New meta-analytic research from our lab on the integration hypothesis suggests that acculturation generally and the integration strategy specifically are much less robust predictors of migrants’ adaptation than previously assumed.
By Natascha Janho, Jonas R. Kunst, & Kinga Bierwiaczonek
It is not unusual in psychological research to encounter rivaling standpoints and controversy regarding a particular theory or paradigm. Indeed, controversies are inherent to scientific advances because they allow the research community to critically evaluate the beliefs they took for granted for a long time. Most recently, in the field of acculturation, new meta-analytic research from our lab suggests that acculturation generally and the integration strategy specifically are much less robust predictors of migrants’ adaptation than previously assumed.
What is acculturation?
When a person leaves their home to move to another country, they can find themselves in a challenging position. They are often members of an ethnic minority group. Their socialization, education, values, traditions, and every experience down to their native language are part of their heritage cultural identity – an identity they do not share with the majority-group members within the new society. This circumstance requires a response, a strategy for successful adaptation. It is this response acculturation research aims to investigate. The term acculturation refers to the bidirectional processes of psychological and cultural change that occur on an individual and group level when people of different cultural backgrounds come into contact.
The integration hypothesis in acculturation psychology
With his seminal model of acculturation, John W. Berry (1997) contributed an influential framework that is still widely used today. According to the model, the acculturation of minority-group members can be categorized into four acculturation strategies.
First, people who follow the assimilation strategy adopt parts of the majority culture while abandoning their heritage culture. Second, those who follow a separation strategy reject the majority culture while maintaining their heritage culture. Third, when immigrants follow the integration strategy, they adopt the majority culture while maintaining their heritage culture. Lastly, when they follow a marginalization strategy, they neither adopt the majority culture nor maintain their heritage culture.
With this four-fold categorization of the different acculturation styles, acculturation theory does not only aim to describe how immigrants relate to different cultures but also to predict their adaptation. In acculturation research, adaptation is often divided into two interconnected but distinct dimensions: psychological adaptation and sociocultural adaptation. These dimensions refer to the ways in which individuals adjust to a new culture after migration or exposure to a different cultural environment.
1. Psychological adaptation: This dimension refers to the mental and emotional well-being of an individual as they adapt to the new cultural context. It encompasses aspects such as self-esteem, life satisfaction, mental health, and overall psychological well-being. In acculturation research, psychological adaptation is often assessed through indicators like stress levels, anxiety, depression, and subjective well-being. A successful psychological adaptation means that the individual is able to maintain a positive sense of self and mental health while navigating the challenges of adjusting to the new culture.
2. Sociocultural adaptation: This dimension focuses on the individual’s ability to function effectively and navigate the social and cultural aspects of the new environment. Sociocultural adaptation involves learning and mastering the skills, norms, and behaviors required to interact with the majority culture and its social institutions successfully. Indicators of successful sociocultural adaptation include social competence, communication skills, and the ability to establish and maintain productive relationships with members of the new culture. This adaptation also involves understanding and adhering to cultural norms, customs, and expectations in various social situations.
Importantly, Berry (2013) postulates with his prominent “integration hypothesis” that the integration strategy leads to the most favorable psychological and sociocultural adaptation for immigrants and minority-group members. In other words, individuals are thought to achieve the best psychological and sociocultural adaptation by adopting the majority culture while maintaining their heritage culture.
Does more research always equal more knowledge?
Psychological theories can influence which measures are considered politically or socially useful. When it comes to the study of acculturation, knowledge can have far-reaching implications. Determining successful adaptation and the policies that could promote it is a core challenge in these globalized times. The integration hypothesis by Berry aims to provide a useful theoretical framework to inform solutions to this challenge, but is it based on reliable evidence?
The importance of acculturation is well reflected in the rapid growth of academic attention the topic has received: more than 13,000 scientific articles related to acculturation were published within less than a century, between 1923 and 2020 (Web of Science, 2021). Generally speaking, it seems plausible that more research on a topic leads to a deeper understanding of the subject matter. When theories get validated or challenged through peer-reviewed research, hypotheses are thought to be supported, disregarded, or become more refined. However, the overall findings within a research field often first become comprehensible through a meta-analysis.
