A candidate perspective on personality testing in the selection process: The use of strategies and criteria for a positive experience

Abstract

The aim was to explore how job applicants use strategies when completing a personality test in a selection process, and further to identify factors that contribute to a positive test experience. The informants were eight former applicants for an actual job including management responsibilities. The applicants had completed the personality test OPQ32_i. We used semi-structured interviews when collecting data, and thematic analysis was used as an analytic approach.. The candidates considered strategies as important during the testing to optimize their own test profile. Strategies used were related to strategic honesty, time use, consistency, and aiming not to obtain middle or extreme scores. Criteria for an optimal test experience were positive expectations for the test, previous experiences with personality testing, high belief in the test, perception of the test as fair and relevant for the job, as well as certain aspects related to the feedback interview. Our study has provided important knowledge about the subjective experience of a personality test situation in a selection process. The use of strategies should be seen in a more holistic perspective than only focusing on faking. The results address the importance of using the feedback interview as an integrated and central element when applying personality tests in personnel selection.

Keywords:

• Personality testing
• candidate perspective
• strategic honesty
• test experience
• qualitative study
• ipsative test
• fair test

PUBLIC INTEREST STATEMENT

In this research we investigated broadly candidates´ experiences with taking a personality test in an actual selection setting for a management position. We identified that candidates use strategies, but these are not specifically aimed at faking to obtain a best possible result. The candidates strives for a good result, but are still motivated to be honest and true to themselves. It is important for candidates to experience the test as fair and necessary and to be able to receive and participate actively in a feedback interview.

1. Introduction

Applicant reactions refer to how candidates perceive and respond to different selection tools based on their own experience (McCarthy et al., 2017). How the employees experience the selection process is related to their intention of recommending the company to others or accepting the job offer (Hausknecht et al., 2004), as well as better work performance (McCarthy et al., 2017). If an organization fails to focus on the experience of the applicant, it can risk losing qualified candidates. This may also lead to a negative start of the employment (Konradt et al., 2017). Personality tests are often used in personnel selection as a means for improving the accuracy of the decision to hire or not hire. Several meta-studies have shown that personality traits can predict later work outcomes, such as work performance (Barrick & Mount, 1991; Judge & Bono, 2001; McCarthy et al., 2017; Salgado, 1997; Schmidt & Hunter, 1998). How the applicants use conscious and unconscious strategies when completing the personality test is important not only for the applicant reactions, but also for the success of the selection process. Faking can be one such strategy, and is often applied when completing a personality test for selection purposes (Birkeland et al., 2006; Boss et al., 2005; Donovan et al., 2003; Griffith et al., 2007; Salgado, 2016; Tett & Christiansen, 2007). Still, quantitative studies of faking behavior in personality testing are criticized for failing in capturing all important facets of this behavior (König, Merz et al., 2012). Moreover, there is a need for research to explore the strategies of applicants more broadly, and limit the focus to one strategy, e.g., faking. The aim of the present research is to explore how candidates experience the selection process, their use of strategies when completing personality tests, and to understand what contributes to a positive test experience for the candidates.

1.1. Personality tests-the OPQ32i

Research on personality and selection is often based in a trait theory perspective (e.g., eks. Barrick et al., 2001; Barrick & Mount, 1991; Hurtz & Donovan, 2000; Judge & Zapata, 2015; Salgado, 1997; Salgado et al., 2015; Schmidt & Hunter, 1998). In the last 40 to 50 years, the five-factor model of personality (FFM; Costa and McCrae, 1992) has received a dominant status in trait theory and personality research. The model includes five central personality traits: neuroticism, extraversion, openness to experience, agreeableness, and conscientiousness. Several personality tests used in personnel selection are based on the FFM. In the present study, we investigate the candidate perspective of completing the personality test Occupational Personality Questionnaire (OPQ32_i). The OPQ32 is partly developed from the Sixteen Personality Factor Questionnaire (16PF), but it is also inspired by the FFM theory and can measure the five personality traits of the FFM (Saville, 2016). OPQ32 measure 32 facets of personality that are relevant to occupational uses such as selection, promotion, counselling, development and team building. The OPQ32 _n has a normative format, while the OPQ32i has an ipsative format with forced-choice units. The candidates are presented with four options and asked to choose the options that are the most and least like themselves. The test consists of 104 collections of options. For each block respondents have to choose one item that is “Most like me” and one “Least like me”. Here is an example of a block: A: I like to do things my own way, B: I recognize weak arguments, C: I take care to follow procedures, and D: I like to spend time with others (Brown & Bartram, 2013). This version of the test was developed to control for faking and is shown to be less affected by faking than the OPQ32 _n (Bowen et al., 2002). The test features consistency scores, while only the normative format includes a scale for detecting social desirability.

