Preferential Selection or Merit Based Selection in Relationship with
Expectancy Confirmation, Self-Efficacy and Performance Feedback
Master thesis, Msc Human Resource Management University of Groningen, Faculty of Economics and Business
December 16th, 2011
“I’m glad I did it, partly because it was worth it, but mostly because I shall never have to do it again.”
~ Mark Twain
Acknowledgement
It took a little longer than expected, but I finally succeeded in the completion of my second Master degree. I wish to thank my parents for supporting and encouraging me unconditionally. They never forced me to complete a second Master and even said that I could stop at any given time. Knowing that they would be proud of me no matter what, has been of great value to me. But most of all I want to thank Sanne Ponsioen for her trust and support. Her uplifting feedback and ever pragmatic
approach were very motivating. I felt confident that I could do this and never hesitated to ask for advice. Although we haven’t been able to prove it in this thesis, I am
ABSTRACT
This thesis examined the influence of preferential and merit selection on self-efficacy and expectancy confirmation. Whereas preferential selection is usually implemented to improve the number or positions of certain group members, merit selection is based on competence hiring. The moderating role of feedback was studied as well. Self-efficacy, expectancy confirmation and performance feedback are rather unexplored concepts in the area of preferential selection. Analyses of variance were estimated with data gathered from an online scenario study (n = 93). Results indicated that people do not score lower on the concepts of self-efficacy or higher on expectancy confirmation when they are selected on a preferential basis as opposed to merit selection. Furthermore, receiving (no) feedback does not moderate this relationship. As such, the anticipated relationships are not demonstrated in this sample.
INTRODUCTION
Traditionally white males have occupied the majority of jobs in all
organizations and industries. Still today, despite many efforts to alter this tendency, there is no such thing as absolute equal opportunity when it comes to employment. Gender, race and disabilities are factors that create enormous boundaries for members of minority groups to get the jobs they hope for. Many attempts have been made to rectify this inequality in the vocational world. However, preferential selection
practices also have their pitfalls. The perception of being selected as part of a program to reduce inequality has many negative results, such as feelings of unhappiness and dissatisfaction (Heilman & Alcott, 2001). But other, less straightforward issues might result from the preferential selection label.
Preferential selection is a controversial subject. It is meant to restore unfair treatment, but may not be all that fair in itself. Furthermore, it seems like it does not always reach its intended goals. How would we feel if we knew or thought that we were selected for a job only because of our race or gender? Would we start
questioning our capabilities? Would we behave differently than if we had assumed to be selected because we were the best in the pool of applicants?
To answer such questions, we want to investigate whether the perception of being selected either on the basis of preferential treatment or on the basis of merit, is related to concepts like expectancy confirmation and self-efficacy. According to Wood and Bandura (1989), self-efficacy is about people’s confidence in their ability to exert the right actions, use the best resources and be sufficiently motivated to attain the desired outcomes. Expectancy confirmation occurs when a perceiver’s
Self-efficacy and expectancy confirmation are well-known topics in the psychological, educational, and organizational sciences. However, self-efficacy and expectancy confirmation in the context of preferential treatment is a rather unexplored area. Researchers often classify self-efficacy as a general trait, rather than a specific state-like concept that can change. As such, self-efficacy is usually investigated as an antecedent instead of an outcome variable. Despite the recent focus on general self-efficacy, the original and more specific definition of self-efficacy as a variable and state-like belief in one’s competence seems interesting in the context of preferential selection (Scherbaum, Cohen-Charash, & Kern, 2006). Since we are interested in the influence of selection on self-efficacy, we choose specific self-efficacy as a dependent variable, because it is much more versatile than other, more lasting definitions of self-efficacy.
Expectancy confirmation is a topic that is often studied through the eyes of the perceiver. That is, many studies focus on the cognitions and behaviors of the perceiver and whether the expectations one has about another individual will be confirmed (Darley, Fleming, Hilton, & Swann, 1988). Less is known however, about how targets process the expectations others have of them and convert them into behavior. We are interested in preferential selection procedures and the consequences of such programs for its beneficiaries. Therefore we want to study expectancy confirmation on the target’s side.
Performance feedback is likely to contribute to an employee’s sense of self-efficacy. Feedback can be a basis for the evaluation of one’s own ability to succeed in future tasks, since it provides insight into past performance (Bandura, 1991).
Furthermore, it has been found that the confirmation of negative expectations
particularly evolves when information about task performance is absent or ambiguous (Swann & Ely, 1984). Therefore the combined influence of selection method and performance feedback on self-efficacy and the expectancy confirmation process seems worth investigating.
We wish to contribute to the existing knowledge of preferential selection by extending the research beyond what is known about the desirable outcomes, to topics in the area of self-image. That is, we aim to discuss how different selection
feedback has any influence on those relationships. In doing so, we will consider self-efficacy as a state. We will attempt to answer the following main research question: “What is the influence of the perceived basis of selection on expectancy confirmation and self-efficacy and what is the role of performance feedback in these
relationships?”
The answer to this main question will be based on a conceptual model of the relationships between selection basis, expectancy confirmation, self-efficacy and performance feedback (figure 1).
THEORETICAL BACKGROUND
In the following sections, the various aspects of interest to this research will be introduced and elaborated. Later on, hypotheses about the possible relationships between the concepts in the model will be presented.
