Institutionen för klinisk neurovetenskap - PDF

Description
Institutionen för klinisk neurovetenskap Psykologprogrammet, termin 9-10 Huvudämne: Psykologi Examensarbete i psykologi (2PS026), 30 poäng Vårterminen 2014 Recruiting without discriminating - An experimental

Please download to get full document.

View again

of 18
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information
Category:

Reports

Publish on:

Views: 16 | Pages: 18

Extension: PDF | Download: 0

Share
Transcript
Institutionen för klinisk neurovetenskap Psykologprogrammet, termin 9-10 Huvudämne: Psykologi Examensarbete i psykologi (2PS026), 30 poäng Vårterminen 2014 Recruiting without discriminating - An experimental study of hiring discrimination on ethnicity, using structured video interview versus unstructured face-to-face interview Karam Hanna Mentor: Bo Melin, institutionen för klinisk neurovetenskap Examiner: Mats Olsson, institutionen för klinisk neurovetenskap 1 Institutionen för klinisk neurovetenskap Psykologprogrammet, termin 9-10 Huvudämne: Psykologi Examensarbete i psykologi (2PS026), 30 poäng Vårterminen 2014 Recruiting without discriminating - An experimental study of hiring discrimination on ethnicity, using structured video interview versus unstructured face-to-face interview Abstract Background: Ethnic discrimination is believed to be present in recruitment processes and this is reflected in the unequal level of employment in Sweden today. Discrimination may come in a more subtle and implicit form, especially when employers use methods that are more susceptible to subjective judgment. Method: This study examined ethnic discrimination in a recruitment process by comparing two recruiting methods. Participants (N=233) were randomized to an experiment group to perform a structured video interview, and to a control group for an unstructured face-to-face interview. Results: No effects were found between the groups after the recruitment process. However, effects were found within the groups showing that Swedish participants had higher odds of advancing in the recruitment process with 298 percent in the experiment group and 495 percent in the control group, compared to non- Swedish participants in each group. Discussion: While the obtained results suggest discrimination was present the study had methodological issues that may have influenced the outcome. These consisted of the control group undergoing a premature selection prior to the recruiting process, and not having the employers blinded to the purpose of the study. Future research should focus on further assessment of video interviewing as a screening method. Keywords: decision-making, discrimination, ethnicity, interview, recruiting 2 Sammanfattning Bakgrund: Etnisk diskriminering förekommer i rekryteringsprocesser och detta avspeglas i den ojämna anställningsfördelningen i Sverige idag. Diskriminering kan ske subtilt och implicit, i synnerhet när rekryterare använder sig av metoder som är mer känsliga för subjektiva bedömningar. Metod: Denna studie undersökte etnisk diskriminering i en rekryteringsprocess genom att jämföra två rekryteringsmetoder. Deltagare (N=233) randomiserades till antingen en experimentgrupp där en strukturerad videointervju genomfördes, eller en kontrollgrupp där en ostrukturerad face-to-face intervju genomfördes. Resultat: Inga skillnader på etnicitet påvisades mellan grupperna efter rekryteringsprocessen. Dock fanns en effekt inom grupperna som visade att svenska deltagare, i jämförelse med ickesvenska hade 298 procents högre sannolikhet att gå vidare i rekryteringsprocessen i experimentgruppen. Motsvarande siffra för kontrollgruppen var 495 procent. Diskussion: Trots att resultaten indikerar att diskriminering förekom i denna studie har metodologiska tillkortakommanden förelegat och kan ha påverkat utfallet. Dessa utgjordes av att kontrollgruppen genomgick en för tidig selektion innan rekryteringsprocessen påbörjades samt att rekryterarna inte var blinda för studiens syfte. Framtida forskning bör fokusera på att vidare utvärdera videointervju som screeningmetod. Nyckelord: beslutsfattande, diskriminering, etnicitet, intervju, rekrytering 3 Recruiting without discriminating - An experimental study of hiring discrimination on ethnicity, using structured video interview versus unstructured face-to-face interview Karam Hanna Introduction Discrimination Discrimination is the type of unjustifiable behavior directed toward a group of people along with its members (Myers, 2010). The Human rights constitution of the United Nations and the Swedish constitutional law states that every individual is born free and with equal value and equal rights (diskrimineringslagen, SFS 2008:567). Researchers have differentiated between old-fashioned discrimination and modern discrimination (Ziegert & Hanges, 2005), where modern discrimination is a subtle form of discrimination, as opposed to the oldfashioned that is typically more direct and hostile (Brief, 1998). In this modern form of discrimination one is usually rational in the act of discrimination and therefore able to disguise