Article summary with How you pronounce things matters by Rakic a.o. - 2011

When you look at a person, different aspects play a role. Individual characteristics and social categories will play a role in this. Another distinction to be made relates to the sensory domain. Psychologists have concentrated on one of many aspects at a time, when they try to understand person perception. The visual domain has obviously played a major role and because of this, some other sensory domains have been left more in the background. However, people we meet are not just static pictures but they talk and they reveal much about themselves with their speech styles and language. The writers of this text investigated whether regional accents result in more negative evaluations of candidates in a job interview independently of their qualifications.

There has been a lot of progress in understanding face perception since Bruce and Young’s face recognition model. This model states that there are identity specific and unspecific aspects of face recognition. Categorical information, like race and gender, can also be extracted during the face-encoding process and these aspects need to be integrated into a comprehensive person perception model. The categorical approach is more common in social psychology. Of course, in this field, the focus of interest is more on social categories the individual belongs to than on the unique individual. Categorical processing appears to be faster than identity processing and it should occur before individualization. Identity recognition and categorization influence each other. Categorizing a person is likely along primary social categories, like gender, age and ethnicity that are almost automatically activated. In social psychology, it is important to understand the different underlying processes that contribute to the activation of one category over the other. The importance of a given category can vary based on the context and the type of information available. The face reveals information about the person, but the voice also plays an important role.

If we want full understanding of person perception, the voice and speech perception need to be integrated with visual person perception. The face is at the centre of the face recognition model, but the model also includes a part called ‘facial speech analysis.’ This opens the door for speech and voice perception. The face and voice are strongly linked. Also, the face and voice contribute to identifying a familiar speaker’s identity. Speech differences also indicate in which social group somebody belongs (gender and age). Some differences are predominantly due to differences in physiology, but others are cultivated, like the use of lower pitch frequencies of American compared to British speakers. You can infer a variety of information from speech, like personality traits and voice attractiveness can influence the ascription of personality traits. It seems that people automatically appear to use voice and speech to infer information about the speaker. Theories of language attitudes distinguish between standard and non-standard varieties of given language. When a language develops, it will undergo a series of changes and among the many varieties within this language, one will be established as the standard one, while others become non-standard. There is social pressure to use standard language, and it acquires superiority on different dimensions that seem to be accepts and used by those speaking that language. Social status is one attribute associated with different language varieties. Using a regional accent or dialect, may indicate lower social status. In one study, American participants perceived British speakers as more intelligent and of higher social status than American speakers. With the use of the Speech Dialect Attitudinal Scale (SDAS), studies have found that dialect speakers were perceived as having lower socio-intellectual status and minor dynamic qualities compared to standard language speakers. This also goes for regional accents. Accents lead to the perception of speaker group membership, which in turn provokes a judgment of group status and of speaker status.

There is pressure to use standard language, but therefore it seems surprising that non-standard dialects and languages persist. One reason they do persist, is because they are an essential part of one’s (ethnic) identity. There are different dynamics between people based on the type language they use. This is explained with the communication accommodation theory. Different speech strategies can be context-based. You’ll probably speak differently to a friend than during a job-interview. Independently of those dynamics, each language variety (standard or non-standard) results in an evaluation of the speaker. Studies have shown that in general, standard language speakers are evaluated more positively across different dimensions, but these differences also varied based on the context as well as the country of the studies. One study found that people speaking with a non-standard dialect were attributed higher loyalty and integrity. It is somewhat difficult to compare some studied, because some concentrate on dialects and others on regional accents. Dialects comprise different words and grammar compared to standard language, so some differences found between the negative perceptions of people with a dialect were due to incomprehension. Speech styles seem to be irrelevant for manual labour positions, but they play a more important role for management positions. Accents should be unrelated to job performance, but they are salient cues within communication and influence evaluations of others.

The experiments

Much evidence shows that different factors influence the evaluation of speakers differently and the writers of this text wanted to test whether accent alone would be sufficient for discrimination in the context of a job interview. They used regional accents versus standard German accent. The aim of this article was to test whether hirability, competence and socio-intellectual status judgments of speakers depend on their regional accents, given only auditory information.

Experiment 1

The aim of the first experiment was to test the influence of different regional accents on speaker’s perceived competence, hirability and socio-intellectual status. They writers wanted to know whether there were differences in job interview outcome when regional accents are compared to a standard German accent. They writers thought that people speaking standard German would be judged as having the highest competence as well as having the highest chance to be hired. They also thought that regional accent speakers should be assigned lower socio-intellectual status.

In the first study, six different speakers were used, four of which had different accents (one woman and three men) and also a man and a woman who spoke standard German. A pilot-study revealed that participants were unable to tell if a voice belonged to the same or a different person (from the same gender). Participants were first told that they would hear part of a job interview and would have to imagine they were the employer and judge the speaker based on what they heard during the interview. After that, participants were presented with the job announcement and the job interview. Afterwards, participants had to rate traits of the speakers on a Likert-scale. Then, participants were told that they would hear the speaker again shortly and they needed to concentrate on their speech style and indicate their impression about the personality of each speaker. The last thing they had to do is to provide demographic information about themselves.

The results showed that speakers with a standard German accent were perceived as more competent than regional accent speakers. No difference was found between different regional accents. Standard German speakers were also rated as more hireable than regional accent speakers. Standard German speakers were also assigned higher socio-intellectual status compared to regional accent speakers. Bavarian speakers were assigned high socio-intellectual status while the Berlin and Saxon speakers were not. Accents can activate very complex set of stereotypes linked to a group and perhaps socio-intellectual status was linked with the general knowledge that Bavaria is a strong economic region of Germany.

Experiment 2

This second experiment was conducted to exclude any possible confound of individual voice characteristics on the findings. In this study, six speakers with regional accents were used and they had to record six statements (one each). All of them were also recorded saying the same sentence in standard German. This resulted in 12 statements, half spoken in standard German and half with different regional accents. Each participant had to evaluate one set. So, one participant heard only one version of each speaker. The participants were told to evaluate different speakers as if they were recruiters. After hearing each speaker, the participants evaluated the speaker on a Likert-scale.

The results show that speakers with a standard German accent were rated as more competent than regional accent speakers. Also, speakers with standard German accent would be hired with a higher probability than the same speakers with a regional accent. The same voices were used for standard German and for regional accents, and the writers can thus conclude that the differences they found were due to accent type and not any other voice or speech characteristics.

In experiment 2, no specific position description was provided. So it seems that the outcome can’t not only be generalized to management positions, but also other positions. However, future research should look into jobs which prefer regional accents and see whether the findings reverse in such contexts. These two experiments also don’t yield a hint that gender is used as a cue for competence ascriptions.