Preferential selection versus merit based selection
Preferential selection generally pertains to initiatives to improve the number or position of certain group members, usually defined by gender or race, within a working environment. Heilman, Block and Stathatos (1997) state that preferential selection programs are acted out to enhance the employment opportunities of ethnic minorities and women. According to Kravitz and Platania (1993), preferential hiring programs should ideally involve weighing both an applicant’s demographic status as well as qualifications in making selection decisions. Such programs are usually not allowed to imply quotas or to push a company to employ or advance incompetent employees, but they may imply the weighing of demographic aspects in hiring decisions (Bennett-Alexander, 1990; Pettigrew & Martin, 1987; Turner, 1990). Particularly, demographic status may be taken into consideration after competence is screened initially (Pettigrew & Martin, 1987). Hence, preferential selection programs have been applied to stop discrimination in the workplace and to battle disadvantage in employment. However, opposition to preferential hiring policies has more and more been expressed by people who should actually benefit most of them (Wilkerson, 1991; Wycliff, 1990). They reason that preferential treatment stigmatizes the
beneficiaries by inducing inferences of incompetence.
considered very important in a selection process, this assumption leads to the
conclusion that the hire is incompetent. This conclusion is based on the reasoning that if the hire in question would actually be competent, he or she wouldn’t have needed preferential treatment to get the position (Pettigrew & Martin, 1987). In addition, in his article on the role of preferential selection, Kravitz (2008) also specifically criticizes one aspect of such practices, namely that it causes onlookers to doubt the qualifications of employees who belong to minority groups. This stigmatization follows from the discounting principle. However, providing clear information about an applicant’s capacity or performance can diminish the stigmatization (Heilman et al., 1997).
So preferential selection entails a decision in which more weight is assigned to one or more non-work related aspects. Merit based selection on the other hand,
implies that capacity and skill were of primary importance to the selection decision. In fact, merit based selection can be seen as the normal procedure for job selection. So those selected on the basis of merit feel that their ability is verified and that they rightfully earned and deserve their position. People who are preferentially selected, have not received any confirmation of their capacity. Therefore, they may be more susceptible to feelings of incompetence (Heilman, Simon, & Repper, 1987).
Although one may conclude that the original initiative of preferential selection was based on sincere intentions to help minorities in finding a job, preferential
selection turns out to have many side effects. An example of a side effect is that beneficiaries often question their own capacity and the reason for their employment.
Expectancy confirmation
A lot of research has been done on how expectancies are likely to create actual behavior in others. One of the most well-known phenomena in this field is the ‘Self-fulfilling prophecy’, first coined by Robert Merton (1948). He described it as follows:
In this statement, Merton clearly describes a process in which an individual expects something from another individual or situation, thereby making that other individual or situation fit the expectancy, even if it was false to begin with (Darley & Fazio, 1980).
Expectancy confirmation is in fact a form of self-fulfilling prophecy. It is a process in which the perceiver’s expectations about a target influence the perceiver's own behavior. When the target responds to that behavior by acting in a way that fits the perceiver's expectations, the expectancies will be confirmed (Darley & Fazio, 1980; Jussim, 1991). In a research by Snyder, Tanke and Berscheid (1977) the behavioral confirmation of stereotypes of attractiveness was investigated. The male perceivers were assigned to interact with a female target that was either attractive or unattractive to them (manipulated by showing the men pictures of attractive or unattractive women, prior to the experiment). It turned out that those men in the ‘attractive picture condition’ actually found their female targets more friendly, warm and attractive than the men in the ‘unattractive picture condition’. Furthermore women who were assigned to men in the ‘attractive picture condition’, actually behaved more friendly and warm, compared to women in the ‘unattractive picture condition’. This was probably caused by the perceiver’s behavior. The male perceivers who expected a warm and friendly female partner, actually started
behaving in that manner themselves. In doing so, they evoked similar behavior in their female partners, thus helping the behavioral confirmation to come about (Snyder et al., 1977).
Expectancy confirmation processes have often been associated with harmful consequences such as stereotyping and discrimination (Copeland, 1993). The majority of studies on expectancy confirmation were conducted with uninformed targets. That is, the targets were unaware of any expectations the perceivers held of them (Hilton & Darley, 1985). But what if a target is actually well aware of the (negative)
expectancies that exist of him or her, because the target is hired for a job on the basis of a preferential selection program in the company?
It is said that when people believe that another person is aware of their stigmatized status, they conclude that the other undervalues them and expects
considers them a beneficiary of preferential selection, are likely to assume that the other expects them to be unqualified. One’s expectancies about a target, even when false, can lead the target to behave in ways that confirm the expectancies. For example, in a research on stereotype threat (Steele & Aronson, 1995), it was found that blacks performed less than whites on several intellectual tasks, when they were in situations where they felt a negative stereotype about their intellectual capacity was present.
Altogether, preferential selection may very well instigate the process of expectancy confirmation. As people are aware of their preferential selection, they might confirm the negative expectancies they assume to be present and start performing sub optimally.
Self-efficacy
People’s performance on their job is strongly determined by the way they think about themselves, in particular about their capacity to cope with challenges. Bandura (1986) described self-efficacy as:
“People’s judgments of their capabilities to organize and execute courses of action required to attain designated types of performances. It is concerned not with the skills one has, but with the judgments of what one can do with whatever skills one
possesses.” (Bandura, 1986: 391).
The concept of self-efficacy has often been studied as a task-related or state-like construct. This is called Specific Self-Efficacy (SSE) (Gist & Mitchell, 1992). Bandura (1986) also referred to self-efficacy as a belief that is determined every time a certain situation comes up (a situation specific belief), rather than a more stable characteristic like personality traits.