A meta-analysis is a statistical technique used to combine and analyze the results of multiple individual studies, providing a more comprehensive understanding of a particular research question. By aggregating data from various sources, it increases statistical power and helps to draw more reliable conclusions. Meta-analyses are particularly useful in identifying overarching trends, reducing inconsistencies between studies, and highlighting areas for further research.
Throughout the years, Berry’s influential model of acculturation (1997) and the derived integration hypothesis have not remained unchallenged, both conceptually and methodologically. In 2003, Floyd Rudmin contested the original concept of the four acculturation tendencies. His criticism focused on the often-weak reliability of measures of acculturation as well as logically implausible correlations between acculturation strategies that are mutually exclusive. Instead, he suggested the field should consider the two dimensions of heritage-culture maintenance and mainstream-culture adoption separately instead of measuring the four strategies with so-called four-fold acculturation measures (with distinct items measuring each strategy).
Berry and colleagues responded to this criticism with empirical evidence that seemed to demonstrate support for the integration hypothesis regardless of how integration was measured or calculated (Berry et al., 2006). Seemingly settling the debate, Nguyen and Benet-Martínez (2013) published a comprehensive meta-analysis that appeared to provide robust support for Berry’s integration hypothesis. Across 83 studies, the researchers found what they described as a strong positive relationship between integration and adaptation.
Thus, many researchers and practitioners have long assumed that the existing evidence overwhelmingly favored the integration hypothesis. However, more recent meta-analytical research from our lab (Bierwiaczoneck & Kunst, 2021) suggests that this conclusion is based on much less robust evidence than previously assumed. Indeed, it seems as if acculturation strategies, including the integration strategy, may be of very little importance when it comes to immigrants’ and minority-group members’ adaptation.
Revisiting the Integration Hypothesis
Specifically, the meta-analytic paper empirically addressed two main points of criticism. The first one is statistical in nature and regards the meta-analysis by Nguyen and Benet-Martínez (2013). In their original meta-analysis, these authors analyzed their data with the Rosenthal approach to random effects (Rosenthal, 1995; Rosenthal & DiMatteo, 2001; Rosenthal & Rubin, 1994), an approach proposed in the 90s that today remains largely unknown to meta-analysts worldwide. This approach estimates random effects as counternull values of fixed effects, a statistic that requires very careful interpretation in the context of other statistics.
The counternull values are considerably larger than, and often twice as large as, the fixed effects, and they are not point estimates of the average effect across studies (Rosenthal & Rubin, 1994). Thus, this original analysis should not be interpreted in the usual way effect sizes are evaluated. This rather uncommon method for analyzing the data increases the risk of misinterpretation.
Therefore, our lab members first reanalyzed the data from Nguyen and Benet-Martínez (2013) using a state-of-the-art robust approach based on the Hedges and Olkin (1985) method to obtain readily interpretable effects. In contrast to the original results, the reanalysis found only a weak and inconsistent correlation between adaptation and the integration strategy. Overall, acculturation could only explain as little as 0.8% to 1.4% of the differences in adaptation. Thus, the first study’s findings gave reason to consider the claim of the integration hypothesis (Berry, 2013) as being based on uncertain evidential grounds.
However, the second study addressed a more fundamental criticism regarding the entire acculturation field. Whereas the definitions of acculturation may differ in their exact wording, most define it as a causal process (Kunst, 2021). To investigate causal relationships, only experimental or longitudinal methods are suitable. Yet, strikingly, the vast majority of papers published on the topic of acculturation are correlational. For instance, 94% of the primary studies in the Nguyen and Benet-Martínez (2013) meta-analysis were based on correlational data (Bierwiaczoneck & Kunst, 2021).
Thus, the divergence between the conception and the actual research conducted on acculturation is a significant limitation because correlational data are not suitable for drawing reliable causal inferences. Therefore, in a second study, our lab members conducted a meta-analysis using only primary studies that utilized longitudinal data. In total, 19 papers with 6791 participants satisfied this inclusion criterion and were included in the meta-analysis.
The findings of the longitudinal analyses were generally in line with the first study. Almost all longitudinal relationships between acculturation strategies and orientations and adaptation approached zero. Furthermore, the effects were often inconsistent (i.e., weakly positive for one longitudinal path but weakly negative for the other).
