No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Is Your Accent Right for the Job? A Meta-Analysis on Accent Bias in Hiring Decisions
Abstract
Standard-accented job candidates are perceived as more hireable than non-standard-accented candidates. Two broad perspectives have emerged as to what drives this effect: (a) that it is a pragmatic response to the perception that non-standard accents can impede job-relevant communication (processing fluency explanation) and/or (b) that non-standard accents signal “otherness” and candidates are devalued as a result (prejudice explanation). This meta-analytic integration of 139 effect sizes ( N = 4,576) examined these two perspectives. Standard-accented candidates were considered more hireable than non-standard-accented candidates ( d = 0.47)—a bias that was stronger for high communication jobs. Other findings, however, are difficult to explain from a processing fluency explanation: candidates’ relative comprehensibility was not a significant moderator of hiring bias. Moreover, the degree of accent bias was associated with perceptions of the candidates’ social status, and accent bias was particularly pronounced among female candidates and for candidates who spoke in foreign (as compared with regional) accents.
... Prior meta-analytic investigations of accent bias in hiring interviews provided us with the foundation to expect that SA applicants may be seen as more hirable than NSA applicants (Fuertes et al. 2011;Schulte et al. 2024;Spence et al. 2022). Although these meta-analyses had different datasets and research questions, they also highlighted some of the reasons why accent bias may occur, with moderators of accent bias including applicant gender (Spence et al. 2022), interviewer judgments of speaker attributes, including competence, warmth (Fuertes et al. 2011;Schulte et al. 2024;Spence et al. 2022), and dynamism (Spence et al. 2022), as well as communicative factors like type of accent or language dialect (Fuertes et al. 2011;Schulte et al. 2024;Spence et al. 2022), and comprehensibility (Spence et al. 2022). ...
... Prior meta-analytic investigations of accent bias in hiring interviews provided us with the foundation to expect that SA applicants may be seen as more hirable than NSA applicants (Fuertes et al. 2011;Schulte et al. 2024;Spence et al. 2022). Although these meta-analyses had different datasets and research questions, they also highlighted some of the reasons why accent bias may occur, with moderators of accent bias including applicant gender (Spence et al. 2022), interviewer judgments of speaker attributes, including competence, warmth (Fuertes et al. 2011;Schulte et al. 2024;Spence et al. 2022), and dynamism (Spence et al. 2022), as well as communicative factors like type of accent or language dialect (Fuertes et al. 2011;Schulte et al. 2024;Spence et al. 2022), and comprehensibility (Spence et al. 2022). ...
... Prior meta-analytic investigations of accent bias in hiring interviews provided us with the foundation to expect that SA applicants may be seen as more hirable than NSA applicants (Fuertes et al. 2011;Schulte et al. 2024;Spence et al. 2022). Although these meta-analyses had different datasets and research questions, they also highlighted some of the reasons why accent bias may occur, with moderators of accent bias including applicant gender (Spence et al. 2022), interviewer judgments of speaker attributes, including competence, warmth (Fuertes et al. 2011;Schulte et al. 2024;Spence et al. 2022), and dynamism (Spence et al. 2022), as well as communicative factors like type of accent or language dialect (Fuertes et al. 2011;Schulte et al. 2024;Spence et al. 2022), and comprehensibility (Spence et al. 2022). ...
To address concerns of subtle discrimination against stigmatized groups, we meta‐analyze the magnitude and moderators of bias against non‐standard accents in employment interview evaluations. Results from a multi‐level random‐effects meta‐analysis (unique effects: k = 41, N = 7,596; multi‐level effects accounting for dependencies: k = 120, N = 20,873) demonstrate that standard‐accented (SA) interviewees are consistently favored over non‐standard‐accented (NSA) interviewees ( d = 0.46). Accent bias is stronger against women compared to men, particularly when evaluator samples are predominantly female, and was strongly predicted by interviewers' stereotypes of NSA interviewees as less competent and, to a lesser extent, as less warm. Accent bias was not significantly impacted by perceptions of comprehensibility, accentedness, accent type, interview modality, study rigor, or job speaking skill requirements.
... We investigate here whether these effects replicate and extend to the context of narrative storytelling in the written form. Given that prejudicial racial attitudes have been observed in a variety of contexts, from foreign-accented job candidates being deemed less hirable (Spence et al., 2024) to foreign-accented characters receiving more negative evaluations (Romero-Rivas et al., 2022), we predict that: H2: Processing fluency will be negatively associated with a negative out-group bias. H3: There will be an indirect positive relationship between the presence of codemixing and an out-group bias through processing fluency. ...
... The first takeaway is that the presence of code-mixing appears to impact narrative processing in ways above and beyond the inclusion of a cross-cultural character alone. Guided by previous work on how people process foreign languages (Avarino et al., 2021;Dragojevic & Giles, 2016;Dragojevic & Goatley-Soan, 2022;Foucart et al., 2020;Gasiorek & Dragojevic, 2023b;Romero-Rivas et al., 2022;Spence et al., 2024), we examined whether code-mixing would produce negative out-group evaluations due to impairments in processing fluency. Although previous work has generally examined the relationship between processing fluency and language attitudes using promotional materials (Gasiorek & Dragojevic, 2024), organizational materials (Gasiorek & Dragojevic, 2023b), or foreign accents (Dragojevic, 2020), less work has examined these relationships within the context of reading a story. ...
This experiment (N = 1,241) investigates the impact of code-mixing, defined as the use of more than one language, on processing fluency, narrative engagement, and cross-cultural attitudes. Using a sample of native English speakers located in the United States, we found that narratives that include code-mixing, a common feature of intercultural communication, felt more difficult to process and, in turn, led to more negative out-group bias and less narrative engagement. These findings integrate and extend intercultural communication and narrative theory and consider the challenges and opportunities that accompany diverse representations of characters in storytelling. Rather than highlight these challenges, however, we consider theoretically based strategies to improve audiences’ reception to cross cultural content and, in doing so, hope to inform communication practices that lead to a greater regard for others.
... These biased evaluations can substantially impact the professional integration of L2-accented speakers. In fact, recent meta-analytic evidence (Spence et al. 2024) shows that L2-accented speakers are systematically perceived as less employable than L1-speaking Downloaded from https://academic.oup.com/applij/advance-article/doi/10.1093/applin/amae042/7712341 by guest on 11 July 2024 candidates. Adding nuance to this pattern, research further suggests that occupational characteristics can affect professional-related evaluations assigned to L2-accented speakers, who are often preferred for low-prestige positions (e.g. ...
... Kalin and Rayko 1978;Singer and Eder 1989) and those with lower communication demands (e.g. Lindberg and Trofimovich 2023;Spence et al. 2024). Until recently, however, it was unclear whether and how listeners consider a speaker's job performance or skill level when evaluating the speaker, such that accent bias may interact with speakers' demonstrated job-relevant performance/skill level. ...
Evidence from Canada suggests that accent bias can be moderated by speakers' demonstrated job-relevant performance and the prestige level of their occupation (Teló et al. 2022). In this study, we replicated Teló et al.'s (2022) work in Brazil. First language (L1) Brazilian Portuguese-speaking listeners rated audio recordings of L1 Brazilian Portuguese and L1 Spanish speakers along continua capturing one professional (competence), one experiential (treatment preference), and one linguistic (comprehensibility) dimension. Our findings challenge the notion of consistent bias, as listeners did not uniformly perceive L1 Brazilian Portuguese speakers as more competent and comprehensible than L1 Spanish speakers, and, in fact, generally preferred treatment provided by L1 Spanish speakers. Complex interactions provided a nuanced account of listeners' evaluations, revealing, among other patterns, that demonstrated performance level and job prestige affected the evaluated dimensions differently depending on the speaker's L1. This replication further expands the initial study by examining the role of four listener variables as predictors of speaker ratings. Greater listener familiarity with the context depicted in the script was associated with the assignment of higher ratings overall.
... This can then have far-reaching implications in various social situations, including employment interviews (Kalin, 1982;Matsuda, 1991;Munro, 2003;Purnell et al., 1999). In fact, it is well documented that a speaker's accent-or the degree to which speech is colored by various ethnic, regional, or L2 features-can lead to job applicants being evaluated less favorably (Brennan & Brennan, 1981;Ryan et al., 1977;Spence et al., 2022), with L2 speakers especially vulnerable to accent bias (Carlson & McHenry, 2006;Hopper & Williams, 1973;Rakić et al., 2011). For instance, in a recent study, applicants who spoke English with a strong L2 accent during a mock job interview received the lowest employability ratings, even though their English skills were irrelevant to the position, which was to teach German (Roessel et al., 2019). ...
... Put differently, whereas an L2-speaking candidate might be as suitable as an L1 French speaker for the position of a pizza cook, which is a low-status job requiring little communication, an L2 speaker was clearly less acceptable as a waitress, which corresponds to a low-status job requiring considerable French communication skills. This explanation would align with results of a recent metaanalysis of 27 studies which revealed a hiring bias favoring standard-accented speakers only for jobs with high communication demands (Spence et al., 2022). Alternatively, as discussed above, in light of a post hoc analysis of response quality, the L2 French cook may have been upgraded in her final evaluation because, unlike the L2 French waitress, she provided more sophisticated answers in her second interview responses. ...