Other recent research has focused on people’s perception of their capacity to perform well on tasks across a wide range of situations. This is called General Self-Efficacy (GSE), and can be seen as a situation-independent competence belief. According to several researchers (Eden, 1988; Judge, Locke, & Durham, 1997) SSE can be seen as a motivational state and GSE is more like a motivational trait (Chen, Gully, & Eden, 2001). General self-efficacy has dominated personality psychology, whereas organizational psychologists view self-efficacy as a specific state (Eden & Aviram, 1993).
According to Tipton and Worthington (1984), when people find themselves in unfavorable situations, they are expected to hold specific expectations about how they can handle those circumstances (SSE). But they also bring to the situation some expectations about their ability to manage difficulties in general (GSE).
Although self-efficacy may indeed be stable to some extent, we expect it to be less stable than traits, and therefore susceptible to change. In the current research, the aim is to find out whether the perception of being selected on the basis of merit or preferential treatment affects self-efficacy differently. Given the possible impact of selection basis on the way people view their ability, we treat self-efficacy as a state-like dependent variable.
Self-efficacy versus Expectancy confirmation
Self-efficacy and expectancy confirmation seem to be two overlapping
concepts. However, although interrelated they are distinguishable. One could roughly say that self-efficacy is an intrapersonal concept. This means that it is a phenomenon that resides within a person. Expectancy confirmation on the other hand, is the result of a more interpersonal concept: the confirmation evolves from the interaction between people (Neuberg, 1994), as it requires one individual to have clear expectations about the target with whom he or she interacts.
Research effort shows that self-efficacy and expectancy confirmation are often combined in research, although no conclusive results regarding their connection can be derived. Heilman and Alcott (2001) proposed a mediating effect of self-efficacy on expectancy confirmation, whereas Jussim (1986) suggested that self-efficacy serves as a moderator in the relation between low expectations of others and expectancy
clearly distinguishable and that each of the two concepts is directly influenced by selection.
The influence of selection on expectancy confirmation
Little is known about the influence of employee selection on the expectancy confirmation process. We do know however, that people tend to show behavior that is in line with the (negative) expectancies others have of them.
On the one hand, people who are confident about their ability, because they are selected on the basis of merit do not need a verification of qualifications (Heilman et al., 1987). In addition, people selected on the basis of their merit are more likely to confirm the expectations that belong to the job by performing in line with the
qualifications.
On the other hand, beneficiaries of preferential selection probably know (or infer) that they are selected on the basis of preferential programs. As mentioned above, preferentially selected people have not received any proof of external confidence in their ability. The lack of such verification makes their ability questionable, which may result in feelings of incompetence. Preferential selection makes beneficiaries aware of the stereotypical expectations that exist about them. They are very likely to show behavior that confirms their stigmatized status. As such, we hypothesize:
Hypothesis 1: People who perceive themselves as preferentially selected, show more expectancy confirming behavior than people who think they are selected on the basis of merit.
The influence of selection on self-efficacy
We expect self-efficacy to be a state rather than a trait concept within the context of preferential versus merit selection. As such, job selection serves as the situational factor that instigates lower or higher levels of self-efficacy among people. It seems likely that people who are preferentially selected become less confident about their capacities and competences, leading to lower levels of self-efficacy.
Hypothesis 2: People who perceive themselves as preferentially selected, experience lower levels of self-efficacy than people who think they are selected on the basis of merit.
Performance feedback as a moderator
It is widely accepted that feedback is very important to an employee’s
performance and motivation, as well as to his or her job satisfaction (Kluger & DeNisi, 1996). Furthermore there are signs that feedback reduces the tendency to confirm expectancies and may have a positive influence on self-efficacy.
Feedback is a form of persuasion that can be transferred to recipients in spoken or written form. Feedback usually consists of actions taken by an evaluator, to offer information about someone's performance. For its advantageous outcomes to come about, it is important that feedback is trusted and accepted by the person who receives it (Ilgen, Fisher, & Taylor, 1979). Furthermore, feedback needs to be related to the behavior in question and perceived as fair, in order to be accepted by the employee it was directed towards (McDowall & Fletcher, 2004). According to Rummler and Brache (1995) performance feedback must meet the following criteria: relevance, accuracy, timeliness, specificity and finally it must be easily
understandable. Evidently, positive or negative content of the feedback message was not mentioned as a requirement.
Heilman and Alcott (2001) found that when targets were aware of the fact that others saw them as preferentially selected, they made performance-enhancing
decisions, when they were informed that their task ability level was high. So it seems that when participants receive feedback on their task abilities they are less inclined to be misled and affected by the negative expectations that exist around them. Therefore, based on these results it can be expected that when people are preferentially selected, and receive no feedback at all about their task ability, they will probably confirm the negative expectancies they think exist about them by behaving accordingly. We therefore pose the following hypothesis:
Feedback is said to affect self-efficacy (Martocchio & Webster, 1992). As mentioned before, self-efficacy refers to how a person judges his or her ability to accomplish the desired outcome on specific tasks (Gist & Mitchell, 1992). Heilman and Alcott (2001) found that when targets were aware of the fact that others saw them as preferentially selected, they showed negative self-views of competence when they received no information about their task ability. Based on these results it can be expected that when people are hired on the basis of preferential selection, and receive no feedback at all about their task ability, they will experience lower levels of self-efficacy when compared to merit based selection. People who are selected on the basis of their merit will be less impacted by the moderating influence of feedback, as they already know they have the ability to perform their job related tasks.
