Measuring language attitudes in context: Exploring the potential of the Personalized Implicit Association Test

Abstract

After decades of relative methodological stagnation, language attitude research is witnessing an influx of new experimental methods originally developed in social psychology. One such measure is the Personalized Implicit Association Test (P-IAT), a reaction-time-based method that measures the association between two concepts. The P-IAT has been used successfully to measure language attitudes, yet presents a number of challenges, like the fact that it measures attitudes void of linguistic or interactional context. This article aims to address that challenge and introduces a contextualized version of the P-IAT, which was used alongside an explicit measurement to explore attitudes towards varieties of Dutch in formal vs. informal settings. While the explicit attitudes show the expected pattern of preference for the standard variety in formal contexts, results from the implicit measurement are not as clear-cut. We discuss potential explanations for these findings and reflect on consequences for future sociolinguistic research using the P-IAT. (Personalized Implicit Association Test (P-IAT), context dependence of language attitudes, sociolinguistics)*

It's time to do more research on the attitude–behavior relation: A commentary on implicit attitude measures

“…This knowledge is largely related to the cultural milieu and social environments people inhabit, a point made early after the development of the IAT (Dambrun et al, 2008; Fazio & Olson, 2003; Olson & Fazio, 2004). In addition to the cultural bases of associations in memory, knowledge about social groups is related to personal experiences, and the activation of any representation varies with context (Olson & Fazio, 2004; Rosseel et al, 2019). Critical to the interpretation of implicit measures, the knowledge people recruit to respond to the IAT has evaluative implications because the measures present unambiguous evaluative concepts such as “good” and “bad.” However, the knowledge activated is not identical to a summary attitude that categorizes the object or social group as “good” or “bad.” Rather, implicit measures are like beliefs, affective reactions, and evaluations about different facets of an object.…”

The influence of ethnicity and language variation on undergraduates' evaluations of Dutch-speaking instructors in Belgium: A contextualized speaker evaluation experiment

“…Belgian Standard Dutch accordingly involves public interaction in a professional, expertise-oriented sphere, while the dialects are associated with personal interaction in the private sphere.' Some researchers have interpreted this division of labour as a sign of the enduring vitality of the standard language ideology (Jaspers and Van Hoof, 2013;Rosseel et al, 2019a). Others have interpreted the public media licensing of Colloquial Dutch and the association of the variety with dynamism as a sign of changes in the traditional standard language ideology (Grondelaers and Speelman, 2013;Grondelaers et al, 2016).…”

“…Flemish experimental research into language attitudes has thus as yet had little eye for the possible impact of contextual and speaker characteristics on speaker assessment (cf. Rosseel et al, 2019a). Almost all speaker evaluation experiments have deployed male speakers, neglecting the potential impact of gender on evaluation outcomes.…”

“…By explicitly or implicitly equating language attitudes to the social meanings or indexicality of speech styles (Grondelaers et al, 2016;Rosseel et al, 2019a,b;Soukup, 2013), and by treating them as indicative for the vitality, or on-going change, of language ideologies, quantitative language attitude studies have gone through a conceptual rapprochement with interpretive and qualitative strands of sociolinguistics and with linguistic anthropology, the field where the notions of language ideology and indexicality were originally developed (Kroskrity, 2010;Silverstein, 1979;Woolard and Schieffelin, 1994). It has also been noted, however, that experimental methodologies sit uncomfortably with the contextualized nature of social meaning that is foregrounded in these fields (Potter & Wetherell, 1987;Soukup, 2013;Rosseel et al, 2019a). Indeed, central there is the observation that the meanings of linguistic variables that mark standard and vernacular varieties 'are not precise or fixed but rather constitute a field of potential meaningsan indexical field, or constellation of ideologically related meanings, any one of which can be activated in the situated use of the variable' (Eckert, 2008).…”

Language Attitudes in a Lingua Franca: The Case of Black South African College Students

“…From a methodological viewpoint, this study also seeks to advance in the investigation of language attitudes in South Africa by moving away from direct elicitation of language attitudes (Formanowicz & Suitner, 2020;Paladino & Mazzurega, 2020;Rosseel et al, 2019). More precisely, we will use an audio Implicit Association Test (IAT), that is, an adapted version of IAT (Greenwald et al, 1998) that incorporates clips uttered in the selected language varieties to measure participants' implicit bias toward them (Hypothesis 1).…”