How Do Situational Cues Influence Honest and Deceptive Impression Management in Selection Interviews? A Grounded Theory Study

Impression Management

People use IM “to control the impression others form of them” (Leary & Kowalski, 1990, p. 34). To do so in an interview, applicants can use verbal and nonverbal behaviors. In verbal IM, applicants use words to create a certain impression of themselves (e.g., by highlighting their expertise), whereas in nonverbal IM they use, for example, their facial expressions and posture (Bolino et al., 2008). Given that verbal IM is relevant in any kind of selection interview—in-person, videoconference, telephone—and applicants have a high level of control over what they say, we decided to focus on verbal IM in the present study.

As explained by Bourdage et al. (2018), verbal IM can be divided into deceptive IM and honest IM. In both forms, applicants try to present themselves as positively as possible (Levashina & Campion, 2006). In deceptive IM, however, they answer dishonestly whereas in honest IM they remain truthful in their answers (Levashina & Campion, 2006).

According to Levashina and Campion (2007), deceptive IM can be divided into four different tactics—slight image creation, extensive image creation, image protection, and ingratiation—which they described in the following way: Applicants show slight image creation when they exaggerate or adapt the truth. However, if they completely lie about themselves (e.g., about their work experience or skills), this is called extensive image creation. If they intentionally conceal—or distance themselves from—negative work events or their own weaknesses, this is referred to as image protection. Finally, if applicants try to appear as similar as possible to the interviewer by lying or give dishonest compliments to the interviewer or the organization, this is called deceptive ingratiation.

According to Bourdage et al. (2018), honest IM can similarly be divided into three different tactics—self-promotion, honest ingratiation, and honest defensive—which they described as follows: When applicants show self-promotion, they emphasize their job-relevant skills, abilities, or qualifications. Honest ingratiation includes communicating actual commonalities (e.g., values or opinions) with the interviewer or organization, or giving them honest compliments. When applicants honestly provide justifications for past negative work-related events and, if applicable, demonstrate what they have learned from them, this is called honest defensive IM. As is evident from Bourdage et al., these honest IM tactics represent counterparts to the deceptive IM tactics from Levashina and Campion (2007).

Some studies also distinguish between assertive and defensive IM (e.g., Ellis et al., 2002; Stevens & Kristof, 1995). However, in these studies, no distinction is made between deceptive and honest IM. Instead, the decisive factor is the intention to create a positive impression with the interviewer (Ellis et al., 2002). In assertive IM, applicants proactively try to create a positive image of themselves by ingratiating or presenting themselves with certain attributes (such as experience and knowledge,Bolino et al., 2008 ; Ellis et al., 2002). In defensive IM, applicants reactively make excuses or justifications to maintain or defend a good impression of themselves (Bolino et al., 2008; Ellis et al., 2002). Such a distinction between assertive and defensive IM could also be made in the different tactics by Levashina and Campion (2007) and Bourdage et al. (2018).

Existing Models of Deceptive and Honest IM

Models from the field of personnel selection (mostly from personality testing but also some that are applicable to selection situations in general) assume that deceptive IM is influenced by situational characteristics as well as by applicant characteristics (Ellingson & McFarland, 2011; Griffith et al., 2011; Levashina & Campion, 2006; McFarland & Ryan, 2000, 2006; Roulin et al., 2016; Snell et al., 1999; Tett & Simonet, 2011). In addition, some models assume that deceptive IM is also influenced by contextual characteristics (Levashina & Campion, 2006; Roulin et al., 2016). In this study, we refer to situational characteristics as factors that are specific to an interview situation (e.g., how interviewees are treated, Levashina and Campion (2006)). In contrast, we refer to contextual characteristics as general condition under which an interview takes place (e.g., the purpose of an interview, Levashina & Campion, 2006).