This leads to the following hypothesis:
METHOD
Participants & Design
This study was a 2 (preferential vs. merit selection) x 2 (performance feedback vs. no performance feedback) between subjects design. Participants were invited to join the scenario study through e-mail, social media and other online channels. They were provided with a link to the questionnaire. Altogether 93 participants (33 males, 60 females) completed the entire survey. The average age was 32.51 years, and all respondents had the Dutch nationality.
Participants were randomly assigned to one of the four conditions. First, participants were divided into the selection conditions: either preferential selection or merit based selection. Later on, participants were assigned to the feedback condition: either no feedback or feedback.
Procedure
When participants entered the survey, they started by reading a short
introduction, after which they were presented with 5 demographic questions about age, gender, nationality and level of education. These were followed by the manipulation story, to induce the perception of either preferential selection or merit based selection. In the merit condition, participants were led to believe they were approached for this questionnaire on the basis of their competence, because the aim of the study was to develop high quality training methods for cognitive performance, and only the most competent students and professionals were approached to participate in this study. In the preferential selection condition, participants were led to believe they were
approached for this study on the basis of their gender. They were told that the
recruiting efforts for this questionnaire were based on preferential treatment, to restore passed imbalance in cognitive performance research.
message emphasized that we had faith in their ability to perform well on the subsequent task.
Next, they performed a second irrelevant task. In this task, participants were to solve a number of anagrams. They were told that they could fill in an ‘X’ if any of the anagrams was too difficult. First they were presented with 7 Dutch anagrams,
followed by 7 English anagrams. Finally, participants were presented with several items about expectancy confirmation and self-efficacy. The survey ended with a short debriefing in which an explanation of the true rationale behind this research was provided.
Measures
Manipulation check. The selection manipulation was followed by a
manipulation check consisting of 4 items, of which 2 were based on the article by Heilman et al., (1992). The items (1) “To what extent do you think you were selected for this research on the basis of preferential selection?”, and (3) “I was selected for this research on the basis of my gender”, served to become the preferential selection manipulation check. Respondents were to answer on a 7-point Likert scale from ‘completely disagree’ to ‘completely agree’. Cronbach’s alpha for this scale was: .52 The items (2) “How effective are you, in your opinion, when it comes to performing tasks?”, and (4) “I was selected for this research on the basis of my qualifications”, served as a manipulation check for merit based selection. Respondents were to answer on a 7-point Likert scale from ‘completely disagree’ to ‘completely agree’.
Cronbach’s alpha for this scale was: .05.
Expectancy confirmation. After the second task, participants were presented
with a number of items on expectancy confirmation, which were based on an article by Darley et al., (1988). Examples of these items are: “I tend to crumble under pressure”, and “I am able to overcome many challenges”. Respondents were to answer on a 7-point Likert scale from ‘completely disagree’ to ‘completely agree’. Cronbach’s alpha of this scale was: .81
Self-efficacy. This was measured with 6 items from the occupational
RESULTS
Manipulation Check
The manipulation check showed by means of a one way analysis of variance that a condition with and without preferential selection was successfully created. (F(1, 91) = 10.11, p = .01). Participants in the preferential selection condition thought more strongly that they were selected based on preferential selection arguments (M = 3.80, SD = 2.21) than participants in the merit condition (M = 2.56, SD = 1.30).
A one way analysis of variance indicated there was no significant effect of merit based selection on the merit manipulation check items (F(1, 91) = 1.35, p = .25). Participants in the merit condition scored higher on the merit manipulation check items (M = 4.13, SD = .94) than participants in the preferential condition (M = 3.90, SD = 1.03), but this difference was not significant. So respondents in the merit
condition were not more convinced of being selected on the basis of their competence than people in the preferential selection condition.
One Way Analyses of Variance1
Expectancy confirmation. The means and standard deviations are presented in
table 1. Table 2 presents the results of the one way analysis of variance.
--- INSERT TABLE 1 ABOUT HERE ---
To test whether people who perceive themselves as preferentially selected show more expectancy confirming behavior than people who think they are selected on the basis of merit (hypothesis 1), a one way analysis of variance was conducted to compare the effect of selection on expectancy confirmation. Results indicated that although individuals in the preferential condition scored slightly higher on expectancy confirmation than merit selected individuals (table 1), there was no significant
difference between the two groups (F(1, 91) = .36, p = .55). This means that people
1
who were preferentially selected did not show more expectancy confirming behavior than those who were selected on the basis of merit, and therefore hypothesis 1 cannot be confirmed (table 2).
--- INSERT TABLE 2 ABOUT HERE ---
Self-efficacy. The means and standard deviations are presented in table 1. A
one way analysis of variance was conducted to test whether people who perceive themselves as preferentially selected, experience lower levels of self-efficacy than people who think they are selected on the basis of merit (hypothesis 2). Table 3 presents the results of the one way analysis of variance.
--- INSERT TABLE 3 ABOUT HERE ---
Results showed that the difference between the groups in self-efficacy scores was not significant (F(1, 91) = .93, p = .34). This means that preferentially selected individuals did not experience higher levels of self efficacy than individuals selected on the basis of their merit. Hence, hypothesis 2 is rejected.
Two Way Analyses of Variance2
Expectancy confirmation. For the next hypotheses, performance feedback
was added to the analysis. First, a two way analysis of variance was carried out to test whether people who perceive themselves as preferentially selected, show more expectancy confirming behavior than people who think they are selected on the basis of merit, and whether this effect is stronger when they receive no performance feedback as opposed to when they do receive feedback (hypothesis 3).
The results of the two way analysis of variance showed that there was no significant interaction effect between the selection condition and the feedback condition on expectancy confirmation (F(1,89) = .01, p = .92). This implies that
2
performance feedback does not moderate the relationship between selection and expectancy confirmation. Hypothesis 3 can therefore not be confirmed (table 2). To illustrate the nature of the insignificant interaction we plotted the interactions in figure 2.
--- INSERT FIGURE 2 ABOUT HERE ---
Self-efficacy. A two way analysis of variance was conducted to test whether
people who perceive themselves as preferentially selected, experience lower levels of self-efficacy than people who think they are selected on the basis of merit, and
whether this effect is stronger when they receive no performance feedback as opposed to when they do receive feedback (hypothesis 4).
A two way analysis of variance tested the level of self-efficacy of people in the preferential condition and the merit condition, who either received performance feedback or no feedback. Results indicated that there was no interaction effect between the selection condition and the feedback condition on levels of self-efficacy (F(1, 89) = .36, p = .55) (table 3). This means that the relationship between selection and self-efficacy is not moderated by performance feedback, so hypothesis 4 is rejected. To illustrate the nature of the insignificant interaction we plotted the interactions in figure 3.
DISCUSSION
The goal of this thesis was to investigate preferential selection and merit based selection in relationship with expectancy confirmation, self-efficacy and performance feedback as a moderator. The main research question was: “What is the influence of the perceived basis of selection on expectancy confirmation and self-efficacy and what is the role of performance feedback in these relationships?”
Theoretical implications
Based on the notion that the reason for selection can be of major influence to one’s behavior and self-view of competence, preferential selection was expected to impact both expectancy confirmation and self-efficacy.
We posed that people who perceive themselves as preferentially selected, show more expectancy confirming behavior than people who think they are selected on the basis of merit. This was expected, because individuals who are selected on the basis of preferential treatment, tend to assume that other people expect them to be unqualified. That assumption can lead the targets to behave in ways that confirm the expectancies, even if they are false. However, the results of this study do not support the hypothesis.
We also stated that people who perceive themselves as preferentially selected, experience lower levels of self-efficacy than people who think they are selected on the basis of merit. This was hypothesized, because the stigma that follows from
preferential treatment, causes beneficiaries to question their ability to execute tasks and manage difficulties, that is; self-efficacy (Tipton & Worthington, 1984). The results however, did not support the hypothesis.
Presumably participants in the current research were already strongly
convinced of their individual merit and therefore did not question their ability, nor did they experience any uncertainty about their personal competence in the first place. This may provide a partial explanation for the lack of significant results to confirm our hypotheses.
certain tasks. In other words, it may be the case that after an interaction with the perceiver, the target first alters his or her self-perception (according to the
expectations), before starting to behaviorally confirm the expectations. In that case, self-efficacy would be the mediator in the selection - expectancy confirmation relationship.
Alternatively however, a lack of feedback may cause an uncomfortable feeling of not knowing whether one performs properly, forcing the stigmatized target to behave according to the expectations, consequently lowering levels of self-efficacy. If this explanation holds, expectancy confirmation and self-efficacy might form a
relationship in which expectancy confirmation precedes self-efficacy. So then, expectancy confirmation would mediate the relationship between selection and self-efficacy.
Jussim (1986) suggested that individuals’ levels of self-efficacy could moderate their reactions to low expectations. That is, people who have faith in their ability to perform, could decide to prove the negative expectations wrong by raising their effort, resulting in higher performance and hence expectancy disconfirmation. So high self-efficacy could buffer individuals against the negative expectancies of other people.
In the current research we included both self-efficacy and expectancy
confirmation as separate dependent variables, because we were convinced they were distinguishable entities. But it seems they are more connected then we thought. In future research, a different design such as the ones suggested above, should be considered when conducting research on the effects of preferential selection.
Since feedback is a widely used external cue that informs individuals about their abilities in spoken or written form (Gist & Mitchell, 1992), it was expected to moderate the relationship between the before mentioned concepts.
We expected that people who perceive themselves as preferentially selected, would show more expectancy confirming behavior than people who think they are selected on the basis of merit, and that this effect would be stronger when they receive no performance feedback as opposed to when they do receive feedback. This was expected, because feedback provides support and direction in ambiguous
preferential selection stigma. Heilman & Alcott (2001) noted that in situations in which there is little certainty about performance, expectancies are often confirmed. The results however, did not support the hypothesis.
We also expected that people who perceive themselves as preferentially selected, would experience lower levels of self-efficacy than people who think they are selected on the basis of merit, and that this effect would be stronger when they receive no performance feedback as opposed to when they do receive feedback. This was hypothesized, because it has been shown that a lack of feedback can result in negative self-views of competence. The results in the current research however, did not support the hypothesis.
It was anticipated that feedback would alter the relationships between the selection method and self-efficacy and expectancy confirmation, especially among individuals in the preferential conditions, because they were expected to be in particular need of information about their ability. The manipulation checks showed that participants in both selection conditions were equally convinced of being selected on the basis of merit. Providing a feedback message or not, is not likely to change that belief. This may explain why there was no significant interaction effect of feedback and selection on expectancy confirmation and self-efficacy. Alternatively, the lack of the expected interaction effect between feedback and selection in this sample, might be caused by the fact that feedback positively influences performance, regardless of the selection condition. This is endorsed by our finding that feedback influences performance.