the true intentions in order to maintain ones self-image (McConahay, 1986). This form of discrimination is believed to exist in our values and attitudes, and is ruled by unconscious psychological factors, and is therefore more difficult to identify (Ziegert & Hanges, 2005). Earlier research In recent years there has been a series of research in the area of ethnic discrimination in the work labor, however the majority of these studies have looked at race specifically, and were conducted in America (Brief et al., 2000; Pager & Western, 2012). There is limited research on discrimination on ethnic background. Carlsson & Rooth (2007) are amongst the few that have looked at hiring discrimination in Sweden. They used a method called correspondence testing, where two identical applications are sent in response to the same position, with the only difference being the names of the fictive applicants, names that are informative in terms of ethnic background. The results showed a 50% higher chance for fictive applicants with typical Swedish names to be called to an interview, after having sent a job application, compared to those fictive applicants of a typical Middle Eastern name, with an identical job application. Stenberg (2004) conducted a similar experiment, although on a relatively small scale, and found that employers who were contacted by phone about a position were 15% (own calculations) more likely to encourage applicants with Swedish sounding names than people with Arabic sounding names. The International Labour Organization (ILO 2006) has also studied hiring discrimination in an experiment on the Swedish labor market, where Swedish people were compared to people born in Sweden but with a Middle Eastern ethnic background. The experiment was designed to include three stages, each with a potential of detecting hiring discrimination. In the first stage employers were contacted via phone and asked if an application form can be sent, and in the next stage the application were sent using the same correspondence testing method as Carlsson & Rooth (2007). The third and final stage was the interviews, where actors were sent to play the fictive applicants. Results of discrimination were found in the second stage (sending application form and waiting for a callback to an interview), which saw applicants with a Swedish sounding name were 50% more likely to be offered an interview compared to applicants with 4 a Middle Eastern sounding name. Interestingly, results from the first and third stage show no indication of ethnic discrimination. Other studies have failed to prove the occurrence of hiring discrimination (Edin & Lagerström, 2006). People that are unemployed, or simply looking to find a new job, are able to register on the Arbetsförmedlingen (Swedish employment agency) and uploaded a resume and a personal letter, in the hopes of being directly contacted by potential employers. When this database was reviewed in their quasi-experimental study, results saw that while women were contacted by employers more often than men, there was no indication of discrimination against ethnic background (Edin & Lagerström, 2006). So far research presents mixed findings in terms of ethnic discrimination, some in which there is no proof of such unequal treatment. Indeed, it is argued that perceived discrimination may as well be that employees base their decisions on other, more covert, traits or qualifications (Pager & Western, 2012). Because even when there are visible indicators of discrimination (e.g. difference between Swedish and non-swedish people in terms of employment) we can still not argue that it is discrimination, seeing as we cannot know with full accuracy that these differences are not in fact a result of other non-visible differences such as language skills (Carlsson & Rooth, 2007). This seems to be a common explanation used to argue against discrimination. This issue was investigated in an official report of the Swedish government (SOU 2005:56). The argument that the existing notable difference in level of employment between the two groups is caused by lack of certain qualities in the non-swedish group, e.g. sufficient language (SOU 2005:115), failed to serve as an explanation, and the conclusion was that it is indeed ethnic discrimination that is the main reason for the inequality in employment. Furthermore, as discussed earlier, discrimination today exists in a more modern form, where it is subtler, and hence more difficult to identify, therefore hiring discrimination is not easily measured and can go by unnoticed (Pager & Western, 2012). Like many other countries, Sweden strives to have a multi-cultural and diverse society, which is why this often presents as an important question in Swedish politics. To achieve this, equal employment is an important issue that needs to be attained (SOU, 2005:56). This however is not reflected in the labor market where people with non-swedish background have a lower level of employment. The difference between level of occupation for people with Swedish background and people with non-swedish background