Stereotyped Reactions to Speakers with Scottish and English Regional Accents

Abstract

Experimental studies of judgements of personality of speakers suggest that judgements based on vocal characteristics alone tend to be consistent but inaccurate. It appears that when making judgements of speakers with minimal cues, judges tend to draw on stereotypes. Lambert et ul. (1960) found that bilingual Canadian speakers were rated differently depending on whether they spoke French or English, and these differences depended on the languages spoken by the judges. Since the different voices were the same speakers, the differences in judgements are unrelated to real characteristics of the speakers, but must be due to differences The present study uses the matched-guise technique to compare ratings of voices with Scottish and English regional accents.Tape-recordings were made of four drama students (2 male, z female), each reading a standard English passage once with an English and once with a Scottish regional accent (all accepted as native speakers by Scottish and English subjects in a pilot study), and of four other professional-class speakers (male Scottish, male English, female Scottish, female English) reading with their normal voices. These 12 voices were arranged in one order counterbalancing sex and region.There were 1x0 Scottish-born subjects (57 males, 53 female) in Glasgow and 59 Englishborn subjects (26 male, 33 female) in London, consisting of groups of teachers, nurses, businessmen, Young Conservatives and adult further education classes.Subjects were told that the experiment was concerned with judgement of personality from voices, and regional accent was not mentioned. The tape was played and, as each voice spoke, subjects rated it on 21 six-point scales-the 20 used by Lambert et al. (1965) plus 'occupational status'. Each subject finally gave his own age, occupation and place of birth.The 12 voices were paired off appropriately Scottish with English, and a profile of 21 difference scores for each of the six pairs of voices was found for each subject. Differences were analysed by t-tests separately for each scale, each pair of voices and for Scottish and English subjects. Since this involved 252 t-tests, only 0.01 levels of significance or better were considered in view of the high probability of obtaining spurious 0.05 significance values. Table I summarizes the scales for which consistent significant differences occurred between Scottish/English pairs of voices.With male speakers, both Scottish and English subjects rated Scottish accented voices lower than English on several scales which appear to be mainly concerned with status. Scottish subjects also rated Scottish voices higher on several scales suggesting a warmer personality, but with English subjects the only one of these scales showing significant differences was ' friendliness'. With female speakers, differences generally tended to be smaller and occurred for fewer scales, particularly for English subjects. Again English accented voices were rated higher than Scottish on some 'status' scales, but there were no scales where Scottish fem...

An Exploratory Experimental Study of the Influence of Patients’ Social Background upon Diagnostic Process and Outcome

“…Studies of the role of language in middle class and working class cultures (see for example Newson and Newson, 1976) document the high value placed upon verbal fluency in the middle class culture, often at the expense of verisimilitude. Previous research using the matched-guise technique has shown non-RP speakers to be seen as significantly more trustworthy than RP speakers (Strongman and IVoosly, 1967;Cheyne, 1970;Giles, 1970). This result is replicated in the present study: scores on the 'honest-dishonest' scale were significantly higher for the RR speaker (t = -1.85, p < 0.05, 1 tailed).…”

Social representations, social attribution and social identity: The intergroup images of ‘public’ and ‘comprehensive’ schoolboys

“…Here, it is sufficient to note that such 'over-estimation' by the lower status group has been found in several studies (e.g. Anisfeld, Bog0 and Lambert, 1962;Cheyne, 1970;Lambert, Hodgson, Gardner and Fillenbaum, 1960). Jaspars and Warnaen (1982) argue that such groups share a common social representation (based on the 'objective' relationship between both groups), but that they use different comparison levels to make social judgements.…”

The competence and warmth of Thai students’ attitudes towards varieties of English: the effect of gender and perceptions of L1 diversity

“…In particular, it has been demonstrated that speakers of varieties perceived as standard tend to be rated highly in terms of status (traits such as intelligence and education) whereas speakers of varieties perceived as non-standard are generally evaluated more favourably in terms of solidarity (traits such as honesty and friendliness). The distinctions made between perceptions of standard and non-standard forms of spoken English have been demonstrated in a large number of studies in L1 Englishspeaking countries, whether through the presentation of speech samples or involving other sociolinguistic instruments, including: Australia (Bradley and Bradley, 2001); England (McKenzie, in press); New Zealand (Bayard et al, 2001); Republic of Ireland (Edwards, 1977); Scotland (Cheyne, 1970); the USA (Fought, 2002); Wales (Garrett, 2010); and the UK more widely (Coupland and Bishop, 2007).…”