So far, there is only one model by Levashina and Campion (2006) that specifically deals with deceptive IM in the interview context. According to this model, applicants must have the capacity, willingness, and opportunity to fake for deceptive IM behavior to appear. These factors, in turn, are influenced by other variables, according to their model: Capacity to fake depends on variables such as oral skills or cognitive ability, but also on knowledge about job roles, or the construct being measured. Willingness to fake is determined by personality traits such as Machiavellianism or integrity. In addition, according to Levashina and Campion, applicants are more willing to fake in an interview if they think they will not be caught or if they feel unfairly treated. The opportunity to fake is mainly determined by the interview itself. Characteristics such as the amount of structure, the length, the purpose of the interview, or the targeted constructs are assumed to play a role here.

With respect to honest IM, Bourdage et al. (2018) used the deceptive IM model by Levashina and Campion (2006) in an initial attempt to investigate the occurrence of honest IM because of a lack of a specific model or one that integrates both deceptive and honest IM. Doing so, Bourdage et al. assumed that the capacity, willingness, and opportunity to show honest IM could influence corresponding behavior. In contrast to using Levashina and Campion’s model, only the self-presentation theory from Marcus (2009) allows that deceptive and honest IM might be considered as different examples of a common category of self-presentation behaviors that are all triggered by the same antecedents. However, neither of the two models makes any predictions about when deceptive IM occurs and when honest IM occurs. Furthermore, so far, it has only been found that these behaviors are positively related but whether they are really triggered by the same antecedents is not sufficiently clear even though initial evidence suggests a different pattern of influencing factors (Bourdage et al., 2018). Therefore, it is especially relevant to understand when deceptive IM occurs and when honest IM occurs. Furthermore, the role of psychological processes in the form of cognitions and affect also remains unclear.

Previous Research on Deceptive and Honest IM in Interviews

Most of the previous research on antecedents of deceptive IM has focused on applicant characteristics but there is very little research dealing with situational characteristics (see the review by Melchers et al., 2020). Concerning the latter, there are a few studies that considered effects of using different interview questions on deceptive IM. Bourdage et al. (2018), for example, found less slight and extensive image creation as well as less image protection when situational questions were used versus when they were not used. Concerning the use of past behavior questions, they found no difference between when these questions were used versus when they were not used. However, Levashina and Campion (2007) found that situational questions led to more total deceptive IM, ingratiation, and slight image creation behavior than past behavior questions. In addition, in contrast to their assumption that follow-up questions would reduce deceptive IM in interviews, Levashina and Campion found that follow-up questions increased deceptive IM. Besides different question types, Bourdage et al. also found more deceptive IM in interviews that were perceived as more difficult and more procedurally unfair. Finally, a recent interview simulation study by Bill et al. (2023) compared traditional interview questions (i.e., questions asking about attitudes, goals, or self-descriptions, for example, see Campion et al., 1997) to structured questions (i.e., situational and past-behavior questions). This study did not find any differences concerning the self-reported use of deceptive IM between the two types of questions. However, it found that interviewees could intentionally improve their interview performance more in answering the traditional questions than in the answering structured questions.

In addition to the interview itself, it has been found that the interviewer can affect deceptive IM. This effect can either be owing to the specific role of the interviewer (e.g., when the interviewer was the potential future supervisor, interviewees used more deceptive ingratiation than when the interviewer was not a potential future supervisor, Bourdage et al., 2018) or owing to the interviewer’s behavior. For example, Wilhelmy et al. (2021) found that interviewees adapted their IM to the interviewer’s IM such as by being more likely to use other-focused tactics in response to self-focused tactics by the interviewer (but whether these tactics were honest or deceptive could not be distinguished in this study). Additionally, in two different vignette studies, Ho et al. (2020) found that the interviewer’s emphasis on the organization’s competitive climate (Study 1) or on the competitive nature of the selection situation (Study 2) led to higher deceptive IM intentions than if this was not mentioned.