According to expectation violation theory, job applicants can be upgraded or downgraded during an interview when their accent does not match employers’ speech expectations. Focusing on the employment of second language French job candidates in Québec, this study explored this issue dynamically in terms of how expectations may impact the trajectory of interview evaluations. Participants included 60 Québec French raters and 6 female job candidates applying to a waitress or pizza cook position, presented through their resumes as either first (L1) or second (L2) language French speakers. Each speaker’s interview audios were presented to raters in expectancy-congruent and expectancy-incongruent scenarios. Raters first provided resume-based employability assessments, then two more evaluations throughout a typical sequence of interview questions. The congruent and incongruent scenarios revealed similar evaluation patterns, where the L2 French cook applicant’s employability improved after initially being downgraded. Implications are discussed regarding listeners’ readjustment of their perceptions following first-impression biases.
... For example, even in societies with strong anti-racism norms, holding negative sentiments regarding individuals who speak with a foreign accent (e.g., "they don't speak English well") is often considered acceptable, so much so that overt discriminatory behaviors based on these prejudicial attitudes are still prevalent in our everyday lives (Gluszek & Dovidio, 2010). Furthermore, prejudicial attitudes which can be attributed to constructs that are not protected by anti-prejudice norms may be used to justify discriminatory behaviors against social groups that are characterized by related constructs (e.g., foreign accent being used to justify not hiring non-White job candidates; Spence et al., 2022). To understand the psychological processes pertaining to prejudice more broadly, therefore, it is essential that we expand our focus beyond the specific social groups for which norms against prejudice are generally afforded. ...
... Furthermore, children in societies where anti-racism norms are prevalent appear to actively suppress verbal remarks and behavioral displays of race-based biases by around middle childhood (Raabe & Beelmann, 2011). But perhaps due to the lack of social norms against accent-based prejudice, explicit expressions of accent-based biases continue to manifest across the lifespan (Kinzler, 2020;Spence et al., 2021Spence et al., , 2022. To understand how racism continues to be expressed in societies where anti-racism norms are wellestablished, therefore, it is essential that we investigate the anti-prejudice norms for constructs that are intricately related to the construct of "race"-in our study, we do this by examining accent. ...
By middle childhood, children become aware that discriminatory behavior is unacceptable; however, the development of their anti‐prejudice sentiments is largely unknown. Across two studies, 333 Australian 5‐ to 10‐year‐olds (51% female, majority White) were asked how acceptable they thought it was to have prejudicial sentiments toward 25 different targets. Children responded privately through a novel digital paradigm designed to minimize social‐desirability biases. With age, children were more likely to display anti‐prejudice sentiments toward targets who are prosocial, vulnerable, and of minority race and linguistic backgrounds. In contrast, they judged prejudice as “okay” for targets who are antisocial and negatively regarded in society. These findings suggest that children's perceptions of prejudice become increasingly nuanced and adult‐like across the primary school years.
... Listeners readily react to differences in accented L2 speakers' pronunciation, often judging people who speak with L2 accents to be less competent, successful, likeable, intelligent, and even less physically attractive (Campbell-Kibler, 2007;Hosoda et al., 2007;Seligman et al., 1972;Teló et al., 2022). With respect to professional communication, which is the focus of this study, Spence et al. (2022) recently reported a large aggregate negative effect of L2 accent on hiring decision in a meta-analysis of 27 studies containing 139 individual effect sizes. In particular, Spence et al. found that accented L2 speakers are particularly downgraded in hiring decisions relative to speakers of standard varieties, whereas speakers of regional varieties of the same language do not carry this penalty. ...
... Considering future work, accent bias-as a complex, multidimensional phenomenonrequires a nuanced treatment. Whereas accented L2 speakers should not be denied employment opportunities simply because they speak with an accent (Hideg et al., 2022;Spence et al., 2022), an employee's linguistic competence is certainly a consideration in many workplace settings. As explained by a university HR instructor interviewed by O'Brien et al. (2022), accent may play little role when recruiting "the best programmers in the world," but as an umbrella term encompassing a person's L2 speaking ability, it might be a key consideration for communication-focused and customer-facing jobs. ...
For many second language (L2) speakers, including immigrants, speaking with an L2 accent can be a source of unfair or biased treatment in many workplace contexts. However, apart from research on language learners, there is currently little knowledge as to what the general public, and especially members of professional communities, know about accent and accent bias. Our goal in this study was to examine the intuitive understanding of accent and accent bias by university students in human resource (HR) management as future gatekeepers to gainful employment. We interviewed 14 students across two four-year university HR programs in Canada asking the students about their prior experience with accent bias and exploring their understanding of the broader construct of accent through thematic interview coding. The students reported multiple examples of accent bias, demonstrating a nuanced understanding of accent, where they characterized accent bias as an unconscious phenomenon, highlighted its experiential component, expressed sensitivity to different linguistic sources of accent, emphasized the role of a listener in L2 communication, and generally showed flexibility and tolerance toward accented L2 speech. We discuss these findings in light of prior work on accent awareness and highlight the importance of dedicated accent-focused training for HR professionals.
... Early research found British perceivers to prefer an RP-accented speaker (perceived higher class) for higher-status occupations and a Welsh-accented speaker (perceived lower class) for lower-status occupations (Giles et al. 1981). More recent work found less favourable evaluations of lower-class-sounding speakers for an entry-level law firm job in the United Kingdom (though factors such as age moderated this; Levon et al. 2021) and an unspecified position in the United States (Kraus et al. 2019), and a recent meta-analysis found that speakers with standard (i.e., higher-class) accents are perceived as more employable than those with non-standard accents (Spence et al. 2024). ...
Social class profoundly affects people in a wide range of ways, leaving its mark on individuals’ behaviour, cognition, experiences and outcomes. Social class also plays a role in social perception, and, in this review, I provide a snapshot of the current body of knowledge related to the social perception of social class. Specifically, I review research examining accuracy and bias in social‐class perceptions from nonverbal cues, in addition to the consequences of these perceptions. Altogether, perceivers show some accuracy in judging individuals’ social class from various nonverbal cues, indicating that nonverbal behaviour provides signal to individuals’ social class and that perceivers can detect this. A large body of evidence simultaneously highlights substantial bias in social‐class perceptions, with social‐class stereotypes meaningfully affecting impressions. Finally, research illustrates the broad range of consequences of social‐class perceptions, perhaps most importantly in the domain of competence and hiring judgments, which can serve to perpetuate inequality.
... When speaking a new language in day-to-day life outside the classroom, ALSs can face several challenges, including accent-induced anxiety (Baquiran & Nicoladis, 2020;Bresnahan, Ohashi, Nebashi, Liu & Morinaga Shearman, 2002;Chakraborty, 2017;Kristiansen, 2001;Park, Hodge & Klieve, 2022;Romero-Rivas, Morgan & Collier, 2022;Spence, Hornsey, Stephenson, & Imuta, 2024). This anxiety often stems from deeply held beliefs about a society's acceptable linguistics standards with regard to accents. ...
Additional language speakers (ALSs) often experience anxiety due to challenges posed by their nonstandard pronunciation. Building on these insights, this paper introduces an instrument, the Accent Anxiety Scale (AAS), specifically designed to assess three sources of anxiety that are experienced by ALSs, including (a) apprehension about negative evaluations from other individuals due to their distinctive speech style, (b) concerns about rejection from the target language community because of their “foreign” pronunciation, and (c) anxieties over potential communication hurdles attributed to the intelligibility of their pronunciation. We evaluated the psychometric robustness of the AAS by analyzing data from a total of 474 immigrant and international student ALSs at a predominantly English-speaking Canadian university. Study 1 focused on immigrants ( N = 203) and employed exploratory factor and correlational analyses to isolate a concise number of internally consistent and valid items for each subscale. Study 2 extended these analyses to international students ( N = 153) and employed confirmatory factor and correlation analyses to further validate the AAS in this population. Study 3 examined international students ( N = 118) at two time points to establish the AAS’s temporal stability. These studies yielded robust psychometric evidence for the factor structure, reliability, and validity of the AAS. The findings not only support the use of the AAS as a research instrument but also offer implications for pedagogical strategies aimed at alleviating ALSs’ accent anxiety.
... There is a rich history of bias in general recruiting against a broad spectrum of attributes, including all protected characteristics, which have been extensively studied and summarised elsewhere [6][7][8]. ...
Artificial Intelligence (AI) assists recruiting and job searching. Such systems can be biased against certain characteristics. This results in potential misrepresentations and consequent inequalities related to people with mental health disorders. Hence occupational and mental health bias in existing Natural Language Processing (NLP) models used in recruiting and job hunting must be assessed. We examined occupational bias against mental health disorders in NLP models through relationships between occupations, employability, and psychiatric diagnoses. We investigated Word2Vec and GloVe embedding algorithms through analogy questions and graphical representation of cosine similarities. Word2Vec embeddings exhibit minor bias against mental health disorders when asked analogies regarding employability attributes and no evidence of bias when asked analogies regarding high earning jobs. GloVe embeddings view common mental health disorders such as depression less healthy and less employable than severe mental health disorders and most physical health conditions. Overall, physical, and psychiatric disorders are seen as similarly healthy and employable. Both algorithms appear to be safe for use in downstream task without major repercussions. Further research is needed to confirm this. This project was funded by the London Interdisciplinary Social Science Doctoral Training Programme (LISS-DTP). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
... At the same time, they may have a strong sentimental connection to Hindi, speaking it at home or with close friends (affective component). Attitudes have been documented mainly in terms of their 'valence (positive or negative evaluations) and intensity (with which an individual holds an attitude)' (McKenzie and McNeill 2023, p 15). Language attitudes determine several important aspects of an individual's life, for instance, hiring (Spence et al. 2024), getting an accommodation (Purnell et al. 1999), judicial trials (Romero-Rivas et al. 2021), and so on. They take on added significance, as in addition to affecting language use, language variation, and change (Garrett 2010;Kircher and Zipp 2022) they also influence language maintenance, as whether to transfer the language to the next generation or not depends on individual perceptions of that language (Sallabank 2013;Kircher et al. 2022). ...
