Search results

We study if the language of administration of a survey has an effect on the answers of bilingual respondents to questions measuring political dimensions. This is done in two steps. In the first we test whether the measurement instruments are equivalent for the same individual in two languages. After measurement invariance is established, we test if latent mean differences are significant across the two languages. We also test if the correlation of the same concept in two languages is equal to one or not. Results show evidence for language effects, the latent correlation is below one, although mean differences are not significant. We use data of the LISS migration panel in a within subject design, respondents answer a questionnaire twice first in Dutch and then in their (second) native language among Arabic, English, German, Papiamento and Turkish.

The present dissertation explores language effects in a comparative survey i.e. to what extent linguistic diversity affects equivalence in a comparative survey. This is done by studying three different dimensions on the challenges of designing a comparative multilingual survey: survey translation, linguistically diverse countries and bilingualism. Guidelines in survey translation do not link assessment criteria and measurement equivalence testing. I propose a systematic procedure to compare versions of a question in different languages before fieldwork which establishes that link. In linguistically diverse countries, survey instruments are translated into more than one language, equivalence is commonly assumed, not tested. I test for invariance distinguishing the response and cognitive processes to a survey question. Finally, I study measurement equivalence within an individual in two languages for political constructs (bilingualism), challenging current methodological approaches by bringing latent variable models. In each dimension, findings aim to contribute to improving comparative survey methodology. ; Esta tesis explora los efectos del lenguaje en una encuesta comparativa: en qué medida la diversidad lingüística afecta la equivalencia de los datos mediante el estudio de tres dimensiones: la traducción de encuestas, países lingüísticamente diversos y el bilingüismo. Las directrices actuales en la traducción de encuestas no vinculan los criterios de evaluación con un test de equivalencia. Se propone un procedimiento sistemático para comparar las versiones de una pregunta que establece dicho vínculo, en diferentes idiomas antes del trabajo de campo. En países lingüísticamente diversos, el cuestionario se traduce en más de un idioma. Se realiza un test de equivalencia que permite distinguir los procesos de respuesta de los cognitivos. Finalmente, se estudia la equivalencia de conceptos políticos en dos idiomas para un individuo (bilingüismo), proponiendo un enfoque metodológico de modelos de variables latentes. ...

AbstractThis work uses a list experiment to directly test for differences in attitudes towards poor, racially different and racially similar immigrants to the Netherlands in terms of (1) overt support and (2) covert support and (3) the difference between the two (i.e., social desirability bias). Results show that the expression of opposition to immigrants is significantly masked, suggesting that anonymous acts (e.g., voting) could capture greater opposition. Racially similar immigrants are subject to greater opposition relative to racially different or poorer immigrants, but only when anonymity is offered to respondents. The implication is that overt estimates (i.e., openly expressed) are misleading as levels of support appear consistent across immigrant groups. Going forward, we suggest the utility of the list experiment to better account for observable downward bias in estimates at least on the aggregate level or in contexts with clear normative pressure to appear tolerant.

Abstract
A highly controlled experimental setting using a sample of questions from the European Social Survey (ESS) and European Values Study (EVS) was used to test the effects of integrating machine translation and post-editing into the Translation, Review, Adjudication, Pretesting, and Documentation (TRAPD) approach in survey translation. Four experiments were conducted in total, two concerning the language pair English-German and two in the language pair English-Russian. The overall results of this study are positive for integrating machine translation and post-editing into the TRAPD process, when translating survey questionnaires. The experiments show evidence that in German and Russian languages and for a sample of ESS and EVS survey questions, the effect of integrating machine translation and post-editing on the quality of the review outputs—with quality understood as texts output with the fewest errors possible—can hardly be distinguished from the quality that derives from the setting with human translations only.