-cultural investigation of Chinese-Canadian

university students (N=203) who completed a

large questionnaire package. The current study

uses data from the Vancouver Index of

Acculturation (VIA; Ryder et al., 2000),

Acculturation Index (AI; Ward & Rana-Deuba,

1999), J. W. Berry’s four-fold acculturation

measure, and the Social Avoidance and Distress

Scale (SADS; Watson & Friend, 1969).

Heritage and mainstream dimension scores

were used to generate several integration

measures following Rudmin (2005). The heritage

and mainstream dimension scores and the

integration measures were then intercorrelated as

well as correlated with Berry’s measure of

integration as well as social anxiety (SAI).

Given the considerable correlation among the

Berry subscales, partial correlation analyses were

performed in order to control for assimilation,

separation, and marginalization. Further, partial

regression analysis was performed to assess the

contribution of each integration measure over and

above the heritage and mainstream dimensions.


Main effects

- Heritage (H): Sum of heritage items on VIA

or AI.

- Mainstream (M): Sum of mainstream items

on VIA or AI.

Arithmetic transformations

- Summation biculturalism = H + M

- Integration biculturalism = H x M

Note: Although discussed in Rudmin (2005),

subtraction biculturalism (H – M) was not

included in the analyses because, unlike the other

measures, it does not describe a dimension

ranging from low to high integration.

Centralized transformations

- Interaction multiculturalism = (H – MH) x

(M – MM); MH and MM are the means of the

heritage and mainstream scales, respectively.


- Expressed multiculturalism = (H – MptH) x

(M – MptM); MptH and MptM are the

midpoints of the heritage and mainstream

scales, respectively.

- Deviation from biculturalism: using polar

coordinates, how many degrees participants’

scores deviate from biculturalism; i.e., the

theta angle of deviation from an integration

reference axis. Scores are reversed such that

larger scores indicate more integration.


Across the VIA and AI, significant zero-order

correlations and partial correlations with Berry’s

integration score were observed for Summation

biculturalism, Integration biculturalism, and

Expressed multiculturalism. Interaction

multiculturalism failed to show a zero-order

correlation with Berry’s integration score on the

AI. There was no effect observed for Deviation

from biculturalism (Table 1).

The main effects of heritage and mainstream

scores as well as all the integration

transformations either showed mildly negative or

no relation to social anxiety. None of the Berry

four-fold acculturation measures were associated

with social anxiety (Table 2).

All integration methods, except for interaction

multiculturalism, were heavily saturated with

heritage and mainstream dimensions (Table 3).

Partial regression analyses showed that no

measure of integration contributed significantly to

the prediction of social adjustment over and

above the effects of the mainstream and heritage

dimensions. This last finding is likely attributable

to the high degree of intercorrelation between

most of the integration measures with the heritage

and mainstream dimensions, discussed above.


When methods of measuring integration

succeed in predicting social adjustment, they

appear to work because they share variance with

the mainstream dimension. This finding is in

keeping with Ryder et al. (2000), who found that

most adjustment effects are carried by the

mainstream dimension.

The study is limited by the fact that one

particular form of adjustment (i.e., social anxiety)

was investigated in a restricted sample population

(i.e., Chinese-Canadian undergraduate students).

As such, the measurement of integration and the

hypotheses associated with it need to be further

studied in other samples with other adjustment


That said, there is little evidence in this study

to support the unique role of integration in the

prediction of adjustment. In general, larger effect

sizes were observed for the VIA compared with

the AI.

The larger implication of this study for

acculturation researchers using bidimensional

methods is that these methods allow for the

assessment of integration. We encourage

acculturation researchers to continue using

bidimensional measures, and to routinely

investigate the effects of integration using these

measures. In time, a sufficient cross-method and

cross-setting database will exist, allowing

researchers to return to the integration-adjustment



Berry, J. W. (1997). Intergration, acculturation, and

adaptation. Applied Psychology: An International

Review, 46, 5-34.

Berry, J. W., & Sam, D. L. (2003). Accuracy in scientific

discourse. Scandinavian Journal of Psychology, 44, 65-


Rudmin, F. (2005). Debate in science: The case of

acculturation. (unpublished manuscript)

Ryder, A. G., Alden, L. E., & Paulhus, D. L. (2000). Is

acculturation unidimensional or bidimensional? A head-

to-head comparison in the prediction of personality, self-

identity, and adjustment. Journal of Personality and

Social Psychology, 79, 49-65.

Ward, C., & Rana-Deuba, A. (1999). Acculturation and

adaptation revisited. Journal of Cross-Cultural

Psychology , 30, 422-442.

Watson, D., & Friend, R. (1969). Measurement of social-

evaluative anxiety. Journal of Consulting and Clinical

Psychology, 33, 448-457.