Furthermore, several studies also examined the effects of warnings from the interviewer on deceptive IM (Bill et al., 2020; Bill & Melchers, 2023; Law et al., 2016). Some of these studies suggest that a warning that deceptive IM can be identified leads to somewhat less deceptive IM than no warning (Bill et al., 2020, Study 2; Law et al., 2016) but other studies found no beneficial effects (Bill et al., 2020, Study 3; Bill & Melchers, 2023). Finally, in a vignette study, Bill and Melchers (2023) found that deceptive IM intentions were also unaffected by a description of a personable interviewer or the use of objective questions.

Besides these quantitative findings, there is also a recent qualitative study by Ho et al. (2021) who examined situational cues of deceptive IM in interviews. They found that some circumstances led to a perceived need to fake in interviews. These circumstances included that applicants were asked about work experiences that they did not have or remember or because they perceived a difference between their own values and interests and those of the interviewer. In other situations, however, the applicants perceived external pressure that caused deceptive IM, for example, when they perceived interview anxiety (e.g., from an intimidating interviewer) or when they felt pressure to answer interview questions immediately (e.g., in stressful interviews).

Furthermore, Ho et al. (2021) reported that applicants had the impression that they could fake without being detected in some situations. This impression was created by certain interview questions (traditional, opinion based, and leading questions), lack of job-related competencies of the interviewers, or when interviewers were less attentive. Additionally, this impression was also created when required competencies could be learned quickly or were not verified. However, Ho et al. did not investigate which kinds of deceptive IM tactics the situational cues led to. Furthermore, it is not yet known whether these cues also affect honest IM and what importance is attached to the associated affect and cognitions. However, knowledge of these aspects would be important to understand IM behavior holistically.

Finally, there have been a couple of studies on situational cues of honest IM. Bourdage et al. (2018) found that the use of past behavioral questions, situational questions, resumé-based questions, and preference-based questions resulted in more honest IM in comparison to when these questions were not used. Additionally, Bill et al. (2023) found more honest IM for traditional interview questions compared to structured interview questions.

Furthermore, Bourdage et al. (2018) found that both interview duration and perceived procedural justice had positive correlations with honest IM. They also found a negative correlation between perceived interview difficulty and honest IM and that interviewing with a potential future supervisor led to more honest IM compared to other interviewers.

There is also evidence that longer preparation time in asynchronous video interviews, which was provided to interviewees before they start to record their answers, led to more honest IM (Basch et al., 2021). Besides these three studies, however, research on honest IM has been very sparse. Therefore, it is important to find out which situational cues trigger this behavior.

Purpose of the Present Research and Research Questions

As described above, we believe there are significant gaps in current research on situational cues influencing deceptive IM (see the review by Melchers et al., 2020) and honest IM. In particular, previous models are limited by the fact that they only considered deceptive IM (e.g., Ho et al., 2021; Levashina & Campion, 2006) or else IM in general, without differentiating between deceptive and honest IM (e.g., Marcus, 2009). Moreover, there has been little insight into the psychological reactions (i.e., the emotions and cognitions) caused by the situational cues and how these influence subsequent IM behavior. These reactions could provide insight into why applicants exhibit either honest or deceptive behavior. For example, it could be that follow-up questions are interpreted by interviewees either as dissatisfaction with the content of the answer or as positively intended interest. If there is such a difference in perception, it seems likely that the same situational cue will also influence subsequent behavior in different ways. Therefore, we consider it important to understand such psychological processes in order to understand IM behavior holistically.

Accordingly, the overall goal of this study is to derive a model that includes both deceptive and honest IM and to show how the psychological processing of these cues affects subsequent IM behavior. Here, we are interested not only in deceptive and honest IM in general, but also, with regard to the interview tactics presented above, in learning which tactics applicants use. To do so, we used Grounded Theory (Glaser & Strauss, 1967) as a qualitative approach, which is particularly suitable for investigating black box phenomena such as the psychological processing of situational cues and phenomena for which theoretical frameworks are missing (Murphy et al., 2017). This general approach is similar to a previous study by Wilhelmy et al. (2016) that dealt with IM on the part of the interviewer and in which Grounded Theory was successfully used to identify interviewers’ IM tactics, antecedents of these tactics, and their intentions to use IM tactics. Thus, Grounded Theory seems suitable for investigating antecedents and related processes of interviewee IM for the present study, in which we want to answer the following research questions:

Research Question 1: How do applicants psychologically process situational cues in general and how does this processing influence subsequent deceptive and honest IM?