1.2. Applicant reactions to the selection process

Research on applicant reactions started in the 1980 s and focused on attitudes, affect, and cognition related to the selection process. The organizational justice perspective by Gilliland (1993) is highly relevant, as this concept proposes that 10 different procedural rules of justice decide the applicant reaction to the selection process. The following five of the rules may be considered especially relevant to how the applicants perceive their own experience with personality testing: job relatedness, whether the test is perceived as valid and relevant for the job; opportunity to perform, to what degree the candidate experiences that he or she can show his or her knowledge in the test situation; reconsideration opportunity, which comprises the experience of possibilities to evaluate and reevaluate the decisions made in the test; feedback, whether the feedback from the test is perceived as informative and succinct; and two-way communication, regarding whether the candidate feels that he or she can be heard and contribute with his or her own opinions.

Much research has been performed on applicant reactions to the selection process, for example, in terms of investigating how the selection process can affect confidence (Hausknecht et al., 2004), what selection methods are most popular among applicants (Nikolaou et al., 2015), and what types of personality tests are preferred among applicants (Bowen et al., 2002). However, less research has been performed concerning how candidates evaluate a selection method based on their ability to use strategies when completing a test.

1.3. The test experience

Previous studies have found that personality testing attain average scores compared to other selection methods when evaluated by the applicants (Anderson et al., 2010; Gilliland & Steiner, 2012). Other studies have reported that the applicants in general are negative to the use of personality tests in a selection process (Hausknecht et al., 2004). Performing a feedback interview concerning the test results (Coaley, 2010), and a perception of the test having high job relevance (Mabon, 2004) can contribute to a positive test experience. However, the selection process can also affect self-confidence negatively (Hausknecht et al., 2004), which can lead to a negative test experience. If one goal is for applicants to have a positive candidate experience, it is important to figure out how the test is experienced by the candidates and what factors contribute to a positive test experience.

1.4. Use of strategies

Candidates may use different types of implicit and conscious strategies when completing the test. One of the strategies that are most often investigated is faking. Faking a personality test in a selection process is common (Birkeland et al., 2006; Boss et al., 2005; Donovan et al., 2003; Griffith et al., 2007; Salgado, 2016; Tett & Christiansen, 2007). Faking occurs when a candidate deliberately tries to change his or her answers in accordance with what he or she believes will be an ideal personality for the specific job. In a meta-analysis, it was found that about 30% (± 10%) fake their scores on different selection tools, including personality tests (Griffith & Converse, 2011). Faking is higher if the candidates believe that faking is necessary to attain the job (Ellingson & McFarland, 2011), or if the candidate can identify the criteria the test is assessing (König et al., 2006; Marcus, 2009). In addition, taking a long time to complete the test is related to faking (Fine & Pirak, 2015; Van Hooft & Born, 2012). Faking can affect criteria validity (e.g., Griffith & Converse, 2011; Jackson et al., 2000) and change the ranking of the candidates (f. eks. McFarland & Ryan, 2000). However, faking does not have to be negative. It has been argued that faking is just normal social behavior (Hogan et al., 2007), and that light faking can be socially adaptive (Marcus, 2009). Furthermore, the process of faking and thinking about faking is not similar across individuals (Ziegler, 2011).

Qualitative studies have found that some candidates fake to a lesser extent, while others fake substantially (König, Merz et al., 2012; Pavlov et al., 2019; Robie et al., 2007). König, Merz et al. (2012) argued that faking is an adaptive process. However, when completing a personality test, candidates in general try to be honest because they suppose that the truth will prevail anyway, they are afraid of being discovered, and they believe that honesty is best for all parts. Some studies have also reported that candidates choose to answer more moderately on personality tests when asked to present themselves in a positive way, because of expectations about the evaluation of their profiles (Kuncel & Tellegen, 2009). When candidates were asked to fake, they tended to answer like they would have normally, albeit with somewhat higher scores (Ziegler, 2011). Although these studies are directly focused on faking, applicants may use different strategies when completing a personality test, strategies related to faking, or other strategies. It is important for researchers to approach this topic broadly, and not specifically look for faking behavior, as this may cloud other strategies. It is also important to explore how applicants experience personality testing in a selection process, both the actual completion of the test and the process as a whole.

There is a lack of qualitative research investigating what contributes to a positive test experience and focusing on strategies, besides faking, applied by the candidate when completing a personality test. In the present study, we use an explorative qualitative research design in studying applicants’ reactions and strategies used when completing the personality test OPQ32_i.

2. Method

2.1. Informants

Applicants who had gone through a selection process in the last six months that included completing the OPQ32_i were recruited to this qualitative interview-based study. We aimed to include a varied sample in terms of age and gender. All informants had applied for a job within management. The informants were recruited through a firm that used the OPQ32_i as a standard procedure in its selection process. The firm gave potential informants the contact information of our researcher in charge of the data collection (author IE), who then contacted them for further information about the study. The recruitment firm then had no further role in the data collection or the study. Eight research interviews were performed with eligible informants, and we ended the data collection after the eighth interview due to theoretical saturation (Tuckett, 2004). The informants were between 35 and 55 years of age, and all except one had performed a personality test in a previous selection process. However, none of them had completed the OPQ32_i before.

2.2. Data collection

Data was collected through semi-structured individual interviews. Following a pilot interview, the guide was slightly revised, and the final interview guide consisted of six main themes: 1) About the test in general, 2) Preparations, 3) Completing the test, 4) Test results and feedback, 5) Honesty versus strategies and techniques used in completing the test, and 6) Experience of the process and the company. All interviews were performed by the second author in a face-to-face situation in September 2016. The interviews lasted from 45 minutes to about 60 minutes and were performed in the recruitment firm´s offices, at the workplace of the informant, or at NTNU in accordance with the informants’ preferences for location. Before the interview, all participants received information about the study and their ethical rights to withdraw their participation at any time, and they signed a participation confirmation. All interviews were recorded, and the interviewer took notes during the interview. The data collection was approved by the Norwegian Data Protection Official for Research (NSD).

2.3. Transcription and analysis

Recordings were transcribed in a detailed playwright format. All speech and sounds were included. The data were analyzed with thematic analysis, including getting to know the data, generating codes, searching for themes, going through the themes, defining and giving names to themes, and producing a written report (Braun & Clarke, 2006). The analysis was assisted by the computer aided qualitative data software program NVivo (NVivo, version 10), and a total of 204 distinct codes were identified. During this process, previous literature related to themes was reviewed. The names of all informants were changed due to anonymity.

3. Results

The analysis of the transcripts resulted in seven main themes: (1) An inherent part of the process, (2) Strategic optimization of the profile, (3) The format and the process limit use of strategies. In regard to criteria for a positive test experience, the following themes emerged: (4) Being an active part of the process (5) Confidence in the test important for perceived fairness, (6) Experienced personal gain through increased self-knowledge, and (7) Test is accepted due to perceived job relevance. Strategic optimization of the profile included four sub-categories, as shown in Figure 1.

Figure 1. Thematic map over the use of strategies and criteria for a positive test experience.