What determines the importance of a small acculturation effect?
A recurring argument in defense of the integration hypothesis is the claim that even a reliable small effect found in research still matters at a large scale. In principle, this is a valid argument. Indeed, it is common in psychology that effect sizes are small. However, the argument overlooks that the small effects observed in a correlational meta-analysis do not imply causality. Longitudinally, there is little evidence for even a small effect of the integration strategy on adaptation. Second, such an argument overly focuses on p-values which are of less importance in high-powered meta-analyses, while ignoring the large heterogeneity of effects, that is, the fact that effects might differ dramatically from one study to another.
One statistic that can give us some insight into how heterogenous effects are is I², which indicates how much of the variability in the results is attributable to heterogeneity instead of random variance. Guidelines suggest that I² values above 50% – 75% percent indicate high heterogeneity (compare Bornstein et al., 2017 for further detail). Such high heterogeneity complicates the interpretation of results substantially. According to Stanley (2017), values above 80% indicate that there might not be a consistent true effect because the consistency of the underlying phenomena is questionable.
According to a recent review (Kunst, 2021), the heterogeneity of the effects of acculturation orientations and strategies on adaptation is problematically high in every existing meta-analysis. Importantly, attempts to identify moderators that may explain a meaningful portion of this heterogeneity have been unsuccessful.
To further demonstrate the problematic nature of high effect heterogeneity in the field of acculturation, we in a recent paper (Bierwiaczonek, Cheung & Kunst, 2022) calculated and visualized the estimated distributions of true effects of acculturation on adaptation with bell curves. The distribution of true effects across studies shows that integration has a negative correlation with adaptation in almost one third of all cases (27% – 30%). Thus, even when we ignore the lack of causal evidence and only focus on correlational studies, the evidence for the integration hypothesis remains highly inconsistent, indicating that in one in three cases, adopting the integration strategy will be associated with worse adaptation.
Implications for the next generation of acculturation research
The new meta-analytic evidence has important implications. The combination of small effect sizes and high heterogeneity makes it difficult to interpret the evidence and identify its societal implications. It underlines the problematic reliance on correlational research to study a causal phenomenon and emphasizes the need for more experimental and longitudinal studies (Kunst, 2021). The proportion of these designs in the field of acculturation is still very small – under 10% for longitudinal studies and 3% for experiments. Although longitudinal studies are not ideal for causal inference, they are better than relying on correlational data.
Furthermore, enhancing the variety of experimental designs should be a key objective for future research, which may involve developing and validating innovative experimental manipulations. While this endeavor is undoubtedly challenging, drawing inspiration from existing work in other domains, such as social identity research, could provide valuable insights (see for example Ben-Nun Bloom et al., 2015; Liu et al., 2019; Wojcieszak & Garrett, 2018 ).
Finally, the finding that acculturation seems to play a minimal role in the adaptation of immigrants and minority-group members has practical implications for policymakers and practitioners. These implications emphasize the importance of focusing on factors that more reliably predict psychological and sociocultural adaptation, such as discrimination experiences and the presence or absence of social support. Thus, it may be meaningful for policymakers to prioritize addressing discrimination experiences and promoting social support systems for immigrants and minority-group members. This focus may include implementing anti-discrimination laws, promoting diversity and inclusion initiatives, and fostering community programs that facilitate social integration.
References to studies from the lab:
Bierwiaczonek, K., Cheung, M. W., & Kunst, J. R. (2022, October 20). Revisiting the integration hypothesis again: High heterogeneity complicates the interpretation of cross-sectional evidence. https://doi.org/10.31234/osf.io/32npb
Bierwiaczonek, K., & Kunst, J. R. (2021). Revisiting the integration hypothesis: Correlational and longitudinal meta-analyses demonstrate the limited role of acculturation for cross-cultural adaptation. Psychological Science, 32(9), 1476-1493. https://doi.org/10.1177/09567976211006432
Kunst, J. R. (2021). Are we facing a “causality crisis” in acculturation research? The need for a methodological (r) evolution. International Journal of Intercultural Relations, 85, A4-A8. https://doi.org/10.1016/j.ijintrel.2021.08.003
(links to other references are provided in text)