Research Question 2: Which different situational cues influence applicants’ deceptive and honest IM in detail and how are they processed individually?

Grounded Theory Approach

Grounded Theory (Glaser & Strauss, 1967) is especially useful for focusing on a particular phenomenon in a certain context (Murphy et al., 2017) and for developing theories to better explain it (Charmaz, 2006). Additionally, it is suited for the exploration of black box phenomena (Murphy et al., 2017), such as the psychological processing of stimuli. Therefore, we used Grounded Theory (Glaser & Strauss, 1967) to investigate situational cues of deceptive and honest IM, related psychological processes, and subsequent deceptive and honest IM in selection interviews. Within Grounded Theory, we used a twin-slate approach (Murphy et al., 2017), which means that we incorporated previous research to define topics in advance that have proven relevant in the past.

In collecting the data, we followed the principle of theoretical sampling (Glaser & Strauss, 1967). In line with this principle, we developed our theory by continuously analyzing the data during the data collection process. These data determined further data collection and consequently, the selection of the appropriate sample with which to examine concepts and their relationship to each other in more detail (Corbin & Strauss, 2015; Glaser & Strauss, 1967). In addition, we used the constant comparative method (Glaser & Strauss, 1967) to condense single incidents into a theory. This method includes the continuous comparison of new data with already coded data, the writing of memos to record initial theoretical considerations, the involvement of a second coder, and the up-aggregation of the data to a higher level theory (Glaser & Strauss, 1967). According to Grounded Theory, data collection stops when theoretical saturation is reached, that is, when no new insights regarding the theory or categories are revealed through further data (Charmaz, 2006). For our study, this was the case after N = 31 interviews with informants. However, we conducted another two interviews that were already scheduled.

Sample

In order to obtain the most accurate first-hand information possible, we focused on informants who had recently participated in a job interview. To obtain a comprehensive picture of interview behavior, we also pre-specified characteristics that might be associated with IM and according to which informants should differ. These were: age, gender, work experience, interview experience, and the type of position applied for. Following the theoretical sampling approach (Glaser & Strauss, 1967), further data collection was also determined by the data that were already collected. Thus, in the data collection process, we found that sector was another important characteristic by which the sample should be diversified. We found that staff shortages in the health sector meant that applicants in this sector felt less need to present themselves in the best possible way, as the demand for labor was correspondingly high. For this reason, we then specifically searched for informants from other sectors. In addition, the initial precondition for participation in our study was that informants had taken part in at least one interview during the last 12 months. However, this time range was reduced to six months after the first interview to ensure more accurate recall.

The sample consisted of N = 33 informants (12 males, 21 females). Their average age was M = 30.06 years (SD = 10.78) and ranged from 18 to 57. On average, informants had completed M = 2.22 interviews (SD = 1.76) during the last six months (n = 32). Overall, they had a mean interview experience of M = 11.33 interviews (SD = 11.29) with a range of 2 to 60 interviews. Of these interviews, M = 56.73 % (SD = 28.59 %) had been successful at the time of the survey, ranging from 0 to 100 %. In addition, several informants had already conducted interviews as interviewers themselves because they held managerial roles and/or were involved in the selection of interns, applicants, or new colleagues. One of them (an HR manager) had acted as an interviewer in about 100 interviews, but of the remaining only four had conducted more than five interviews (M = 2.25, SD = 4.20, Mdn = 0, for the remaining 32 informants).