The dominance of one powerful, ‘killer’ language, particularly in formerly colonized countries, has been studied mainly in the context of language education in a multilingual scenario. However, attitudes towards linguistic hegemony remain largely unexplored, an issue that takes on added significance in India, a country known for its exceptional linguistic diversity. The ethnographic study attempts to bridge the gap by analyzing the implicit and explicit language attitudes toward English and Hindi language hegemony among Indians representing a diverse spectrum of linguistic, geographical, and socioeconomic backgrounds. Explicit attitudes, consciously held and reported perceptions, were investigated through one-on-one and group interviews with 60 participants while implicit attitudes, unconsciously held beliefs, were explored through field observations. The study also examined whether English is still viewed as a remnant of colonialism in India. The findings of the thematic analysis reveal that English has largely shed its colonial associations, with socioeconomic status, and the medium of instruction significantly influencing these attitudes. With its exploration of Hindi hegemony in addition to English, the research offers fresh insights into the evolving linguistic landscape of India and contributes to a more nuanced understanding of language dynamics in a post-colonial, globalized society.
... For example, women-far more frequently than men-are criticized for their accent or their voice. A recent meta-analysis showed that women, but not men, were affected by non-standard accent bias such that women were perceived as less hirable (Spence et al., 2022). Another study found that when young women speak with a vocal fry (low pitched and creaky sounding), they are considered less competent, educated, trustworthy, and hirable (Anderson et al., 2014). ...
While much bias and discrimination research has focused on individual identity analysis, for example, gender, race, or ethnicity, what is not fully understood is how other social categories of difference influence discrimination. Using qualitative content analysis, this study examines facets of women leaders' identity that influence their experiences of bias. Respondents were 913 women in four United States industries in which top leadership roles are male dominated but the industries overall are female dominated or gender balanced. The findings revealed 30 distinct identity factors that were used as the basis for discrimination. The variety of the factors and the variation of stigma within each factor show that women leaders are considered “never quite right” as almost any facet of their identity can be declared problematic for a specific woman in a specific context. Human resource leaders should create inclusive policies so that women can express their authentic selves at work. Leadership teams and supervisors should be trained to recognize identity factors that may be the focus of criticism and, thus, serve as veiled bases for discrimination.
... Language-based stigma tends to be examined by broadly focusing on the negative effect of non-standard accents on others' judgments (e.g., Fuertes et al., 2012;Lippi-Green, 2012;Spence et al., 2024). However, not all nonstandard accents are perceived the same nor are all people with a non-standard accent alike. ...
... However, assessor evaluations may be biased by irrelevant factors such as applicant race (Quillian et al., 2017), gender (Schaerer et al., 2023), attractiveness (Hosoda et al., 2003), and accents (Spence et al., 2024). ...
Machine learning (ML) models are increasingly used for personnel assessment and selection (e.g., resume screeners, automatically scored interviews). However, concerns have been raised throughout society that ML assessments may be biased and perpetuate or exacerbate inequality. Although organizational researchers have begun investigating ML assessments from traditional psychometric and legal perspectives, there is a need to understand, clarify, and integrate fairness operationalizations and algorithmic bias mitigation methods from the computer science, data science, and organizational research literatures. We present a four-stage model of developing ML assessments and applying bias mitigation methods, including 1) generating the training data, 2) training the model, 3) testing the model, and 4) deploying the model. When introducing the four-stage model, we describe potential sources of bias and unfairness at each stage. Then, we systematically review definitions and operationalizations of algorithmic bias, legal requirements governing personnel selection from the United States and Europe, and research on algorithmic bias mitigation across multiple domains and integrate these findings into our framework. Our review provides insights for both research and practice by elucidating possible mechanisms of algorithmic bias while identifying which bias mitigation methods are legal and effective. This integrative framework also reveals gaps in the knowledge of algorithmic bias mitigation that should be addressed by future collaborative research between organizational researchers, computer scientists, and data scientists. We provide recommendations for developing and deploying ML assessments, as well as recommendations for future research into algorithmic bias and fairness.
... Effortless comprehension of speech leads to more positive evaluations of the speaker in terms of effect, status, and solidarity (Dragojevic & Giles, 2016). Conversely, when speech is harder to understand, speakers will be perceived more negatively (e.g., foreign compared to native accents with or without background noise; Spence et al., 2024). Most of the processing fluency literature concentrates on foreign-accented speech compared to native speech. ...
Accents provide information about a speaker's geographical, socio-economic, and ethnic background. An additional important variable in the social evaluation of accented speech is the listener's own accent. Just as with any in-group marker, there is a preference for accented speakers that sound like us. Here we employed an auditory version of the implicit association test to quantify own-accent bias. At a Welsh university, we recruited two groups of participants born and raised in distinct regions within the UK, Wales and England. These regions have a long-standing history of national rivalry. In Experiments 1 and 2 we show that the magnitude of the implicitly measured own-accent bias in both groups was comparable to biases based on visible group membership (e.g., race). In addition, Experiment 2 shows that this implicitly measured bias was large compared to the explicitly reported preference. The effect sizes of the in-group preference reported here may have societal impact.
... However, assessor evaluations may be biased by irrelevant factors such as applicant race (Quillian et al., 2017), gender (Schaerer et al., 2023), attractiveness (Hosoda et al., 2003), and accents (Spence et al., 2024). ...
Machine learning (ML) models are increasingly used for personnel assessment and selection (e.g., resume screeners, automatically scored interviews). However, concerns have been raised throughout society that ML assessments may be biased and perpetuate or exacerbate inequality. Although organizational researchers have begun investigating ML assessments from traditional psychometric and legal perspectives, there is a need to understand, clarify, and integrate fairness operationalizations and algorithmic bias mitigation methods from the computer science, data science, and organizational research literature. We present a four-stage model of developing ML assessments and applying bias mitigation methods, including (a) generating the training data, (b) training the model, (c) testing the model, and (d) deploying the model. When introducing the four-stage model, we describe potential sources of bias and unfairness at each stage. Then, we systematically review definitions and operationalizations of algorithmic bias, legal requirements governing personnel selection from the United States and Europe, and research on algorithmic bias mitigation across multiple domains and integrate these findings into our framework. Our review provides insights for both research and practice by elucidating possible mechanisms of algorithmic bias while identifying which bias mitigation methods are legal and effective. This integrative framework also reveals gaps in the knowledge of algorithmic bias mitigation that should be addressed by future collaborative research between organizational researchers, computer scientists, and data scientists. We provide recommendations for developing and deploying ML assessments, as well as recommendations for future research into algorithmic bias and fairness.
... This may be because information gleaned via direct experience with applicants is likely to be more media rich than alternative sources of information (Lengel & Daft, 1984). In the interview, interviewers can not only listen to what the applicant says, but also notice how they say it (e.g., with an accent; Spence et al., 2024), how they look (e.g., physical attractiveness, dress/attire; Barrick et al., 2009;Chang & Cortina, 2023), and even how they smell (Baron, 1986). Interviews involve personal interactions between interviewers and applicants, with opportunities for two-way communication and immediate feedback. ...
The current study examines the riskiest forms and sources of potential bias in the employment interview. A mixed methods survey focused on interviewers' attention to various potentially biasing applicant characteristics, how interviewers learn about such characteristics, the traits of interviewers who are sensitive to such characteristics, and how knowledge of such characteristics affects interviewers' opinions of applicants. Participants were 680 professional interviewers from the U.S., U.K., and Canada. High risks of bias were associated with six applicant characteristics (as targets of bias), three interviewer characteristics (as predisposing toward bias), and three sources of information in the interview process. Interviewers commonly rationalized their judgments in job‐relevant terms. These results have implications for understanding and limiting the risk of common selection biases.
... Their findings suggest that language plays a key role in the sense that less proficient immigrants encounter greater discrimination. More recent studies (Schmaus & Kristen, 2022;Spence et al., 2022), on the contrary, have determined that accent as opposed to linguistic competence or even a foreign surname plays a fundamental role in discrimination. Accents can thus be linked to a negative categorization and evaluation of immigrant workers. ...