is estimated to as much as 20% (Statistiska centralbyrån [SCB], 2007). Psychology of the mind; Heuristics and biases How is it that discrimination takes place? Everyday we are faced with difficult tasks that require careful judgment and decision-making and we do not always have enough resources, hence we employ different heuristics; mechanisms in our brain that we use as mental shortcuts to facilitate everyday choices and decisions (Tversky & Kahneman, 1973). These heuristics are useful to some degree, but are dangerous at times as they may result in systematic errors (Tversky & Kahneman, 1974). When we make decisions we are not always as rational as we would believe ourselves to be. We do not always weigh the pros and cons before making decisions, and many times our assessment of certain events are incorrect or biased (Zajnoc, 1980). A part from these heuristics, we also tend to have different cognitive biases, which are flaws in our judgment (Haselton, Nettle & Andrews, 2005). How do employers judge and evaluate applicants in a hiring process? This knowledge of how people function may shed some light on the subject and explain why discrimination can take place. Confirmation bias This is the process of seeking out evidence and information that supports ones belief, and selectively finding and molding information to fit into that belief (Nickerson, 1998). 5 Although this is a less deliberate and less explicit process, so when people use this bias (e.g. looking for and collecting evidence to argue for a certain hypothesis) they are doing it unawares. Confirmation bias was illustrated cleverly in an experiment where participants were recruited on basis of their attitude on death penalty. Half of the group was in favor and the other half was against it. They were instructed to read two articles, fictional studies on the positive and negative aspects of death penalty, to later evaluate the research studies and assess the credibility of their results. When they were asked about their viewpoints afterwards participants tended to stick to their original attitude despite there being strong support against it. They were good at finding evidence or details that supported their belief, and they disregarded all that contradicted their viewpoint (Lord, Ross & Lepper, 1979). In another study by Wason (1960) participants were presented with a series of three numbers and their task was to identify the rule behind the combination of that series. The series was (2,4,6) and participants were able to create their own series using the same rule and the experiment leader would give feedback on whether or not their series were applicable to the rule. This was a difficult task for participants to accomplish. The actual rule was as simple as any ascending sequence however, many identified the rule as each number is triple as its predecessor and they would therefore come up with a series according to that rule. They only tested examples that would fit the rule instead of a series that would disconfirm the rule (e.g. 3, 7, 4). This, Wason argued, goes to show that we tend to seek confirmation over falsification. The affect heuristic Affect plays an important role in our reasoning on our daily issues (Zajonc, 1980). Our judgment and decision-making are heavily influenced by our emotions to different stimuli. This is an automatic process that functions as a guide to the processing and judgment of stimuli, that comes afterwards. As Zajnoc so nicely put it We buy the cars we like , chose the jobs and houses we find attractive and then justify it by various reasons . The affect heuristic was illustrated in a study where participants were instructed to give their opinions on different techniques used within different areas, such as cars and chemical industries with the task to provide pros and cons to each technique. The results showed an inverse correlation between the perceived risk and benefit respectively that are associated with the techniques. When people had a positive opinion of a specific technique they tended to rate it as more beneficial and less harmful or risky. And when they had a more negative outlook on a specific technique they tended to be more resourceful on risks and less on benefits (Finucane, Alhakami, Slovic & Johnson, 2000). Another phase of this study saw the participants read short statements with arguments for each technique. Some had more focus on the benefits of certain techniques, while others focused on the low risks of those same techniques. This resulted in having the participants change their viewpoint on the risks of a technique after reading a paper that acclaimed the benefits of that technique, and vice versa. Availability heuristic When trying to estimate the frequency of classes or probability of events we tend to use a mental shortcut based on availability of information of a certain event, and on the degree of access we have to that information. In other words, events or classes that are easier to remember also become more likely to happen according to our judgment (Tversky & Kahneman, 1973). A good illustration of this