Practical implications
Although neither of the hypotheses can be confirmed on the basis of the current sample, this thesis provides some insights into preferential selection and its outcomes.
The finding that beneficiaries of preferential selection do not show more expectancy confirmation than merit selected individuals, implies that the
As we have seen, our merit manipulation did not succeed. This means that people were equally convinced of being selected on the basis of merit, regardless of their selection basis. Managers should be aware that it is useful to emphasize among beneficiaries that, apart from preferential selection, they are also selected on the basis of their merit. According to the attributional framework mentioned by Major,
Feinstein, & Crocker (1994), if a preferential selection procedure makes clear that selection is based on individual merit as well as gender, it should reduce uncertainty about personal competence. So managers should treat and appraise their employees as equally competent, because that confirms their feeling of being hired for their merit.
We found that participants who received feedback scored higher on the second task than those who did not receive feedback. So regardless of the selection condition, feedback improved performance. We also found an interaction effect of gender and feedback on expectancy confirmation. Receiving feedback led women to score higher on expectancy confirmation than men, while receiving no feedback led men to score higher on expectancy confirmation than women (see results). These additional findings can be meaningful in practice. Our finding that feedback results in better performance on certain tasks (in our case the English anagrams), is not necessarily new, but it reaffirms the importance of feedback. It implies that in practice everyone gains from feedback, regardless of their selection basis. However, our second finding knocks down this conclusion. Feedback does not prevent women from behaviorally confirming expectations, so simply giving all employees the same amount of feedback will not yield optimal results for everyone. Instead, managers should adjust the
feedback to every individual employee. In doing so, they should especially distinguish between men and women, since they seem to respond differently to feedback.
Limitations and suggestions for future research
As in every scientific research, there were some limitations in this one. In the following sections, these will be discussed.
vs. 33). Although this is not problematic, a more even representation of both sexes would have been better.
Next, the manipulation checks provided notable results. Although the preferential selection manipulation proved to be significant this was not the case for the merit manipulation. This means that there was no difference between the
conditions regarding their belief about whether they were selected on the basis of their merit. As such, people in the preferential condition were not less convinced of their capacity, so they believed they were selected on the basis of their merit as well. This may have caused a bias in the results. Furthermore, the manipulation check for the merit based selection had a very low reliability. This may explain why that
manipulation was unsuccessful.
It can be noticed that only the selection manipulation was checked and not the feedback manipulation. This omission is a limitation that should be adjusted if this research were to be replicated. Moreover, participants either received no feedback or feedback, and the content of the feedback was (apart from encouraging) rather neutral and not directly related to the performance on the preceding task. The results might have turned out differently if positive or negative feedback would have been provided.
Additionally, expectancy confirmation is a phenomenon that evolves from the interaction between perceiver and target (Neuberg, 1994). Expectancy confirmation as reported by the target only, may be a slightly unilateral view of the concept. However, it would have been more difficult to perform a research in which respondents
participate in ‘perceiver-target-pairs’, since it would have created a barrier for
participation. Besides, a data collection method like that would have been beyond the scope of this research. Although the expectancy confirmation scale may have resulted in a somewhat one-sided representation of the construct, the scale was reliable.
performance, but rather the (un)certainty about the basis of selection in combination with one’s initial self-efficacy. In the current research we haven’t been able to draw such conclusions, since we did not measure initial self-efficacy. We decided not to measure initial self-efficacy because we wanted to keep the survey length below 10 minutes, to keep the threshold for participation low. If we had measured self-efficacy, we would have been able to see how self-efficacy changes as a result of the situational factor selection, by looking at the difference between initial and post experimental measures of efficacy. In that case we would have controlled for trait-like self-efficacy, and would have been able to draw conclusions about pure state self-efficacy
Preferential selection versus merit based selection is a complicated issue in practice. The design of the current research may not resemble the true situation of being selected on either of those grounds. Of course this is an issue in scientific research in general, but it seems to apply here in particular. An investigation in a real organizational setting, in which hires are interviewed in person, or approached to participate in surveys based on their own occupational setting, may result in a better reflection of the real world. A qualitative research could have provided insights that are neglected in the current quantitative research. In the future, more attempts should be made to collect data in the field, in order to get more detailed results.
Although not part of the initial hypotheses the influence of feedback on the anagrams was explored. This yielded one significant result. It turns out those participants who received feedback after the first task, scored higher on the second task, than those who did not receive feedback (see results). This means that regardless of the selection condition, feedback resulted in higher performance. This reaffirms the importance of feedback in organizations, and provides an interesting ground for further investigation.
In a study on the consequences of preferential selection among working women, Heilman (1994) concluded that women need feedback about their competence in order to drive out self-doubts about their performance ability. We explored the data to check whether this was also the case in our sample. We found no interaction for gender and feedback on self-efficacy, but results did indicate that when no feedback is provided, men score higher on expectancy confirmation than women. However, when feedback is provided, women score higher on expectancy
confirmation was significant (see results). This means that in the current sample, feedback has negative results for women, because it makes them behaviorally confirm expectancies. This is a striking finding, which is inconsistent with Heilman’s (1994) results. This is interesting for future research.
Feedback impacts performance, so it is important for organizations. Although feedback improves performance regardless of the selection procedure in the current sample, feedback might still be able to help restore the imbalances that people may feel as a result of different selection methods. But this suggestion is still to be investigated further.
The interaction effect of feedback and gender on expectancy confirmation, demonstrated that feedback has negative results for women, because it increases their tendency to confirm expectancies. However, the feedback that was given to
participants here, was not clearly positive or negative. Moreover, the feedback was not specifically related to their performance on the experimental task. A different pattern of results might be found if clearly directed and positive (vs. negative) feedback are administered to preferentially (vs. merit) selected individuals in an experiment. It seems likely that a positive feedback message that is directly related to the performance in question, diminishes the tendency to confirm expectancies. But that is yet to be demonstrated in future research.
Altogether it is clear that more attention to feedback in selection research is needed
Conclusion
The present research investigated the effects of job selection on individual’s self-efficacy and expectancy confirmation. In addition, we wanted to study the role of performance feedback as a potential moderator in that relationship. The intention was to find out whether people score lower on self-efficacy and higher on expectancy confirmation when they are preferentially selected as opposed to when they are selected on the basis of merit. We found that individuals are not affected by their selection basis when it comes to self-efficacy and expectancy confirmation. Moreover, this result is not different when individuals do or do not receive feedback.
which means that feedback results in higher levels of expectancy confirmation among women in comparison with men. This is striking because women in particular are expected to benefit from feedback, especially in gender-based preferential selection situations.
REFERENCES
Bandura, A. (1986). Social foundations of thought and action: A social cognitive
theory. Englewood Cliffs, NJ: Prentice Hall.
Bandura, A. (1991). Social cognitive theory of self-regulation. Organizational
Behavior and Human Decision Processes, 50: 248-287.
Bennett-Alexander, D. D. (1990). The state of Affirmative Action in employment: A post-Stotts retrospective. American Business Law Journal, 27: 565-597.
Brown, I., & Inouye, D. K. (1978) Learned helplessness through modeling: The role of perceived similarity in competence. Journal of Personality and Social Psychology, 36: 900-908.
Brutus, S., & Ryan, A. (1998). A new perspective on preferential treatment: The role of ambiguity and self-efficacy. Journal of Business and Psychology, 13(2): 157-178.
Chen, G., Gully, S. M., & Eden, D. (2001). Validation of a New General Self-Efficacy Scale. Organizational Research Methods, 4(1): 62-83.
Copeland, J. T. (1993). Motivational approaches to expectancy confirmation. Current
Directions in Psychological Science, 2(4): 117-121.
Darley, J. M., & Fazio, R. H. (1980). Expectancy confirmation processes arising in the social interaction sequence. American Psychologist, 35(10): 867-881.
Darley, J. M., Fleming, J. H., Hilton, J. L., & Swann, W. B. (1988). Dispelling
Negative Expectancies: The impact of Interaction Goals and Target Characteristics on the Expectancy Confirmation Process. Journal of Experimental Social Psychology, 24: 19-36.
Eden, D., & Aviram, A. (1993). Self-Efficacy Training to Speed Reemployment: Helping People to Help Themselves. Journal of Applied Psychology, 78(3): 352-360.
Gist, M. E., & Mitchell, T. R. (1992). Self-efficacy: A theoretical analysis of its determinants and malleability. Academy of Management Review, 17: 183-211.
Gist, M. E., Schwoerer, C., & Rosen, B. (1989). Effects of Alternative Training Methods on Self-Efficacy and Performance in Computer Software Training. Journal
of Applied Psychology, 74(6): 884-891.
Goffman, E. (1963). Stigma: Notes on the management of spoiled identity. Englewood Cliffs, NJ: Prentice Hall.
Harris, M. J., Lightner, R. M., & Manolis, C. (1998). Awareness of Power as a Moderator of Expectancy Confirmation: Who’s the Boss Around Here? Basic and
Applied Social Psychology, 20(3): 220-229.
Heilman, M. E. (1994). Affirmative action: Some unintended consequences for working women. In L. L. Cummings & B. M. Staw (Eds.), Research in
organizational behavior, vol. 16: 125-169. Greenwich, CT: JAI.
Heilman, M. E., & Alcott, V. B. (2001). What I Think You Think of Me: Women's Reaction to Being Viewed as Beneficiaries of Preferential Selection. Journal of
Applied Psychology, 86(4): 574-582.
Heilman, M. E., Block, C. J., & Lucas, J. A. (1992). Presumed incompetent?: Stigmatization and Affirmative Action efforts. Journal of Applied Psychology, 77: 536-544.
Heilman, M. E., Block, C. J., & Stathatos, P. (1997). The Affirmative Action stigma of incompetence: Effects of performance information ambiguity. Academy of
Heilman, M. E., Simon, M. C., & Repper, D. P. (1987). Intentionally Favoured, Unintentionally Harmed? Impact of Sex-Based Preferential Selection on Self-Perceptions and Self-Evaluations. Journal of Applied Psychology, 72(1): 62-68.
Hilton, J. L., & Darley, J. M. (1985). Constructing other persons: A limit on the effect.
Journal of Experimental Social Psychology, 21(1): 1-18.
Ilgen, D. R., Fisher, C. D., & Taylor, M. S. (1979). Consequences of individual feedback on behavior in organizations. Journal of Applied Psychology, 64: 349-371.
Judge, T. A., Locke, E. A., & Durham, C. C. (1997). The dispositional causes of job satisfaction: A core evaluations approach. Research in Organizational Behavior, 19: 151-188.
Jussim, L. (1986). Self-fulfilling prophecies: A theoretical and integrative review.
Psychological Review, 93: 429-445.
Jussim, L. (1991). Social perception and social reality: A reflection-construction model. Psychological Review, 98: 54–73.
Kelley, H. H. (1972). Attribution in social interaction. In E. E. Jones, D. E. Kanouse, H. H. Kanouse, H. H. Kelley, R. E. Nibett, S. Valins, & B. Weiner. (Eds.), Attribution:
Perceiving the causes of behavior. Morristown, NJ: General Learning Press.
Kelley, H. H., & Michela, J. L. (1980). Attribution theory and research. In M.
Rosenzweig, & L. Pater. (Eds.), Annual review of psychology, vol. 31. Palo Alto, CA: Annual Reviews.
Kluger, A. N., & DeNisi, A. (1996). The effects of feedback interventions on performance: a historical review, a meta-analysis, and a preliminary feedback intervention theory. Psychological Bulletin, 119: 254-284.
Kravitz, D. A., & Platania, J. (1993). Attitudes and Beliefs About Affirmative Action: Effects of Target and of Respondent Sex and Ethnicity. Journal of Applied
Psychology, 78(6): 928-938.
Major, B., Feinstein, J., & Crocker, J. (1994). Attributional ambiguity of Affirmative Action. Basic and Applied Social Psychology, 15: 113–141.
Martocchio, J. J., & Webster, J. (1992). Effects of feedback and cognitive playfulness on performance in microcomputer software training. Personnel Psychology, 45: 553-578.
McDowall, A., & Fletcher, C. (2004). Employee development: An organizational justice perspective. Personnel Review, 33: 8-29.
Merton, R. K. (1948). The bearing of empirical research upon the development of social theory. American Sociological Review, 13: 505-515.
Merton, R. K. (1957). Social theory and social structure (Rev. ed.). New York, NY US: Free Press.
Neuberg, S. L. (1994). Expectancy-confirmation processes in stereotype-tinged social encounters: The moderating role of social goals. In M. P. Zanna, & J. M. Olson (Eds.),
The psychology of prejudice: The Ontario symposium, vol. 7: 103-130.
Pettigrew, T. F., & Martin, J. (1987). Shaping the organizational context for Black American inclusion. Journal of Social Issues, 43: 41-78.
Rigotti, T., Schyns, B., & Mohr, G. (2008). A Short Version of the Occupational Self-Efficacy Scale: Structural and Construct Validity Across Five Countries. Journal of
Career Assessment, 16(2): 238-255.
Rummler, G. A., & Brache, A.P. (1995). Improving performance: How to manage
Scherbaum, C. A., Cohen-Charash, Y., & Kern, M. J. (2006). Measuring general self-efficacy: A comparison of three measures using IRT. Educational and Psychological
Measurement, 66: 1047-1063.
Snyder, M., Tanke, E. D., & Berscheid, E. (1977). Social perception and interpersonal behavior: On the self-fulfilling nature of social stereotypes. Journal of Personality
and Social Psychology, 35(9): 656-666.
Steele, C. M., & Aronson, J. (1995). Stereotype threat and the intellectual test performance of African Americans. Journal of Personality and Social Psychology, 69: 797-811.
Swann, W. B., & Ely, R. J. (1984). A Battle of Wills: Self-Verification versus Behavioral Confirmation. Journal of Personality and Social Psychology, 46(6): 1287-1302.
Tipton, R. M., & Worthington, E. L. (1984). The measurement of Generalized Self-Efficacy: A study of Construct Validity. Journal of Personality Assessment, 48(5): 545-548.
Turner, R. (1990). The past and future of Affirmative Action. New York: Quorum Books.
Wilkerson, I. (1991). A remedy for old racism has a new kind of shackles. New York
Times, first section, p. 1.
Wood R., & Bandura, A. (1989). Social cognitive theory of organizational management. Academy of Management Review, 14: 361-384.
APPENDIX
Figure 1: Conceptual model of the Relationships between Selection, Expectancy Confirmation, Self-Efficacy and Performance Feedback.
Table 1: Descriptive Statistics for each of the Experimental Conditions on Expectancy Confirmation, and Self-Efficacy.
Feedback No Yes Expectancy confirmation M (SD) M (SD) Preferential 3.42 (1.03) 3.22 (.68) Selection Merit 3.29 (.84) 3.12 (.90) Feedback No Yes Self-Efficacy M (SD) M (SD) Preferential 3.98 (1.06) 4.18 (1.25) Selection Merit 3.63 (.87) 4.11 (1.11) Note: n = 93 Preferential selection
Merit based selection
Expectancy confirmation
Self-efficacy
Table 2: One- and Two Way Analyses of Variance for Expectancy Confirmation
Source df MS F p
Selection (one way anova) 1 .26 .36 .55
Error 91 .74
Selection (two way anova) 1 .31 .42 .52
Feedback 1 .78 1.05 .31
Selection x feedback 1 .01 .01 .92
Error 89 .75
Table 3: One- and Two Way Analysis of Variance for Self-Efficacy
Source df MS F p
Selection (one way anova) 1 1.12 .93 .34
Error 91 1.20
Selection (two way anova) 1 .98 .82 .37
Feedback 1 2.59 2.16 .15 Selection x feedback 1 .43 .36 .55 Error 89 1.20 Interaction 2,5 3 3,5 4 4,5 5 5,5 6 Preferential Merit Selection M e a n Ex p e c ta n c y C o n fi r m a ti o n No feedback Feedback
Interaction 2,5 3 3,5 4 4,5 5 5,5 6 Preferential Merit Selection M e a n S e lf-Ef fi c a c y No feedback Feedback