With regard to the highest general school-leaving qualification, 18.20 % had an intermediate school-leaving certificate or equivalent and 81.80 % had a high school diploma. Concerning the highest vocational qualification, 12.10 % had recognized vocational training, 15.20 % a technical college qualification or a professional qualification, 27.30 % a bachelor’s degree, 30.30 % a master’s degree or diploma, and 15.20 % had no vocational training qualification.

In addition, 6.10 % of the informants reported being job seekers, 42.40 % were employed, 6.10 % were in vocational training, 36.40% were university students, and 6.10% were high school students. The average working time per week was M = 22.36 hours (SD = 18.27) and ranged between 0 to 50 hours. On average, informants had 9.26 years (SD = 9.91) of work experience with a range from 0 to 36 years.

In addition, concerning job sectors, 21.2 % were from the industry or manufacturing, 15.2 % from health and social work, 12.1 % from research and science, 9.1 % from public administration, 6.1 % from trade and distribution, 3.0 % from education and training, 18.2 % none, and 15.2 % other industries (also cf. Table 1 for an overview).

Informants were recruited via social media such as LinkedIn and Facebook, and via e-mail distribution lists of the first author’s university. In addition, in line with common recommendations (Waller et al., 2015), our study was recommended to new informants by informants who had already been interviewed. Informants participated in the study voluntarily and received no compensation.

Data Collection

We conducted semi-structured interviews via a video conferencing platform to investigate which IM strategies were used in past selection interviews. We also asked about situational cues for these behaviors and related psychological processes. Interviews were conducted by a master’s student and a doctoral student specializing in work and organizational psychology. Prior to the interview, informants completed an online survey. In this survey, we described the content of the study and asked them to complete an informed consent form. In addition, we asked them for demographic information, their work experience, and their interview experience (during the past 6 and 12 months), the medium through which past selection interviews were conducted, and some other basic information about the structure and content of these selection interviews.

Interview Guide

First, a literature search about situational cues of IM was conducted. The findings from this search were considered in the development of the interview guide. The first guide covered the following topics: medium of communication, first impression, duration of the interview, warning of being detected while using deceptive IM, interview structure, aspects concerning the interviewer, aspects concerning the interviewee, and communication before and during the interview. In line with parallel data collection and data analysis according to Grounded Theory (Glaser & Strauss, 1967), the guide was adapted constantly during the data collection. New aspects were integrated, irrelevant aspects were removed, the order was adjusted, and questions were reworded for better understanding. Each change in the guide was documented. The sixth and final version of the guide covered the following topics: interview structure, first impression, duration of the interview, medium of communication, geographic location of the job, aspects concerning the interviewer, aspects concerning the interviewee, and communication before and during the interview.

In formulating the questions for the interview guide, we considered the suggestions by Brinkmann and Kvale (2015). In order to obtain a broad and unbiased range of information, the guide always started with an open-ended question. Specifically, we first asked informants about general salient memories of past selection interviews, followed by a question about behaviors exhibited in order to make a good impression. In addition, informants were asked about possible incidents that caused them to change their strategy.

Although we changed the guide over time, the overall structure remained the same across versions. Each situational topic was introduced with an open-ended question. For example, for the “interview structure” topic, the question “Can you please describe to me the rough flow of the interview?” was used for this purpose. The following main questions covered further aspects of the respective topic (e.g., “Did you talk about your strengths and weaknesses during the interview at all?”). In addition, we touched on possible IM behavior at this level (e.g., “When you had the opportunity to talk freely about yourself [e.g., during the self-introduction], what did you do to make as good an impression as possible?”). The main questions were followed by more detailed questions aimed at possible emotions (e.g., “What feelings did this trigger inside you?”), cognitions (e.g., “What went through your mind?”), intentions (e.g., “What did you want to achieve by doing this?”), specific behaviors (e.g., “How did you react to this?”), or the type and strength of possible IM shown (e.g., “Were you always 100% honest when answering?”).

The interviews were conducted via a video conferencing platform and were recorded. The recordings of the interviews were automatically transcribed and manually revised. The interview duration was between 24 and 85 minutes (M = 57.55, SD = 10.31). Before the recording began, we reiterated the purpose of the study and explained the procedure. As deceptive IM in selection interviews is potentially a sensitive topic, we explicitly informed the informants once again before recording that the data would be treated confidentially and pseudo-anonymized. Additionally, we turned off the cameras during the interview so that the informants felt less observed. Furthermore, we also used softer wording for deceptive IM during the interview (e.g., we asked informants whether they had fibbed or told the truth 100% of the time instead of asking them directly whether they had lied).

After the start of the recording, we started asking questions according to the interview guide. The order and level of detail of the questions varied as we individually addressed what the informants reported. This approach is in line with Brinkmann and Kvale (2015), who pointed out that in qualitative interviews, it may be at the discretion of the interviewer to deviate from the interview guide and be sensitive to the informants’ responses.

Data Analysis

All interviews were analyzed sentence by sentence. In line with the idea of investigator triangulation (Denzin, 1978), we used two independent coders for each interview to reduce possible personal biases and to account for different perspectives on the data. A total of six coders were involved in the coding process, with one main coder who participated in the coding of all 33 interviews.

The first step of the analysis consisted of open coding (Corbin & Strauss, 2015). Accordingly, the coders assigned code names to individual words, sentences, or text passages. Following the twin-slate approach (Murphy et al., 2017), the coders based the code names on the terminologies of the existing literature if there were already defined constructs, or, if this was not the case, on the informants’ literal statements. Thus, a hierarchically arranged coding tree grew and was increasingly enriched while the data was being analyzed, which already initially included the main categories Situational Cue, Behavior, Intention, Cognition, and Affect on account of the research questions. With a growing number of interviews that were analyzed, the level of abstraction of the codes became increasingly higher. The method of constant comparison (Glaser & Strauss, 1967) prevented the establishment of redundant codes. By continuously writing memos (Glaser & Strauss, 1967) during the analysis, initial theoretical considerations were recorded. The document memos, on the other hand, contained the summary of the most important content of each interview.

In the second step of the analysis and in line with the idea of investigator triangulation (Denzin, 1978), the coders discussed their individual results from a maximum of two interviews in a coding meeting and agreed on a consensus of the codes and hierarchy levels to be assigned. Changes in the code system were recorded in detail in a coding diary and considered in further analyses. During the evaluation process, 1344 different codes were developed. Thematically similar codes were arranged close to each other in the code system to enable the codes to be found quickly.

In the third step of the analysis and after all the interviews were analyzed, we adapted the structure of the code system. Codes of the main categories Behavior, Intention, Cognition, and Affect that were similar in content were aggregated into concepts. For example, the behavior codes “communicate interest”, “display motivation”, and “signal readiness” were aggregated into the concept “presenting personal drive.” In this aggregation, codes were removed that occurred without situational cues, that had occurred before or after the interview, or that were not specific enough even when we reconsidered the raw data. For the aggregation of situational cues, only those that had actually occurred in the interview situation and that were relevant for subsequent behavior were considered. Cues that were supposedly related in content were checked for a common effect path. By this we mean that they led to comparable emotions and/or cognitions and to comparable IM behavior. If this was the case, these cues were also combined into one concept. For example, the cues “questions about characteristics” and “questions about strengths” were similar in content and had the same effect path. Thus, they were aggregated into the concept “questions to test suitability.” If no common effect path could be identified, the cues were considered separately.

In the fourth step of the analysis, we focused only on the cues whose behavior was clearly attributable to IM and/or the aggregated intentions “make a good impression” or “do not make a negative impression”, as these intentions were a prerequisite for IM. Exceptions to this focus were cues that, in addition to meeting the aforementioned criteria, had no effect on the behavior or at least on applicants’ IM behavior. These cues were labeled with “no effect.” This exception made it possible for us to express that not every cue had an impact on IM for all applicants. Due to the aforementioned restrictions, the concepts became more and more condensed.

In the fifth step, we used axial coding (Strauss, 1987) to create and strengthen relationships between the individual concepts that formed a preliminary theory. To do so, concepts were divided into clusters in terms of their related affect and IM. For example, concepts that were always accompanied by negative affect and whose behaviors could always be assigned to deceptive IM formed one possible cluster. Another cluster consisted of concepts that were perceived positively in some sentences, negatively in others, and were accordingly associated with honest or deceptive IM. Within the clusters, the theoretical models that included affect and/or cognitions were then formed.

In the sixth and final step, categories were formed by further aggregating concepts that were related in content and occurred together in these clusters. Owing to the abundance of data and the range of situational cues, three different models crystallized from the data. Therefore, the resulting categories for the main category Situational Cue were thematically organized into groups, according to which the different models were structured. In addition, behavior was also organized into “impression management” and “other behavior” in order to present the categories more comprehensibly. In addition, and to answer Research Question 1, we created a main model that summarized the commonalities regarding the psychological processing across all situational cues. In doing so, we also included situational cues that did not make it into the two different models developed before.

In summary, individual codes were initially aggregated into concepts and these were condensed into categories over the course of the analyses. The final categories with their concepts were recorded in the category book (see Table 2) and illustrated with definitions and interview quotes (also cf. Murphy et al., 2017). We included the interview quotes to make the data more vivid and to strengthen the credibility of the data (Glaser & Strauss, 1967). The final models were developed and visualized separately according to groups of the identified situational cue categories. In addition, regarding the psychological processes, we condensed commonalities across all situational cues to develop a main model. Each step of the process from raw data to theory building was recorded in detail (Corbin & Strauss, 2015). Taking existing literature into account when analyzing our own data allowed us to develop theoretical models that are connected to existing knowledge (Murphy et al., 2017).

Member Checks

We conducted so-called member checks (Charmaz, 2006) to verify that the generated models were accurate and consistent with informants’ perceptions. To do so, we summarized the key findings, presented them again to five informants, and asked them whether they had any comments on these results that they would like to share with us (also see Bluhm et al., 2011; Locke & Velamuri, 2009). Overall, informants agreed with our findings and had no additional comments.

The Impact of Psychological Processing on Impression Management

One of our main aims was to find out how applicants psychologically process situational cues in general and how this processing influences subsequent deceptive and honest IM (Research Question 1). To investigate this question, we condensed commonalities across all situational cues with respect to applicant affect, cognition, and IM (see Fig. 1; definitions of all categories and concepts are shown in Table 2).

The Impact of Psychological Processing on Impression Management. Note. POJ-Fit = person-organization-job-fit; IM = impression management

Full size image

Across different situational cues, we found a pattern that positive affect (represented by dashed arrows in the figures) mainly led to^{Footnote 1} honest assertive IM. For example, in response to a biographical question, an informant reported: “it was something positive, with which I could also present myself well” (Inf15). In line with this example, another informant reported: “I didn’t feel under pressure … it was a very relaxed and pleasant conversation. I behaved naturally … I didn’t have to pretend much” (Inf13). Additionally, negative affect (represented by solid arrows) was associated with honest assertive IM and deceptive IM.

Interestingly, we found that negative affect, as opposed to positive affect, was not only mainly followed by a certain type of IM but also by rejections of a potential job offer (e.g., “The questions were very personal … the conversation was not on an equal footing. … there was no feedback … it was more of an interrogation. … I thought this was very inappropriate. … I was quite angry after the interview and already knew that I didn’t want to take the offer, regardless of whether I was accepted or not.” Inf9). In this regard, our results parallel meta-analytic findings from the recruiting domain that there is a higher likelihood that applicants reject job offers when recruiters are perceived as unfriendly (Chapman et al., 2005).