The workplace is currently one of the main places of discrimination for socially vulnerable groups such as immigrant workers, who are often required to take on highly stigmatized, menial jobs under supervisors who subject them to daily mistreatment and racism. This study adopted a qualitative approach to 42 semi‐structured interviews of Ecuadorian immigrant workers residing in Spain to explore the processes of discrimination these laborers feel in their everyday workplaces. The findings clearly indicate that immigrant workers can be victims of daily discrimination, which is evidenced by the higher degree of scrutiny and lower levels of trust they suffer compared to their Spanish counterparts, and by their supervisors’ lack of compliance with contractual agreements. As these immigrants are obliged to take on less qualified jobs, they suffer from a lack of recognition and a sense of being undervalued. This analysis also gathered evidence of interviewees’ daily humiliations imparted by their supervisors—and even, at times, by work colleagues—in the form of racial slurs, verbal abuse, and unequal treatment, leaving them feeling powerless and helpless. Most of our respondents in fact find themselves in a predicament they do not know how to confront and cannot reject. All of these factors lead to feelings of humiliation and lack of independence.
This study sought to reshape accent prejudice through an awareness course and to explore course effects on learners’ willingness to communicate and their attitudes toward the social attractiveness of Indian English as a variety of World Englishes. Ten learners with strong accent prejudice were purposefully selected for the treatment. Pre- and post-treatment questionnaires and retrospective interviews showed that the course was significantly effective in lowering accent prejudice and reshaped learners’ perceptions toward a World Englishes variety in terms of social attractiveness and willingness to communicate. These findings imply the criticality of accent exposure and instruction for redressing accent prejudice.
Research has documented significant evidence of workplace bias in the United States, yet less focus has been given to the Latino working population, particularly related to those in high‐paying professional roles, despite the growing presence of Latinos in the US workforce. In this study, we integrate two complementary theories, that is, impression formation and expectancy violation theories, to examine the factors—including the presence of a Latino accent—that may lead to Latino bias during the personnel selection process of an information technology manager. We test our hypotheses using two separate studies. Results from an experimental study (Study 1; N = 458) suggest that managers make decisions based on the candidates' ethnicity during the evaluation of candidates' resumes and interviews, and we find evidence of bias against Latino candidates with a Latino accent and in favor of Latino candidates with an American accent. Next, using a time‐lagged experiment and path analysis (Study 2, N = 328), we find support for the mediation role of expectancy violation. The disclosure of a Latino accent was related to changes in expectancy violation, which in turn were related to same‐direction changes in the ratings of personnel selection outcomes. These results suggest that the presence of a Latino accent is meaningful in personnel selection decisions and that expectancy violation is one of the mediating mechanisms by which these relationships occur.
This study examined the relationship between American listeners’ stereotypes, processing fluency, and attitudes toward speakers of five different foreign accents: French, Hindi, Mandarin, Russian, and Vietnamese. For each accent, (a) listeners’ stereotypes toward the national group marked by the accent and (b) listeners’ fluency processing speech produced in the accent were both positively associated with their attitudes toward speakers who spoke in that accent, even after controlling for the other factor. These findings support the theoretical claim that language attitudes are a function of both stereotypes and processing fluency.
This study reports the outcomes of a speaker evaluation experiment conducted in a higher educational context in Flanders, in which we investigated the influence of language variety (Standard vs. Colloquial Dutch), accent (Flemish vs. Eastern European) and name (Flemish vs. Eastern European) on students’ perceptions of a female university instructor. The results suggest that participants exhibited a relatively high level of tolerance toward both non-conforming speech and non-conforming speakers. Whereas Colloquial Dutch scored lower than Standard Dutch on standardness, we observed no negative effects of Colloquial Dutch on the teacher’s didactic competence, status, or social attractiveness. We did not find any evidence of a negative bias triggered by an Eastern European name. Whereas the Eastern European accent did have a negative impact on the teacher’s social attractiveness, we did not find any evidence of an accent bias affecting judgements of the teacher’s didactic competence and status.
Interacting with speakers of different accents is a prevalent global phenomenon. Given the considerable influence of accents in daily life, it is important to conduct a comprehensive review of listeners’ accent attitudes. This paper provides an integrative summary of research on accent attitudes, drawing from the Affective, Behavioral, and Cognitive (ABC) perspectives. We begin by outlining the social meaning of accents and laying out the theoretical foundations of the ABC approach. Then, we organize and integrate existing research findings using the ABC framework. Next, we illustrate how the perspectives intersect by discussing pertinent research findings. Drawing from various sociocultural contexts over many years, this review underscores the significant impact of accents on people’s thoughts, feelings, and behaviors. The review concludes by discussing limitations, proposing future directions, highlighting real-world relevance, and suggesting areas for research expansion.
Purpose: High-stakes decision-makers, including human resource (HR) professionals, often exhibit accent biases against second language speakers in professional evaluations. We extend this work by investigating how HR students evaluate simulated job interview performances in English by first and second language speakers of English.
Design/methodology/approach: Eighty HR students from Calgary and Montreal evaluated the employability of first language (L1) Arabic, English, and Tagalog candidates applying for two positions (nurse, teacher) at four points in the interview (after reading the applicant's resume, hearing their self-introduction, and listening to each of two responses to interview questions). Candidates' responses additionally varied in the extent to which they meaningfully answered the interview questions.
Findings: Students from both cities provided similar evaluations, employability ratings were similar for both advertised positions, and high-quality responses elicited consistently high ratings while evaluations for low-quality responses declined over time. All speakers were evaluated similarly based on their resumes and self-introductions, regardless of their language background. However, evaluations diverged for interview responses, where L1 Arabic and Tagalog speakers were considered more employable than L1 English speakers. Importantly, students' preference for L1 Arabic and Tagalog candidates over L1 English candidates was magnified when those candidates provided low-quality interview responses.
Originality: Results suggest that even in the absence of dedicated equity, diversity, and inclusion (EDI) training focusing on language and accent bias, HR students may be aware of second language speakers' potential disadvantages in the workplace, rewarding them in the current evaluations. Findings also highlight the potential influence of contextual factors on HR students' decision-making.
Purpose
The purpose of this paper is to clarify the influence of linguistic profiling on perceived employability. In doing so, we recognize multiple factors that can contribute to this relationship. Using the systems theory framework (STF) of career development, we categorize these factors based on whether they primarily reside within the societal system, the organizational system, or the individual system. Subsequently, we construct and present an open systems model that depicts the influence of linguistic profiling on perceived employability in context with other societal, organizational and individual factors.
Design/methodology/approach
The STF provides a theoretical perspective on how to contextualize linguistic profiling and perceived employability within the scope of career development. We employed an integrative literature review method to locate existing research studies that investigated the influences of linguistic profiling and perceived employability. The literature search process, coupled with inclusion and exclusion criteria, resulted in 46 publications retained for analysis.
Findings
Findings provided evidence of language ideologies that contribute to linguistic profiling globally. Linguistic profiling usually results in reduced perceived employability. There was variation in the research findings based on the context of the research studies. The context consisted of the societal, organizational and individual systems in which the research study was conducted. Findings supported our construction of an open systems model of the influence of linguistic profiling on perceived employability, which is rooted in the STF.
Originality/value
This study highlights the negative influence of linguistic profiling on perceived employability. It also demonstrates how the STF can be used to contextualize the linguistic profiling problem within societal and organizational systems.
In linguistics, the (foreign) accent has been defined as a set of patterns linked to segmental (phones) and suprasegmental (intonation, melody, rhythm, and tone) pronunciation that differ from standard and non‐standard varieties of a language. Other definitions go further by adding that a speaker's (foreign) accent provides information about their geographical, socio‐economic, and ethnic background. In this entry, we argue that an accent, as other traits of a speaker's identity, such as gender, age, or social status, not only affects language processing but also the evaluation of the speaker. We then explain how social categorization, stereotypes, and linguistic (dis)fluency are factors that influence language attitudes and (neuro)cognitive processes, and consequently modulate native speakers' behavior. Finally, we describe the consequences foreign‐accented speech has on everyday life from the point of view of both foreign and native speakers.
Speaking with a non‐native English accent at work is a prevalent global phenomenon. Yet, our understanding of the impact of having a non‐native accent at work is limited, in part because research on accents has been multidisciplinary, fragmented, and difficult for scholars to access and synthesize. To advance research on accents in the workplace, we provide an interdisciplinary and integrative review of research on non‐native accents drawing from the communications, social psychology, and organizational sciences literatures. First, we briefly review the dominant approaches taken in each literature. Second, we organize and integrate extant research findings using a 2 × 2 framework that incorporates the two main theoretical perspectives used to explain the effects of accents—stereotypes and processing fluency—and the two primary categories of workplace outcomes examined—interpersonal (i.e., others' evaluations of speakers with non‐native accents, such as hiring recommendations) and intrapersonal (i.e., non‐native‐accented speakers' own evaluations and experiences, such as sense of belonging). To facilitate future research, we end by articulating a research agenda including theoretical and methodological expansions related to the study of accents, identifying critical moderators, adopting an intersectional approach, and studying group‐level and potential positive effects of speaking with non‐native accents.
In this study, participants were presented with two tasks: the Implicit Association Test (IAT), and a mock trial task. In both tasks, the auditory stimuli were produced by native or foreign-accented speakers, and presented either free of noise or mixed with background white noise, to estimate the role of processing fluency on jurors’ appraisals. In the IAT, participants showed positive implicit biases toward native speech, and negative implicit biases toward foreign-accented speech. In the mock trial task, participants gave much harsher sentences to the foreign-accented than native defendant, but only when defendants’ statements were free of noise. Moreover, we found that participants’ implicit biases were a relevant predictor of the sentences they gave to the defendants. Our results suggest that categorization/stereotyping is the main mechanism responsible for the effect of defendants’ accents on jurors’ appraisals, and that members of an estimated group who violate social norms are punished more severely.
This study estimates empirically derived guidelines for effect size interpretation for research in social psychology overall and sub‐disciplines within social psychology, based on analysis of the true distributions of the two types of effect size measures widely used in social psychology (correlation coefficient and standardized mean differences). Analysis of empirically derived distributions of 12,170 correlation coefficients and 6,447 Cohen's d statistics extracted from studies included in 134 published meta‐analyses revealed that the 25th, 50th, and 75th percentiles corresponded to correlation coefficient values of 0.12, 0.24, and 0.41 and to Cohen's d values of 0.15, 0.36, and 0.65 respectively. The analysis suggests that the widely used Cohen's guidelines tend to overestimate medium and large effect sizes. Empirically derived effect size distributions in social psychology overall and its sub‐disciplines can be used both for effect size interpretation and for sample size planning when other information about effect size is not available.
The study of language attitudes is concerned with the social meanings people assign to language and its users. With roots in social psychology nearly a century ago, language attitudes research spans several academic disciplines and draws on diverse methodological approaches. In an attempt to integrate this work and traverse disciplinary boundaries and methodological proclivities, we propose that language attitudes—as a unified field—can be organized into five distinct—yet interdependent and complementary—lines of research: documentation, explanation, development, consequences, and change. After highlighting some of the key findings that have emerged from each area, we discuss several opportunities and challenges for future research.
Evidence suggests that accents can be typically more powerful in activating ethnicity categorization than appearance. Concurrently, some social categories, such as ethnicity, can be linked with other categories, such as religion. We investigate how people categorize those who belong to a (mis)matching pair of categories? In the present study, we investigated Germans’ categorization of women either wearing a headscarf (Muslim religious symbol), or not, and speaking either standard German or German with an Arabic accent. The “Who Said What?” paradigm and multinomial modelling yielded that category memory, indicative of subtyping, was best for nonprototypical targets (i.e., headscarf and standard German accent, no headscarf and Arabic accent). In contrast, in-group targets (no headscarf and standard German accent) were individually remembered better than all other targets, whereas nonprototypical targets (no-headscarf and Arabic accent) were not remembered individually at all. These findings are discussed in terms of intersectionality and category prototypicality.
Using a correspondence field experiment, the study reported in this article has investigated if immigrant job applicants with equivalent qualifications are treated differently in the Finnish labour market. The study consists of 5000 job applications that were sent out to 1000 advertised positions by five applicants of Finnish, English, Iraqi, Russian and Somali backgrounds, who differed only in their names. The findings show that applicants of immigrant origin receive significantly fewer invitations for a job interview than the native candidate, even if they possess identical language proficiency, education and vocational diplomas. However, the extent of discrimination is not equally distributed among the immigrant groups. Rather, job applicants from non-European backgrounds seem to suffer a significantly greater labour-market penalty. The findings clearly suggest that, despite anti-discrimination legislation and measures aimed at promoting equal employment opportunities, discrimination continues to remain a serious barrier to immigrants’ labour-market integration in a Nordic welfare society.
Nonnative-accented speakers face prevalent discrimination. The assumption that people freely express negative sentiments toward nonnative speakers has also guided common research methods. However, recent studies did not consistently find downgrading, so that prejudice against nonnative accents might even be questioned at first sight. The present theoretical article will bridge these contradictory findings in three ways: (a) We illustrate that nonnative speakers with foreign accents frequently may not be downgraded in commonly used first-impression and employment scenario paradigms. It appears that relatively controlled responding may be influenced by norms and motivations to respond without prejudice, whereas negative biases emerge in spontaneous responding. (b) We present an integrative view based on knowledge on modern forms of prejudice to develop modern notions of accent-ism, which allow for predictions when accent biases are (not) likely to surface. (c) We conclude with implications for interventions and a tailored research agenda.
This study investigates Americans’ attitudes toward the four major accents of South African English (SAfE) and several of their subvarieties in a hypothetical U.S. employment scenario. Participants perceived that SAfE accents possess positive language personality traits in comparison with standard American English; however, respondents identify SAfE speakers as foreign and perceive specific SAfE-accented varieties to be superior and more dynamic (e.g., General/Cultivated White SAfE and Indian SAfE) in relation to other SAfE-accented speakers (e.g., Mesolect Black SAfE and Cape Flats SAfE), even when they do not correctly identify the speaker’s country of origin.
Stereotype research emphasizes systematic processes over seemingly arbitrary contents, but content also may prove systematic. On the basis of stereotypes' intergroup functions, the stereotype content model hypothesizes that (a) 2 primary dimensions are competence and warmth, (b) frequent mixed clusters combine high warmth with low competence (paternalistic) or high competence with low warmth (envious), and (c) distinct emotions (pity, envy, admiration, contempt) differentiate the 4 competence-warmth combinations. Stereotypically, (d) status predicts high competence, and competition predicts low warmth. Nine varied samples rated gender, ethnicity, race, class, age, and disability out-groups. Contrary to antipathy models, 2 dimensions mattered, and many stereotypes were mixed, either pitying (low competence, high warmth subordinates) or envying (high competence, low warmth competitors). Stereotypically, status predicted competence, and competition predicted low warmth.
The present research unites two emergent trends in the area of language attitudes: (a) research on perceptions of nonnative speakers by nonnative listeners and (b) the search for general, basic mechanisms underlying the evaluation of nonnative accented speakers. In three experiments featuring an employment situation, German participants listened to a presentation given in English by a German speaker with a strong versus native-like accent (in Studies 1–3) versus a native speaker of English (in Study 1). They evaluated candidates with a strong accent worse than candidates with a native(-like) pronunciation—even to the degree that the quality of arguments was of no relevance (Study 1). Study 2 introduces an effective intervention to reduce these discriminatory tendencies. Across studies, affect and competence emerged as major mediators of hirability evaluations. Study 3 further revealed sequential indirect influences, which advance our understanding of previous inconsistent findings regarding disfluency and warmth perceptions.
Psychological research has neglected people whose accent does not match their appearance. Most research on person perception has focused on appearance, overlooking accents that are equally important social cues. If accents were studied, it was often done in isolation (i.e., detached from appearance). We examine how varying accent and appearance information about people affects evaluations. We show that evaluations of expectancy-violating people shift in the direction of the added information. When a job candidate looked foreign, but later spoke with a native accent, his evaluations rose and he was evaluated best of all candidates (Experiment 1a). However, the sequence in which information was presented mattered: When heard first and then seen, his evaluations dropped (Experiment 1b). Findings demonstrate the importance of studying the combination and sequence of different types of information in impression formation. They also allow predicting reactions to ethnically mixed people, who are increasingly present in modern societies.
Two experiments tested the prediction that heavy foreign-accented speakers are evaluated more negatively than mild foreign-accented speakers because the former are perceived as more prototypical (i.e., representative) of their respective group and their speech disrupts listeners’ processing fluency (i.e., is more difficult to process). Participants listened to a mild or heavy Punjabi- (Study 1) or Mandarin-accented (Study 2) speaker. Compared to the mild-accented speaker, the heavy-accented speaker in both studies was attributed less status (but not solidarity), was perceived as more prototypical of their respective group, disrupted listeners’ processing fluency, and elicited a more negative affective reaction. The negative effects of accent strength on status were mediated by processing fluency and sequentially by processing fluency and affect, but not by prototypicality. Theoretical, methodological, and practical implications are discussed.
Discrimination can be defined as any behavior that denies persons' certain rights because they belong to specific groups. In recent years, literature on discrimination in organizations suggests that considerable progress has been made to address barriers to job entry for women and minorities; however, more subtle discrimination still exists and impacts on their career progress. This article discusses discrimination in organizations, specifically focusing on distinctions between individual and institutional discrimination and theories of disparate impact and disparate treatment.
Most research on ethnicity has focused on visual cues. However, accents are strong social cues that can match or contradict visual cues. We examined understudied reactions to people whose one cue suggests one ethnicity, whereas the other cue contradicts it. In an experiment conducted in Germany, job candidates spoke with an accent either congruent or incongruent with their (German or Turkish) appearance. Based on ethnolinguistic identity theory, we predicted that accents would be strong cues for categorization and evaluation. Based on expectancy violations theory we expected that incongruent targets would be evaluated more extremely than congruent targets. Both predictions were confirmed: accents strongly influenced perceptions and Turkish-looking German-accented targets were perceived as most competent of all targets (and additionally most warm). The findings show that bringing together visual and auditory information yields a more complete picture of the processes underlying impression formation.
Applying a multilevel approach to meta-analysis is a strong method for dealing with dependency of effect sizes. However, this method is relatively unknown among researchers and, to date, has not been widely used in meta-analytic research. Therefore, the purpose of this tutorial was to show how a three-level random effects model can be applied to meta-analytic models in R using the rma.mv function of the metafor package. This application is illustrated by taking the reader through a step-by-step guide to the multilevel analyses comprising the steps of (1) organizing a data file; (2) setting up the R environment; (3) calculating an overall effect; (4) examining heterogeneity of within-study variance and between-study variance; (5) performing categorical and continuous moderator analyses; and (6) examining a multiple moderator model. By example, the authors demonstrate how the multilevel approach can be applied to meta-analytically examining the association between mental health disorders of juveniles and juvenile offender recidivism. In our opinion, the rma.mv function of the metafor package provides an easy and flexible way of applying a multi-level structure to meta-analytic models in R. Further, the multilevel meta-analytic models can be easily extended so that the potential moderating influence of variables can be examined.
