Language attitudes, family language and generational cohort in Catalonia: new contributions from a multivariate analysis

ABSTRACT

The effect of the generational cohort on language attitudes has hardly ever been studied. However, especially in the bilingual territories in Spain, the literature stresses the importance of being Spanish-speakers or non-Spanish-speakers in the construction of these attitudes. Focusing on the case of Catalonia (Spain), this paper intends to analyse the impact of the family language and the generational cohort on the language attitudes towards Catalan and Spanish that is built by immigrant children (aged 14–16). The nature of the analysis carried out as a result of an attitude questionnaire answered by 527 young people allows differentiation of the effect of both variables individually when included in the same explanatory model. Results indicate that being Spanish-speaking or non-Spanish-speaking is influential in the construction of attitudes towards Spanish and towards Catalan. Moreover, the generational cohort is only influential for attitudes towards Catalan and when a migratory transit has been experienced or not. Since both variables are included in the same model, it is suggested that the effect of the generational cohort is moderated by being Spanish-speaking or not. The new perspectives opened up by these results are discussed as regards the promotion of intercultural communication and a common project of coexistence.

El efecto de la cohorte generacional sobre las actitudes lingüísticas es un aspecto que prácticamente nunca se ha estudiado. Por su parte, especialmente en los territorios bilingües del Estado español, la literatura reitera la importancia del ser Hispanohablante o no Hispanohablante en la construcción de estas actitudes. Situados en Cataluña (España), el objetivo de este artículo es analizar el impacto de la condición lingüística familiar y la cohorte generacional sobre las actitudes lingüísticas hacia el catalán y el castellano que construyen los descendientes de migrados (14–16 años). La naturaleza del análisis realizado, fruto de la aplicación de un cuestionario de actitudes a un total de 527 jóvenes, permite diferenciar el efecto de ambas variables individualmente y si se incluyen en un mismo modelo explicativo. Los resultados indican que el ser Hispanohablante o No Hispanohablante es influyente tanto en la construcción de las actitudes hacia el castellano como hacia el catalán. Por otro lado, la cohorte generacional solamente lo es en el caso del catalán y si se ha vivido o no un tránsito migratorio. Al incluir ambas variables en un mismo modelo, se apunta a que el efecto de la cohorte generacional se encuentra modulado por ser o no Hispanohablante. Se discuten las nuevas perspectivas que estos resultados abren de cara a la potenciación de la comunicación intercultural y un proyecto de convivencia común.

KEYWORDS:

• Language attitudes
• Spanish-speaking – non-Spanish-speaking
• generational cohort
• Catalonia (Spain)
• young immigrants

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Cecilio Lapresta-Rey is Lecturer in Sociology at the Department of Geography and Sociology of the University of Lleida (Spain). His research focuses on language and identity, language attitudes and integration in the case of children of immigrants in Catalonia. Currently, his work focuses on the analysis of the integration processes of these youngsters.

Ángel Huguet is Professor in Developmental and Educational Psychology at the Department of Pedagogy and Psychology of the University of Lleida (Spain). His research interest includes bilingual education, second-language acquisition, psycholinguistics and sociolinguistics.

Alberto Fernández-Costales is Lecturer in Applied Linguistics at the Department of Educational Sciences of the University of Oviedo. Among other fields, his research interests lie on language teaching, multilingualism, language attitudes and the role of English as a lingua franca.

ORCID

Cecilio Lapresta-Rey http://orcid.org/0000-0002-3411-7077

Notes

1. Warner and Srole (1945) differentiate between immigrants born overseas (P generation) and born in the host society – the United States in this case – (F generation). In turn, among the first, P1 would be those who had arrived at an age older than 18, and P2 those who arrived at an age younger than 18. In the case of those born in the host society, they differentiate between the children from immigrant parents born in the host society (F1) and the grandchildren (F2).

2. Delving deeper into the analytical use of cohorts, Waters and Jiménez (2005) stress that these allow adjustment of the generational study, since the individuals from different generations, but from a similar generational cohort, have similar experiences in their host societies. This is due to the fact that they may have similar experiences in belonging to the same generational cohort. This also allows the overcoming of a ‘supposed’ sequential time conception since, as stated above, each generation has different cohorts and each cohort exists in different generations. Complementarily, this allows segregation according to one of the elements that are repeatedly shown to be significant in the study of experiences and outcomes of the descendants of immigrants, i.e. which is being are born or not in the host society and having one or both parents immigrant which have one or both immigrant parents (generational cohorts 2.0 and 2.5).

3. A GLM is not a correlational technique, and it is not necessarily predictive. This type of model analyses the joint effect of two variables (usually nominal variables) as regards the differences in the mean of a third variable (the dependent variable). Although the nominal variables can be analysed separately, the study of their impact is more reliable when considered jointly. In this sense, the interaction of both variables is considered taking into account the fact that it might change the results of the individual analysis (Field, 2012).

4. Only in this case has the student’s t-test been used for mean comparison, since the mean values are compared for the same group in two different variables.

Additional information

Funding

This work was supported by the Ministry of Economy and Competitiveness of the Government of Spain [grant number EDU2014-54093-R] and Government of Catalonia – Agència de Gestió d’ajuts Universitaris i de Recerca (AGAUR) [grant number 2014 SGR 208].