heuristic was when people were asked to estimate the most common causes of death. The majority of people overestimated the likelihood of typically airplane accidents and underestimated diabetes as a common cause of death, even though it is the other way around. This can be explained by the fact that for instance media tend to report events that are more chocking more frequently and more intensely. Hence, we usually hear more about airplane accidents, murders and fires etc. and 6 not so much about how people are dying from diabetes or heart diseases. So when this question is asked, people can recall these big events more easily, and because this information spring to mind, it seems to us that it is more common than it actually is, and therefore more likely to occur (Tversky & Kahneman, 1974). In-group bias People seek to benefit members of one's own group, in-group, over members of others group, out-group (Brewer, 1979). Tajfel, Billig, Bundy, & Flament (1971) had participants sit in a room together and presented with an amount of dots on a screen, and later guess the number privately. Based on their guesses they were told that they were either overestimators or underestimators, and that neither was better or worse than the other. They were later given the task to allocate monetary rewards and fines to the other subjects that participated (this was also done privately). Those other participants were identified by a coded number and were labeled as either an overestimator or an underestimator. How they chose to allocate the rewards and punishments was strictly confidential and they would have no contact with the other participants. The results saw participants being strongly biased to other participants with the same label, choosing to benefit participants of their own group rather than participants from the other group . Having shed some light on these different psychological mechanisms that are present in our judgment and decision-making, it might provide a better understanding of how discrimination can occur in different situations, one of them being recruiting processes. Recruiting people is a big responsibility and finding the right person for the job is not always an easy task, but an important one nonetheless. Therefore it is important that these employers can do their job and base their decisions on valid information. There must not be any other factors involved such as biases and heuristics, because clearly, they are not always good. That is why there are different recruiting methods used, each with its pros and cons. Predicting future job performance Resume One of the most commonly used methods in recruiting for entry level jobs is the classical resume which includes detailed information about the applicant, such as education, earlier experience, and also personal details like name, age, sex and sometimes even a photograph (Hutchinson, 1984). There is very little research on the validity of the hiring decisions made by employers when using resumes as a source in judging the eligibility of the applicants (Gatewood & Feild, 2001). The intention of this study was to increase diversity and reduce potential discrimination, and resumes are believed to be a method that gives room for discrimination seeing as it is the first impression applicants will make on the employers (Gatewood & Feild, 2001). Interview The interview as a method has been shown to have a lower predictive validity (0.14) compared to other methods such as mental ability tests, in a meta-analysis by Hunter & Hunter (1984). Predictive validity in this context is measured in terms of correlation between interview ratings and future job performance, according to Cohen s classification; 0.10 to 0.29 equals weak correlation, 0.30 to 0.49 equals average correlation, and 0.50 to 1.0 equals strong correlation (Cohen, 1988). Huffcut & Winfred (1994) responded to these earlier results by referring to subsequent meta-analysis that has shown a much higher predictive validity (0.47) for the interview (Wiesner & Cronshaw, 1988). They also suggested that the structure of the interview had a moderating effect on the predictive validity; more specifically the mean validity was twice as high for the structured interview than that of the unstructured (0.62 and 0.31 respectively). This was in alliance with findings by Marchese and Muchinsky (1993) who presented a correlation between interview structure and predictive validity (0.45). In their 7 meta-analysis of the findings from 85 years of research Hunter & Schmidt (1998) evaluated several recruiting methods and their ability to predict future work performance. The methods that ranked the highest on predictive value were work sample tests, general mental ability tests and structured interviews. According to them, methods that had a lower predictive value for work performance were those that are more sensitive to subjective interpretation, such as unstructured interview. Structured interviews based on a thorough profile for the specific position has a high validity (0.51),
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks