Organizers:
Dagmar Strohmeier & Christiane Spiel, University of Vienna, Austria

Chair:
Christiane Spiel, Universtity of Vienna, Austria

Presenters:
(1) Hildegunn Fandrem, University of Stavanger, Norway
(2) Dagmar Strohmeier, University of Vienna, Austria
(3) Rens van der Schoot, Utrecht University, The Netherlands
(4) Elisabeth Stefanek, University of Vienna, Austria

Title of Panel:
Peer Relations in Immigrant Adolescents: Methodological Challenges and Key Findings

Overall Abstract:
This panel brings together researchers from Austria, Norway and The Netherlands. Drawing from studies on peer relations in immigrant youth, the three papers will address important methodological challenges and will present possible solutions.

Paper 1:
Hildegunn Fandrem
Methodologigal challenges in an immigrant study in Norway
The presentation will focus on methodological challenges in an immigrant study in Norway. The study investigates psychological and sociocultural adaptation among immigrant adolescents compared to natives in Norway. More specifically depression and bullying were studied in relation to different variables. First, problems concerning small immigrant samples, generalization and issues concerning immigrants as a homogeneous group will be discussed. Second, important considerations when using specific concepts or cultural categories (as e. g. 'depression') will be focused. Third issues related to measures on ethnic versus native peer contact will be raised. Last the measurement of acculturation strategies will be discussed in relation to behavior.

Paper 2:
Dagmar Strohmeier, Rens van der Schoot & Christiane Spiel
Friendship homophily in multicultural classes: Effects of personal preferences and structural opportunity
Homophily is a preference principle referring to the tendency that friendships between similar people occur at a higher rate than among dissimilar ones. However, the pattern to have friends from the same cultural group is not only a result of personal preference but also of social structure (e.g., the cultural make-up of a class room). Main goal of the present study is to predict individual homophily taking both individual and class room variables into account.
The cultural make-up of 59 classes and friendship data of 1100 4th to 9th graders were available for analyses. To calculate diversity on class level an index was used that captured both the number of different cultural groups and their relative representation in class. Homophily on individual level was estimated using an index controlling for both class size and number of available peers of the same cultural group. HLM analyses were conducted. Using gender and cultural group as individual variables and diversity as class variable we were able to explain 31.6% out of 83% of the variance in homophily on the individual level and 83.3% out of 17% of the variance located at the class level.

Paper 3:
Rens van der Schoot
Learning more from data: How to move beyond Traditional Null Hypothesis Testing

In practice, researchers have specific expectations about their research questions formulated with inequality constraints. Phrases like “The mean outcome in both experimental groups is expected to be larger than in the control group” and “women score higher than men in each condition” can be found in many papers. These specific expectations may be derived from theories, empirical evidence, or both and will be called informative hypotheses. Yet, despite these expectations, most investigators still rely on classical null hypothesis testing and ignore the empirical evidence that has already been accumulated. In the current presentation alternative methods will be introduced to actually use prior information in the data analyses. A running example will be used about the goal to be accepted by friends as underlying function of aggressive behaviour in immigrant adolescents.


Paper 4:
Elisabeth Stefanek, Dagmar Strohmeier, Hildegunn Fandrem & Christiane Spiel
Depressive symptoms in native and immigrant adolescents: The role of critical life events and daily hassles

In this study, 1) native Austrian and first and second generation immigrant adolescents were compared regarding their level of depressive symptoms, critical live events and daily hassles, and 2) three theoretical models (independent effects, partially mediated, fully mediated) of the associations between these constructs were tested in the three groups of adolescents. The sample comprised 330 native Austrians (51.2% girls), 120 first generation immigrants (49.2% girls), and 159 second generation immigrants (54.1% girls) aged 14 to 19 (M = 15.61). Data were collected via self assessments. Compared with native youth, first generation immigrants reported more depressive symptoms and more daily hassles related to parents, the self, leisure, romantic partner, and future whilst second generation immigrants reported more daily hassles related to parents, school, and romantic partner. Multiple group structural equation models revealed that the fully mediated model fitted the data best. Controlling for gender, daily hassles fully mediated the path between critical live events and depressive symptoms in all three groups of adolescents. Results highlight the importance of daily hassles for depressive symptoms in immigrant youth.