3.1. Personality testing as an inherent part of a selection process

All informants had expected a personality test to be included in the selection process from the beginning. According to one of the informants:

“I was absolutely not surprised. It was something I expected. I had already completed a test like this at another occasion. It is what everyone does in a selection process.”—Henry

This quote illustrates something several of the informants communicated: they expected a personality test based on previous experiences. This quote echoes the other participants.

The informants also expected a personality test as a natural part of the selection process as a statement of the professionalism of the process:

“I expected it. It has kind of been (.) I would have been more surprised if there was no test, than that the test was presented, to say it that way. [I would have] almost been skeptical, [wondering] if they were not serious.”—Elise

Therefore, this natural expectation of the test led to some of the informants not being so nervous about taking the test.

3.2. Strategic optimization of the profile

The informants reported that they did not aim to change the test results drastically, but that they tried to do simple improvements, but still be honest and true to themselves. Many of them used words indicating that they tried to optimize the result. They mentioned: “You always strive to get the best result.”—Henry “The goal is to get a good profile.”—Elise

Their goal about optimizing the test results was related to a desire about getting the job:

“I want to get the job, so I have to try to make that happen.”—Robert

There were several ways the informants talked about strategies related to the test. Firstly, their experience indicated a contrasting approach to honesty, both as a strategy to optimize the test results and as a moral intensive. They viewed honesty as an important ideal. The informants communicated: “I have to act in line with my basic values, the way I really am” (Eric) and, “It would be completely wrong of me as a human being to not be honest” (Linda). They also talked about how it occurred to them during the test that honesty may be the best option, also strategically:

“I tried to be honest because as I filled out the test, I realized that this was the best strategy. Because it is almost impossible to affect it, one can just answer in line with his or her own guts.”—Linda

Secondly, the informants were concerned about the time limit for the test. Although there was no real time limit for completing the test, they were informed in writing before starting the test that it normally took about 45 to 60 minutes to complete. However, most of the informants perceived this as a requirement. The informants said: “It was important for me to know that I was not too far from the average time” (Karoline), “I speeded up and made some mistakes, but I finished within the time limit” (Robert), and “It is hard to answer correctly when it is supposed to happen so quickly” (Martin).

Thirdly, seven out of eight of the informants talked about wanting to avoid average scoring, for example:

“It is maybe in line with the way I am. I don’t need to be the best, but at least I will not be average or invisible. So, that is maybe why I had a clear answer, but not extreme.”—Karoline

Forth, the informants wanted to appear consistent; that is, they were concerned about giving similar answers to similar questions. An example of this can be illustrated in the following quote:

“(…) after you notice what you have answered before, you feel like you need to adjust these answers so they are better aligned. I would like to score something high that I had previously scored high on the test.”—Henry

The informants also talked about the specific measure of consistency, which they were told about in the feedback interview. Some of the informants valued the score on this consistence measure over presenting themselves the way they actually are. Others, however, felt like they had to be true to themselves regardless of the consistency, although they expressed uncertainty about how this would look in the consistence measure.

3.3. The format and process limit the use of strategies

There were several aspects that limited the use of strategies. The informants communicated that the way the test was constructed limited their use of strategies. One of the informants said:

“I thought about having some sort of plan for taking the test. But I am really bad at things like that. And when I started the test, I realized that I was better off without a plan. Because it was not possible, with all the items being placed all around, here and there.”—Andrea

Nevertheless, two of the informants communicated that the format contributed to them being more strategic. Being repeatedly exposed to the same dilemmas made them think more about their answers before responding. Furthermore, almost all informants reported that they felt that the feedback interview could disclose their strategies.

3.4. Candidate agency

The participants communicated that they wanted to actively participate in the feedback interview to gain control over the perception of their scores and their personality profile. They wanted the opportunity to correct misperceptions that may have come from the test. Example of quotes illustrating this are: “Without the feedback, they may have gotten the wrong impression of me” (Andrea), “Most of it was correct, and then I got to make some corrections in the feedback interview” (Eric), and “But I mostly felt that I agreed, apart from a couple of things that I felt I had to give a more in depth answer to” (Karoline). Further, the informants believed that their comments would be considered by the test administrator in the end evaluation.

3.5. Confidence in the test important for perceived fairness

There was a consensus among the informants that the test was serious and that the testing had been conducted correctly. The informants generally had a high trust in the test. They communicated: “It was a serious test, and everything was done by the book, so …, then it is fair” (Robert), and “This type of testing is a clever way to get a more correct result” (Henry). The perception about a fair test was connected to the feedback interview. One of the informants said:

“And I have to say, if I were to order a test like that, and it was not evaluated together with the candidate to exclude obvious errors, I would have trusted it less.”—Henry

3.6. Personal gain through increased self-knowledge

The informants experienced that taking the test was a process of learning that could result in increased self-knowledge. All of them claimed that the test-results reflected who they were, but also that there were some surprising results. One of the informants expressed:

“And I noticed that the test in many ways reflected the way I know myself. I know myself pretty well. But there were also some nuances appearing. And one thinks like, ‘Ah, that’s true, maybe I am a little like that.’ I learned something about myself in a way.”—Martin

Receiving unexpected results and gaining self-knowledge were expressed as somewhat difficult, but still valuable:

“And sometimes it can be a bit painful, a bit difficult to process this type of feedback (…). But if I choose to look at it constructively, I will grow and learn from it.” —Linda

3.7. Perceived job relevance

The informants experienced that the test was relevant for the managing job for which they were applying:

“The test is relevant for a management position. It is important to check that we have the personality they are looking for.”—Martin

However, the experience of relevance of the single items were lower:

“And I don’t see the relevance of each question. But they would not have been asked if it was not relevant.”—Andrea

“Some of it was relevant, and some was not. Maybe not super relevant, really, but I trust that the results serve a purpose. That’s what I felt, at least.”—Eric

Nevertheless, the informants considered the test to be professional and relevant, which increased the perception of the test as job-relevant and contributed to acceptance of the test in the selection process.

4. Discussion

The main findings from this study were that the informants used strategies to optimize their test results. The results indicate that different criteria throughout the selection process are important to create a positive test experience: before taking the test (high expectations, high belief in a just test), criteria during the testing (high job relevance and participation) as well as after the test, related to the use of the test (increased self-knowledge).

4.1. Motivation and use of strategies

Although some of the strategies the informants used were similar to faking, some were unrelated to faking, such as, for example, trying not to spend more time on the test than what the information indicated was normal. Almost all the informants claimed that they had to hurry to complete the test. This experience of time pressure can reduce the reliability of the test (Sjöberg, 2010). Using the right amount of time was important enough to the informants that they prioritized this over making time to answer correctly. The reasoning behind this strategy may be that they tried to identify an objective criterion (the set time limit) to make an effort to obtain a high score on said criterion and, hence, gain a feeling of control. This may be used as a compensation for that there are no “right” or “wrong” answers on the test, or it may be a compensation for an experienced loss of other strategy possibilities, or it may be that the informants aimed to answer quickly to camouflage their attempts of using strategies to optimize their results.

The informants were concerned about honesty, which guided their responses. This is in line with the findings by McFarland and Ryan (2000), that applicants think that dishonesty in the selection process is wrong regardless of the situation. It is also supported by research showing that applicants who are motivated to answer honestly often do so because of a moral conviction (Robie et al., 2007). In our study, dishonesty was perceived as wrong and the applicants often responded honestly, although this does not indicate a total absence of dishonesty. In some cases, the aim to optimize the test scores was so strong that the participants did not want to or could not act morally right. However, throughout the test, the informants discovered that honesty also was a tactic to obtain a “good score.” The informants’ motivation to be honest and present themselves as positively as possible could be conflicting. It is possible that this conflict could be reduced by using honesty more strategically to optimize the test profile.

In our study, the informants communicated that they did not want to appear average. This is in line with previous studies showing the same tendencies (Dunlop et al., 2012; König, Merz et al., 2012; König, Wong et al., 2012). Most previous research on this has been executed with personality tests with a normative format. Here, we show that this also occurs when completing personality tests with an ipsative format, even though identifying middle scores are not as easy in ipsative as it is in normative personality tests. The informants also aimed to avoid extreme scores and preferred to attain a score a little above middle to high, in line with findings from Borkenau and Zaltauskau (2009) and Kuncel and Tellegen (2009). However, avoiding middle scores seemed to be more important for the informants than avoiding extreme scores. This may seem like a good strategy, given the research has shown that leaders often have high scores on the personality traits of extraversion, conscientiousness, and openness to experience (Howard & Howard, 2001). The context of an occupation can moderate the response patterns in various tests (Dunlop et al., 2012). Thus, the aim to avoid middle scores may be more apparent in a selection process for a management position, compared to a middle-level position.

The informants also aimed to appear consistent, which was also expressed by König, Merz, et al. (2012). This strategy may lead to some characteristics of the person not being reflected well enough by the personality test, even though that is not the intent of the applicant. A reason for using this strategy is because it is unpleasant to answer in a way that contradicts oneself. This is in line with the cognitive dissonance theory (Festinger, 1957) and the self-consistence theory (Aronson, 1999), in which dissonance is caused by inconsistent thoughts and actions that are not in line with one’s own self-concept. Similar findings have also been reported by Robie et al. (2007), who found that the informants were not able to present themselves inconsistently with their own personality. The importance of a high score on this measure of consistency can be interpreted as the same as the finding of the importance of time spent on the test: it is an objective criterion where the informants can receive a high score.

4.2. Test-format and the feedback interview

The format of the test and the feedback interview shaped the use of strategies for the informants. The forced-choice format limited the informants use of strategies, in line with previous research on ipsative tests with a forced-choice format (f. eks. Bartram, 2007; Bowen et al., 2002; Christiansen et al., 2005; Griffith & McDaniel, 2006; Salgado et al., 2015). The informants did not understand the logic of the test nor, consequently, the tactically best ways to respond, in line with König, Wong et al. (2012). However, for two of the informants in our study, the format contributed to increased use of strategies. They experienced an increased level of reflection when they were exposed to the dilemmas presented in the test multiple times in different items.

The feedback interview influenced the use of strategies by increasing the informants’ motivation to be honest and reliable, consistent with previous research (Coaley, 2010; Toplis et al., 2005). None of the informants had experinced being confronted with their use of strategies in the feedback interview, but rather they had a strong perception that the test could disclose their use of strategies and they did not want to be discovered as dishonest.

4.3. Use of strategies and social competence

Previous research has focused on how faking is negative, contributes to lower criteria validity (e.g., Griffith & Converse, 2011; Jackson et al., 2000) and changes the rating of applicants (e.g., McFarland & Ryan, 2000). However, some studies have also claimed that faking is just normal social behavior (Hogan et al., 2007). It is possible that faking is a natural behavior performed to create a good impression of oneself. Faking does not seem to have a negative effect on job performance, but rather, it can be a necessary skill in a job (Ingold, Kleinmann, Köning, and Melchers, 2015). Our informants’ way of faking may be more in line with the concept of light faking (Marcus, 2009), which is seen as socially adaptive and useful for selection purposes. When performing light faking, in terms of impression management, informants make a compromise that reflects themselves, while extreme faking involves trying to create an ideal profile (Robie et al., 2007). We interpret that our informants more frequently engage in light faking, and that their use of this strategy may not be only negative.

4.4. Factors contributing to a positive test experience

Our informants generally had a positive experience of the personality test. Previous research has documented that personality tests are often experienced as neutral (Gilliland & Steiner, 2012; Steiner & Gilliland, 2001) or even negative by the applicants (Hausknecht et al., 2004). Positive applicant reactions are related to several work outcome, such as better work performance, increased perception of the organization as attractive, and a higher probability of accepting the job (McCarthy et al., 2017). Thus, it is important to more deeply examine the possible explanations for the applicants’ positive test experience in our study, especially since this is one of the few studies investigating applicant reactions to an ipsative personality test. Our study indicates that such explanations may be due to the informants’ previous experiences with personality tests, as their expectations and the belief in the test influence their perception of the test and, thus, the way they completed the test. Since the test was considered to be a natural and expected part of the process, the participants did not view the test negatively. This can be seen in line with the expectancy theory (Bell et al., 2004), emphasizing that expectations are important in the relationship between previous experiences and reactions to the test. In addition, the belief in the test makes the selection process experienced as fair, which can contribute to a positive test experience. Previous research has supported this claim, reporting that belief in tests is related to evaluations of fairness (Lievens et al., 2003), and that selection methods that are perceived as fair are preferred by applicants (Gilliland & Hale, 2005; Weichmann and Ryan, 2003).

The following four test criteria were found in our study contributing to a positive test experience: Firstly, it was important that the informants experienced the test as relevant for the job. This is in line with findings from previous research (Hausknecht et al., 2004; Ryan & Huth, 2008). In our study, the applicants could not identify the job relevance in each test item; however, they did not doubt the job relevance of the test as a whole. Secondly, the informants wanted to participate actively in the process. This was probably mostly granted by the feedback interview. A study by Coaley (2010) also showed that feedback contributed to the experience of the selection process as attractive, and Gilliland (1993) emphasized that feedback is essential for fairness. The feedback interview may have affected the informants experiencing the test positively, as it was dialogue-formed. Moreover, the feedback interview gave the informants a possibility to review or adjust their answers in line with Gilliland’s (1993) procedural rules of justice. Thirdly, the informants’ perception of a usefulness of the test for their own self-awareness may have contributed to a positive test experience. Although this has not been investigated much in previous research, it seems important for the informants. All informants claimed that they changed the way they saw themselves based on the results of the test. Furthermore, Kahlke and Schmidt (2000) assumed that it is possible that applicants can review their own self-perception after the test. However, our study also shows that only small changes from one’s own perceived personality lead to an altered self-perception, not large changes. Forth, it is possible that the way the test allows for the use of strategies can contribute to a positive test experience. The format of the test may contribute to a balanced use of strategies, where the applicants focus on both answering morally as well as optimizing their profile.

4.5. Strengths and limitations

A strength of the present study is that the sample consisted of actual job applicants. Most previous qualitative research has utilized fictional job applicants (Kuncel & Tellegen, 2009; Robie et al., 2007; Ziegler, 2011), with the exception of the study conducted by König et al. (2012), where part of the sample was actual job applicants. We had a sample of eight informants. The data collection was halted due to theoretical saturation, which richened the data material allowed us to perform an in-depth analysis of the data.

Semi-structured research interviews have the advantage of letting the informant say what they want, when they want to, albeit having a structure and ensuring covering of the central topics being explored. The interviews were performed after the selection process was completed, and the time-span after taking the test varied. We also considered interviewing the participants during the selection process, immediately after completing t the test, as previously done by Ziegler (2011). However, we deemed this not applicable in an actual selection process, since it could possibly affect the way the participants answered the test.

A possible limitation with the study is the use of the recruitment firm in the recruitment of informants, as this may have affected the voluntary participation. In addition, using the offices of the recruitment firm may have not been optimal, as it may have led to social desirability-guided answering patterns. However, the use of these offices was practical, especially for participants who did not wish to perform the interview at their previous/current job. Moreover, the informants themselves chose where they wanted the interviews to take place.

As all of the informants were candidates for management position, the perception of a personality test as an natural part of the selection process might not be transferrable to candidates for other, middle-level positions

We made a conscious choice not to ask the participants directly about faking, but used the term “strategies” instead. The terms “lying” and “faking” are problematic as different forms of lying have very different outcomes. However, previous literature has found that applicants admit to a considerable amount of faking when asked directly (Donovan et al., 2003; König, Wong et al., 2012). Hence, we cannot exclude that we may have obtained different results if we asked directly about faking. Nevertheless, this could also have affected our insight into the more nuanced picture of strategy use, such as strategic honesty.

4.6. Practical implications

An important practical implication of our findings is that recruiters using ipsative personality tests need to carefully consider the use of strategies when implementing the test. . We do not know how transferable our results are to personality testing in modern computerized adaptive testing. Although we still believe that the use of strategies to optimize test results will be relevant. Our study revealed that informants use different strategies, even though they had completed an ipsative personality test, which is often used because the use of this type of test is believed to reduce chances of faking compared to using normative tests (jf. Bartram, 2007; Christiansen et al., 2005; Jackson et al., 2000). However, one needs to consider that some possibilities to use strategies may also contribute to a positive test experience. The information about normal time use represented a constraint for the informants, and implications to deal with this needs to be considered. One possibility is to inform the candidates that this time limit is only an estimate used as a guide, still one needs to consider that having some thoughts about a time limit may be positive, as longer time to answer is related to more faking (Fine & Pirak, 2015; Van Hooft & Born, 2012).

In this study, belief in the test due to previous experiences of testing was important and contributed to a positive test experience. This should be considered in practical test applications. It is probably fruitful to ask the applicants about previous test experiences and be prepared to discuss and possibly dampen the negative perceptions. Furthermore, the test administrator should be aware that not all applicants have a high belief in the test and the fairness of the test as our participants.

Lastly, we found that the feedback interview was important in two areas: to make the applicants feel more involved in the process, which lead to a more positive test experience, as well as self-development; and to reduce the use of strategies. Not all selection processes that employ a personality test utilize feedback interviews, but according to our results, this is recommended.

4.7. Conclusion

In this study of applicant reactions to taking a personality test, we found that the overall aim of the applicants was to create good test results. This made complete honesty not possible. There is a need for more qualitative studies investigating the applicant reactions of personality tests, especially those with an ipsative format, as this is one of the first studies to investigate this. In this current study, the ipsative format constrains, but does not exclude, the use of strategies. The findings concerning the applicants aiming to obtain a consistent score and answer within the time limit is important. This has received limited attention previously and needs to be replicated in different samples. In particular, there is a need for research to investigate applicant reactions of personality testing in an ipsative format in a sample consisting of applicants to positions outside of management.

Future research needs to investigate more holistically into the use of strategies, particularly the classic way of investigating and interpreting faking. Previous studies of faking have often considered that participants fake when they choose the highest possible score on personality traits that seem positive for the position (Kuncel & Tellegen, 2009). However, in our study, the informants deliberately tried to avoid the highest scores.

Although our study focused on applicant reactions to an ipsative personality test, several of the findings have general relevance, regardless of the test format. This especially adheres to what constitutes a positive test experience. We found that the criteria that contribute to a positive test experience for the applicants are: perceived quality of the test, experiencing the test as fair, seeing the relevance of the test in the actual position, and, in particular, a two-way communication-based feedback interview. McCarthy et al. (2017) emphasized the need for future research to investigate the complex issue of applicant privacy that protects the employer, but also focuses on the fairness of the selection process, especially in light of possible new selection tools emerging in the market in forms of online interactive tests or neurobiological tests. Although information from big data is important (McCarthy et al., 2017), we would like to argue that insight from qualitative studies is equally important in unraveling the complex issue behind applicant attitudes towards new selection tools. The knowledge from our study can serve as a base for further investigation of applicant reactions to new selection tools, the importance of the feedback interview, and the complex use of strategies in the selection process.

Competing interests

The authors declare no competing interests.

Additional information

Funding

None of the authors received any funding for this work.

Notes on contributors

Ingvild Saksvik-Lehouillier

Saksvik-Lehouillier and Langvik are associate professors in the research group Occupation, psychocardiology and sleep (OPS). Eriksen has been a graduate student affiliated with this group. The research group focuses on multidisciplinary projects related to health psychology, clinical psychology, community psychology, personality psychology and work- and organizational psychology.