This article analyses the influence of accent on discrimination against immigrants by examining the hypothesis that accent leads to discrimination only in more prejudiced individuals, merely because people speaking with a native accent are perceived to be better qualified than those whose accent is non-standard. In Study 1 (N = 71), we found that only prejudiced individuals use accent to discriminate against immigrants. In Study 2 (N = 124), we replicated this effect and found that the influence of accent on discrimination is mediated by the perceived quality of the accent. Study 3 (N = 105) replicated the previous results even after controlling for the effect of stereotyping. These results are the first experimental illustration of the hypothesis that accent triggers intergroup discrimination only among prejudiced individuals because they evaluate native accents as being qualitatively better than accents of immigrants, thereby legitimizing ingroup bias.
Objective: Discrimination against nonnative speakers is widespread and largely socially acceptable. Nonnative speakers are evaluated negatively because accent is a sign that they belong to an outgroup and because understanding their speech requires unusual effort from listeners. The present research investigated intergroup bias, based on stronger support for hierarchical relations between groups (social dominance orientation [SDO]), as a predictor of hiring recommendations of nonnative speakers. Method: In an online experiment using an adaptation of the thin-slices methodology, 65 U.S. adults (54% women; 80% White; Mage = 35.91, range = 18–67) heard a recording of a job applicant speaking with an Asian (Mandarin Chinese) or a Latino (Spanish) accent. Participants indicated how likely they would be to recommend hiring the speaker, answered questions about the text, and indicated how difficult it was to understand the applicant. Results: Independent of objective comprehension, participants high in SDO reported that it was more difficult to understand a Latino speaker than an Asian speaker. SDO predicted hiring recommendations of the speakers, but this relationship was mediated by the perception that nonnative speakers were difficult to understand. This effect was stronger for speakers from lower status groups (Latinos relative to Asians) and was not related to objective comprehension. Conclusions: These findings suggest a cycle of prejudice toward nonnative speakers: Not only do perceptions of difficulty in understanding cause prejudice toward them, but also prejudice toward low-status groups can lead to perceived difficulty in understanding members of these groups.
Two experiments examined the effects of processing fluency—that is, the ease with which speech is processed—on language attitudes toward native- and foreign-accented speech. Participants listened to an audio recording of a story read in either a Standard American English (SAE) or Punjabi English (PE) accent. They heard the recording either free of noise or mixed with background white noise of various intensity levels. Listeners attributed more solidarity (but equal status) to the SAE than the PE accent. Compared to quieter listening conditions, noisier conditions reduced processing fluency, elicited a more negative affective reaction, and resulted in more negative language attitudes. Processing fluency and affect mediated the effects of noise on language attitudes. Theoretical, methodological, and practical implications are discussed.
Systematic reviews and meta-analyses have become increasingly important in health care. Clinicians read them to keep up to date with their field [1],[2], and they are often used as a starting point for developing clinical practice guidelines. Granting agencies may require a systematic review to ensure there is justification for further research [3], and some health care journals are moving in this direction [4]. As with all research, the value of a systematic review depends on what was done, what was found, and the clarity of reporting. As with other publications, the reporting quality of systematic reviews varies, limiting readers' ability to assess the strengths and weaknesses of those reviews.
Several early studies evaluated the quality of review reports. In 1987, Mulrow examined 50 review articles published in four leading medical journals in 1985 and 1986 and found that none met all eight explicit scientific criteria, such as a quality assessment of included studies [5]. In 1987, Sacks and colleagues [6] evaluated the adequacy of reporting of 83 meta-analyses on 23 characteristics in six domains. Reporting was generally poor; between one and 14 characteristics were adequately reported (mean = 7.7; standard deviation = 2.7). A 1996 update of this study found little improvement [7].
In 1996, to address the suboptimal reporting of meta-analyses, an international group developed a guidance called the QUOROM Statement (QUality Of Reporting Of Meta-analyses), which focused on the reporting of meta-analyses of randomized controlled trials [8]. In this article, we summarize a revision of these guidelines, renamed PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analyses), which have been updated to address several conceptual and practical advances in the science of systematic reviews (Box 1).
Box 1: Conceptual Issues in the Evolution from QUOROM to PRISMA
Completing a Systematic Review Is an Iterative Process
The conduct of a systematic review depends heavily on the scope and quality of included studies: thus systematic reviewers may need to modify their original review protocol during its conduct. Any systematic review reporting guideline should recommend that such changes can be reported and explained without suggesting that they are inappropriate. The PRISMA Statement (Items 5, 11, 16, and 23) acknowledges this iterative process. Aside from Cochrane reviews, all of which should have a protocol, only about 10% of systematic reviewers report working from a protocol [22]. Without a protocol that is publicly accessible, it is difficult to judge between appropriate and inappropriate modifications.
Google Scholar (GS), a commonly used web-based academic search engine, catalogues between 2 and 100 million records of both academic and grey literature (articles not formally published by commercial academic publishers). Google Scholar collates results from across the internet and is free to use. As a result it has received considerable attention as a method for searching for literature, particularly in searches for grey literature, as required by systematic reviews. The reliance on GS as a standalone resource has been greatly debated, however, and its efficacy in grey literature searching has not yet been investigated. Using systematic review case studies from environmental science, we investigated the utility of GS in systematic reviews and in searches for grey literature. Our findings show that GS results contain moderate amounts of grey literature, with the majority found on average at page 80. We also found that, when searched for specifically, the majority of literature identified using Web of Science was also found using GS. However, our findings showed moderate/poor overlap in results when similar search strings were used in Web of Science and GS (10–67%), and that GS missed some important literature in five of six case studies. Furthermore, a general GS search failed to find any grey literature from a case study that involved manual searching of organisations' websites. If used in systematic reviews for grey literature, we recommend that searches of article titles focus on the first 200 to 300 results. We conclude that whilst Google Scholar can find much grey literature and specific, known studies, it should not be used alone for systematic review searches. Rather, it forms a powerful addition to other traditional search methods. In addition, we advocate the use of tools to transparently document and catalogue GS search results to maintain high levels of transparency and the ability to be updated, critical to systematic reviews.
Presents a novel approach to conducting meta-analysis using structural equation modeling.
Structural equation modeling (SEM) and meta-analysis are two powerful statistical methods in the educational, social, behavioral, and medical sciences. They are often treated as two unrelated topics in the literature. This book presents a unified framework on analyzing meta-analytic data within the SEM framework, and illustrates how to conduct meta-analysis using the metaSEM package in the R statistical environment.
Meta-Analysis: A Structural Equation Modeling Approach begins by introducing the importance of SEM and meta-analysis in answering research questions. Key ideas in meta-analysis and SEM are briefly reviewed, and various meta-analytic models are then introduced and linked to the SEM framework. Fixed-, random-, and mixed-effects models in univariate and multivariate meta-analyses, three-level meta-analysis, and meta-analytic structural equation modeling, are introduced. Advanced topics, such as using restricted maximum likelihood estimation method and handling missing covariates, are also covered. Readers will learn a single framework to apply both meta-analysis and SEM. Examples in R and in Mplus are included.
This book will be a valuable resource for statistical and academic researchers and graduate students carrying out meta-analyses, and will also be useful to researchers and statisticians using SEM in biostatistics. Basic knowledge of either SEM or meta-analysis will be helpful in understanding the materials in this book.
Systematic reviews should build on a protocol that describes the rationale, hypothesis, and planned methods of the review; few reviews report whether a protocol exists. Detailed, well-described protocols can facilitate the understanding and appraisal of the review methods, as well as the detection of modifications to methods and selective reporting in completed reviews. We describe the development of a reporting guideline, the Preferred Reporting Items for Systematic reviews and Meta-Analyses for Protocols 2015 (PRISMA-P 2015). PRISMA-P consists of a 17-item checklist intended to facilitate the preparation and reporting of a robust protocol for the systematic review. Funders and those commissioning reviews might consider mandating the use of the checklist to facilitate the submission of relevant protocol information in funding applications. Similarly, peer reviewers and editors can use the guidance to gauge the completeness and transparency of a systematic review protocol submitted for publication in a journal or other medium.
Prejudice against a social group may lead to discrimination of members of this group. One very strong cue of group membership is a (non)standard accent in speech. Surprisingly, hardly any interventions against accent-based discrimination have been tested. In the current article, we introduce an intervention in which what participants experience themselves unobtrusively changes their evaluations of others. In the present experiment, participants in the experimental condition talked to a confederate in a foreign language before the experiment, whereas those in the control condition received no treatment. Replicating previous research, participants in the control condition discriminated against Turkish-accented job candidates. In contrast, those in the experimental condition evaluated Turkish- and standard-accented candidates as similarly competent. We discuss potential mediating and moderating factors of this effect.
A fundamental model of stereotypes, the stereotype content model (SCM, Fiske, Cuddy, Glick, & Xu, 2002), postulates that stereotypes of many social groups vary on two core dimensions: warmth and competence. In line with traditional gender stereotypes, the SCM predicts women to be perceived as warmer than men, and men to be perceived as more competent than women. Research on implicit measurement of stereotypes suggests that, next to people’s underlying beliefs, a major predictor is the tendency to favor one’s own group (Rudman, Greenwald, & McGhee, 2001). We examined gender stereotypes concerning warmth and competence, using Implicit Association Tests (IATs, Greenwald, McGhee, & Schwartz, 1998) and drawing on diverse samples of women and men in eastern and western Germany (i.e., students and non-students; total N = 384). On the warmth dimension, women and men showed a traditional women-warmth stereotype, confirming predictions of the SCM. On the competence dimension, associations of own gender and competence were observed for both men and women, suggesting the impact of self-favoring processes. Findings are discussed with respect to social role theory (Eagly, 1987) and the changing roles of women.
This study applies concepts and methods from the domain of Implicit Social Cognition to examine language attitudes toward foreign and U.S. accented speech. Implicit attitudes were measured using an Implicit Association Test (IAT) that incorporated audio cues as experimental stimuli. Explicit attitudes were measured through self-report questionnaires. Participants exhibited a pro-U.S. accent bias on the IAT measure but a pro-foreign accent bias on explicit measures. This divergence supports the conclusion that implicit and explicit attitudes are separable attitude constructs resulting from distinct mental processes and suggests that language attitudes research—which has traditionally measured only explicit attitudes—would benefit by incorporating indirect measures. The Associative-Propositional Evaluation Model is proposed as a comprehensive and consistent theory to explain the cognitive processing of language attitudes.
Two studies examined the compensation hypothesis that members of both high- and low-status groups associate high-status groups with high levels of competence and low levels of warmth on the one hand, and low-status groups with low levels of competence and high levels of warmth, on the other. Building upon existing linguistic relations between the French and the Belgians, Study 1 had standard, i.e. French, and non-standard, i.e. Belgian, speakers rate the linguistic skills, competence, and warmth of both groups and report their meta-stereotypes. As predicted, both groups of participants saw the French as more skilled linguistically than Belgians and evaluated standard speakers as more competent than warm and non-standard speakers as more warm than competent. This pattern also emerged in respondents’ meta-stereotypes. Study 2 revealed that compensation was less marked among a third group of Francophone speakers, i.e. Swiss, even if the latter respondents seemed well aware of the pattern guiding Belgian and French representations of each other. We discuss the implications of the findings in terms of motivated intergroup stereotypes.
We propose and test a new theory explaining glass-ceiling bias against nonnative speakers as driven by perceptions that nonnative speakers have weak political skill. Although nonnative accent is a complex signal, its effects on assessments of the speakers' political skill are something that speakers can actively mitigate; this makes it an important bias to understand. In Study 1, White and Asian nonnative speakers using the same scripted responses as native speakers were found to be significantly less likely to be recommended for a middle-management position, and this bias was fully mediated by assessments of their political skill. The alternative explanations of race, communication skill, and collaborative skill were nonsignificant. In Study 2, entrepreneurial start-up pitches from national high-technology, new-venture funding competitions were shown to experienced executive MBA students. Nonnative speakers were found to have a significantly lower likelihood of receiving new-venture funding, and this was fully mediated by the coders' assessments of their political skill. The entrepreneurs' race, communication skill, and collaborative skill had no effect. We discuss the value of empirically testing various posited reasons for glass-ceiling biases, how the importance and ambiguity of political skill for executive success serve as an ostensibly meritocratic cover for nonnative speaker bias, and other theoretical and practical implications of this work. (PsycINFO Database Record (c) 2013 APA, all rights reserved).
Meta-analysis is an indispensable tool used to synthesize research findings in the social, educational, medical, management, and behavioral sciences. Most meta-analytic models assume independence among effect sizes. However, effect sizes can be dependent for various reasons. For example, studies might report multiple effect sizes on the same construct, and effect sizes reported by participants from the same cultural group are likely to be more similar than those reported by other cultural groups. This article reviews the problems and common methods to handle dependent effect sizes. The objective of this article is to demonstrate how 3-level meta-analyses can be used to model dependent effect sizes. The advantages of the structural equation modeling approach over the multilevel approach with regard to conducting a 3-level meta-analysis are discussed. This article also seeks to extend the key concepts of Q statistics, I2, and R2 from 2-level meta-analyses to 3-level meta-analyses. The proposed procedures are implemented using the open source metaSEM package for the R statistical environment. Two real data sets are used to illustrate these procedures. New research directions related to 3-level meta-analyses are discussed. (PsycINFO Database Record (c) 2013 APA, all rights reserved).
The effects of status ascriptions (high vs. low) and accent (standard American vs. Mexican-American) upon perceptions of speaker status and solidarity are examined. Consequences of the independent variables for a novel dependent variable, perceived control of speech style, are examined also. One hundred and twenty-two respondents rated one of eight message versions in a between-group design. Results indicated, among other things, that: (1) status ascriptions and accent combined additively to affect judgements of speaker status, as did status ascriptions and diversity; (2) lexical diversity and accent affected both solidarity judgements and status judgements interactively, although the nature of the interactive effect differed across measures; and (3) lexical diversity was an especially potent determinant of perceptions of control of speech style. It is suggested that a clear picture of linguistic and non-linguistic determinants of judgements of speaker status is emerging.
Unequal outcomes in professional hiring for individuals from less privileged backgrounds have been widely reported in England. Although accent is one of the most salient signals of such a background, its role in unequal professional outcomes remains underexamined. This paper reports on a large-scale study of contemporary attitudes to accents in England. A large representative sample ( N = 848) of the population in England judged the interview performance and perceived hirability of “candidates” for a trainee solicitor position at a corporate law firm. Candidates were native speakers of one of five English accents stratified by region, ethnicity, and class. The results suggest persistent patterns of bias against certain accents in England, particularly Southern working-class varieties, though moderated by factors such as listener age, content of speech, and listeners’ psychological predispositions. We discuss the role that the observed bias may play in perpetuating social inequality in England and encourage further research on the relationship between accent and social mobility.
Intersectionality theory states that social identities are interdependent and that the meaning of identities derives from their relationship to one another. In this chapter, we describe how taking an intersectional approach to the study of prejudice confrontations may advance our understanding of confrontations by targets and nontargets of prejudice. Given that the social identities of targets and nontargets shape prejudice confrontations, the intersectionality of identities may also be expected to affect confrontations. Using the intersectional frameworks of double jeopardy, ethnic prominence, and intersectional invisibility we develop hypotheses describing how intersectionality may be incorporated into future research on prejudice confrontations. We then preview findings from research in our lab that directly examined how race and gender intersect to predict whether women of color will confront racist and sexist evaluators. We conclude by considering the applications of findings in this area to policy and practices that encourage stigmatized individuals to confront prejudice.
Mor Barak, Managing Diversity: Toward a Globally Inclusive Workplace
Managing Diversity won the prestigious Academy of Management’s George Terry Book award for “the most outstanding contribution to the advancement of management knowledge” and received the CHOICE Award for Outstanding Academic Titles by the Association of College and University Libraries.
“An excellent resource to develop, theorize, and work out the inclusive workplace in a very comprehensive, encompassing, and interdisciplinary way. .. Boxes, tables, graphs, and figures as well as practical examples and empirical illustrations… make the book very interesting for both the conceptual, pedagogical research interest and the practical, educational interest.”
- Cordula Barzantny, Academy of Management Learning & Education Journal
This book introduces a unique and refreshing prism that is highly useful for managers and scholars alike. The authentic examples and case studies bring the content to life and make this book a very interesting and captivating read. Managing Diversity is a ‘must read’ for managers who need to effectively manage today’s diverse work force in order to survive and thrive in the global economy.
- Alan D. Levy, Chairman and CEO Tishman International Companies
Successful management of today’s increasingly diverse global workforce is among the most important challenges faced by corporate leaders, human resource managers, and management consultants. In the Third Edition of this award-winning book, Michàlle E. Mor Barak argues that exclusion is one of the most significant problems facing today’s diverse workforce, and she provides strategies for unleashing the potential embedded in a multicultural and diverse global workforce.
Key Features:
• Offers up-to-date information and statistics on the new realities of the global workforce, including demographic, legislation, and social policy trends around the world
• Analyzes the causes and consequences of workforce exclusion, highlighting the groups commonly excluded in various countries and providing theories that explain exclusion and inclusion in the workplace
• Provides an original and comprehensive model of the Inclusive Workplace suggesting policies, procedures and programs that facilitate its implementation
New to This Edition
• New and revised diversity case examples from around the world
• Updated statistics on global workforce trends and new legislations and social policies in different countries
• New information about leadership in diversity management
• Up-to-date research on diversity management outcomes
• Assessment tools for organizational diversity climate and for inclusion-exclusion with data on their psychometric properties
A password-protected instrucot teaching site at… includes PowerPoint slides, chapter overviews and outlines and test questions.
Michàlle E. Mor Barak is a professor at the University of Southern California with a joint appointment at the School of Social Work and the Marshall School of Business. She holds the Lenore Stein-Wood and William S. Wood Professorship of Social Work and Business in a Global Society.
Using quantitative methods, this article examines the effect of foreign accents on job applicants’ employability ratings in the context of a simulated employment interview experiment conducted in the USA. It builds upon the literature on aesthetic labour, which focuses largely on the role of physical appearance in employment relations, by shifting attention to its under-investigated auditory and aural dimensions. The results suggest that the managerial respondents actively discriminate in telephone-based job interviews against applicants speaking Chinese-, Mexican- and Indian-accented English, and all three are rated higher in non-customer-facing jobs than in customer-facing jobs. Job applicants who speak British-accented English, especially men, fare as well as, and at times better than, native candidates who speak American English. The article makes a contribution to the sociological literatures surrounding aesthetic labour and discrimination and prejudice against migrant workers.
This study was designed to determine how ethnicity, the amount of perceived accent or dialect, and comprehensibility affect a speaker's employability. Sixty human resource specialists judged 3 female potential applicants. The applicants represented speakers of Spanish-influenced English, Asian-influenced English, and African American Vernacular English. When the speaker's perceived accent or dialect was minimal, perceived ethnicity did not affect employability. However, all speakers with maximally perceived accents or dialects were given a lower employability rating. Thus, speakers with a maximally perceived accent or dialect should consider accent or dialectal modification if their comprehensibility or prospective employability is compromised. © 2006 by the American Counseling Association. All rights reserved.
One of the chief goals of most second language learners is to be understood in their second language by a wide range of interlocutors in a variety of contexts. Although a nonnative accent can sometimes interfere with this goal, prior to the publication of this study, second language researchers and teachers alike were aware that an accent itself does not necessarily act as a communicative barrier. Nonetheless, there had been very little empirical investigation of how the presence of a nonnative accent affects intelligibility, and the notions of “heavy accent” and “low intelligibility” had often been confounded. Some of the key findings of the study—that even heavily accented speech is sometimes perfectly intelligible and that prosodic errors appear to be a more potent force in the loss of intelligibility than phonetic errors—added support to some common, but weakly substantiated beliefs. The study also provided a framework for a program of research to evaluate the ways in which such factors as intelligibility and comprehensibility are related to a number of other dimensions. The authors have extended and replicated the work begun in this study to include learners representing other L1 backgrounds (Cantonese, Japanese, Polish, Spanish) and different levels of learner proficiency, as well as other discourse types (Derwing & Munro, 1997; Munro & Derwing, 1995). Further support for the notion that accent itself should be regarded as a secondary concern was obtained in a study of processing difficulty (Munro & Derwing, 1995), which revealed that nonnative utterances tend to require more time to process than native-produced speech, but failed to indicate a relationship between strength of accent and processing time.The approach to L2 speech evaluation used in this study has also proved useful in investigations of the benefits of different methods of teaching of pronunciation to ESL learners. In particular, it is now clear that learner assessments are best carried out with attention to the multidimensional nature of L2 speech, rather than with a simple focus on global accentedness. It has been shown, for instance, that some pedagogical methods may be effective in improving intelligibility while others may have an effect only on accentedness (Derwing, Munro, & Wiebe, 1998).
As the workplace becomes increasingly global, organizations are more likely to employ individuals with non-native accents. The present study looked at the influence of accents on the evaluation of job applicants during an interview. In addition, a path model was developed to understand the accent condition–hiring recommendation relationship. Participants were asked to evaluate an applicant with one of three accents (Midwestern US, French, Mexican) at two understandability levels (low and high) by listening to an audiofile. The results showed that the applicant with the Midwestern US accent was seen as more hirable than the applicant with the French low understandability accent. The path model indicated that the accent condition–hiring recommendation relationship was mediated by similarity, interpersonal attraction, and understandability.
Lambert's use of the “matched‐guise” technique to study stereotyped impressions of personality characteristics from contrasting spoken dialects and languages has been extended to investigate three other evaluative dimensions in relation to British regional and foreign accents. 177 Ss were required to rate the “aesthetic”, “communicative” and “status” contents of various accents presented both vocally and conceptually. Although a generalised pattern of ranking accents across these dimensions emerged, the factors of age, ser, social class and regional membership were found to be important determinants of evaluation. The social and educational significance of these findings were discussed.
The number of recent immigrants and refugees in the United States is growing dramatically. Among key reasons for migration is search for adequate employment and hope for opportunities to develop occupationally. However, recent immigrants and refugees face multiple obstacles in their career development in the United States. This article uses social cognitive career theory to examine the role of relocation circumstances, stressors of migration, acculturation patterns, and oppression on the career development and employment functioning of recent immigrants and refugees. Specific suggestions for career counselors working with this population are provided.
The present research explored how White college students may exhibit response patterns associated with a subtle and rationalizable contemporary bias, aversive racism. In the study, higher and lower prejudice-scoring participants evaluated applicants for admission to their university, for whom information about high school achievement and college board scores (aptitude and achievement test scores) was independently varied as strong or weak. As predicted, discrimination against Black applicants relative to White applicants did not occur when the credentials were consistently strong or weak; however, discrimination by relatively high prejudice-scoring participants did emerge when the credentials were mixed and hence ambiguous. Moreover, relatively high prejudice-scoring participants weighed the different, conflicting criteria in ways that could justify or rationalize discrimination against Black applicants. The implications of these data for understanding contemporary racism and their relation to the shifting standards model of bias are considered.
Research concerning native speaker reactions to the speech of language learners has generally flowed from an interpersonal perspective (i.e. speaker/listener). On the other hand, the broader social psychological domain of language attitude research has typically been based on an intergroup perspective (i.e. ingroup/outgroup). The two perspectives are elaborated here and shown to provide complementary interpretations of evaluations of non-native speech. From the intergroup viewpoint, evaluations are based upon two processes: identification of the speaker's social group attributes and group-based inferences. The particularly complex role of identification for non-native speakers is discussed in detail. Within the interpersonal sphere, inferences regarding speaker competence and generalized negative affect are emphasized.
Purpose
Although statistical evidence clearly demonstrates discrimination against foreign‐accented individuals in the workplace, surprisingly little research attention has been paid to how such individuals are evaluated when they apply for jobs. Thus, the aim of this paper is to examine the effects of applicant accent on access‐related employment decisions across four jobs that differed on job status and communication demands.
Design/methodology/approach
The study used a 3 (applicant accent: Standard American English, French, and Japanese) × 2 (job status: low vs high)×2 (communication demands: low vs high) mixed‐factorial design, and data from 286 college students at two different locations.
Findings
Results show that in comparison with French‐accented applicants, Japanese‐accented applicants fared worse on employment‐related decisions, especially for jobs that had high communication demands, even after controlling for applicant understandability and location. French‐accented applicants were viewed as favorably as, or more favorably than, Standard American English‐accented applicants.
Research limitations/implications
Applicant accent was confounded with applicant names. Thus, it is not known whether the obtained results are due to applicant accent, names, or both.
Practical implications
It was found that organizations could do one of the following: use structured interviews; train interviewers on potential biases against foreign‐accented applicants; and provide more individuating information to reduce the effects of accent‐based stereotypes on employment‐related decisions.
Originality/value
The paper considers the communication demands of jobs and job status as influences on the evaluation of foreign‐accented applicants.
Purpose
Despite the fact that Hispanics are the largest and fastest growing segment of the population and that 44 percent of Hispanics of 18 years of age and older speak English less than very well, research examining the impact of Spanish‐accented English on employment‐related decisions has been scarce. Therefore, the present study aimed to examine the effects of the accent (standard American English and Mexican Spanish) of a hypothetical job applicant on employment‐related judgments and hiring decisions.
Design/methodology/approach
Participants made employment‐related decisions (i.e. job suitability ratings, likelihood of a promotion, and hiring decision) and judgments of personal attributes (i.e. perceived competence and warmth) of a hypothetical applicant for an entry‐level software engineering job. The accent of the applicant was manipulated using the matched‐guise technique.
Findings
Results showed that compared to an applicant with a standard American‐English accent, one with a Mexican‐Spanish accent was at a disadvantage when applying for the software engineering job. The Mexican‐Spanish‐accented applicant was rated as less suitable for the job and viewed as less likely to be promoted to a managerial position. In addition, fewer participants decided to hire the Mexican‐Spanish‐accented applicant than the standard American English‐accented applicant.
Practical implications
Given the negative evaluations of the Mexican‐Spanish‐accented applicant, recruiters and interviewers should be selected who do not view foreign accents negatively. Furthermore, organizations should make a conscious effort to regard foreign accents as assets to their businesses.
Originality/value
This research contributes to our understanding of how foreign accents influence decisions that have important economic consequences for individuals.
The present study investigated the effects of ethnicity, accent, and job status in the selection interview. Two hundred and ten student subjects viewed simulated, videotaped selection interviews in which applicant ethnicity, applicant accent, and job status were manipulated in a completely crossed 3 × 2 × 2 factorial design. Two sets of ratings were obtained, selection decision ratings of the job applicant, and subjective importance ratings based on the previous selection decisions. Significant main effects were obtained for ethnicity and job status in selection decision as predicted. Contrary to predictions however, the main effect for accent was not found significant. Further analysis of the importance ratings indicated a significant main effect for accent, but not for ethnicity. These findings indicated that applicant ethnicity and job status were important in selection interview decision and accent was not. Contrary to this however, subjects considered that accent was important in their selection decision ratings, while ethnicity was not.