Furthermore, when looking at deceptive IM, we found that it was usually preceded by negative affect. For example, with regard to a personal question, a female informant reported: “That’s when I felt unfairly treated in comparison to men and that’s why I ... lied” (Inf28). This result is in line with findings from Bourdage et al. (2018) and suggestions from Levashina and Campion (2006) that perceptions of unfair treatment increase applicants’ willingness to use deceptive IM. Another informant reported: “I was surprised by one question. … I wasn’t prepared for this. … a little panic had spread through me, my stomach tightened for a moment and I probably also held my breath for a moment and needed a bit, probably a second or two, to catch myself. ... I didn’t lie but … my answer was a bit exaggerated.” (Inf23).

The situational cues that seemed to trigger negative and/or positive affect varied across the different models that we identified, which can also be seen in the Figs. 2, 3, 4. However, overall, our results suggest that when a situational cue elicits positive affect in applicants, they present themselves predominantly honestly whereas they answer predominantly dishonestly when they previously experienced negative affect.

Situational Cues with Regard to the Interviewer. Note. IM = impression management

Full size image

In addition to affect, we also found cognitions that were followed by honest assertive IM. Several informants^{Footnote 2} reported that they used honest assertive IM when they thought the interviewer was actually interested in them as a person (which we refer to as personal assessment, for example “one interviewer was interested in my personal nature”, Inf15, or “Inwardly you thought ... they are also interested in the person,” Inf30). Another pattern that we found was that honest assertive IM was often used when informants thought that they were a good fit for the organization and the job (i.e., POJ-Fit given) such as “I fit in perfectly here” (Inf17) or “I knew that I was suitable” (Inf24). Thus, our results also show that cognitions of being assessed as a person (personal assessment) and fitting the organization and the job are crucial for whether applicants honestly promote themselves.

Situational Cues of Deceptive and Honest Impression Management

To address Research Question 2, we investigated which different situational cues influence applicants’ deceptive and honest IM in detail, and how they are processed individually. Based on our data, we were able to identify different groups of situational categories that we organized thematically into different models. These models were related to the interviewer (Fig. 2) and interview content (Figs. 3 and 4). Furthermore, as explained below, we found many situational cues that affect deceptive and honest IM but that have received little or no attention so far.

Interviewer Cues

Regarding the interviewer, we found a pattern that closeness was associated with applicants’ IM behavior. In these cases, interviewers conveyed a sense of closeness and approachability through a positive relationship with the applicant. In another qualitative study, Wilhelmy et al. (2016) identified a corresponding IM intention on the part of the interviewer, which they referred to as “signaling closeness”. Until now, it was unclear to which psychological processes and behavior closeness can lead on the part of the applicant.

Under closeness, we summarized when interviewers were known to the applicants (e.g., “Yes, I knew them [the interviewers] all,” Inf33), were responsive to them (e.g., “The new managing director supported me in it and said she thinks it’s good if I finish everything properly and then start the new job,” Inf16), met them on an equal footing (e.g., “Both … signaled their interest to me and not as ... if this is a compulsory conversation that they have to get over quickly,” Inf12), and gave them positive feedback (e.g., “if someone [the interviewer] says ... that [I] was very sympathetic to him and [he] can imagine that [an employment] very well,” Inf17; cf. Table 2 for the definitions of all categories and concepts). Overall, we found that closeness was followed by positive affect and, honest assertive IM (e.g., “Both … signaled their interest to me. ... [I] stuck to the truth and emphasized my strengths,” Inf12).

In addition to closeness, we identified three more categories of situational cues concerning the interviewer that all resulted in negative affect. First, we found a pattern that distance—that is, interviewers portraying their superiority over applicants and appearing to have little approachability—was associated with applicants’ IM. This category includes, for example, situations in which the interviewers demonstrated their power or position (e.g., “I was really interrupted a lot, [the interviewer said] ‘No, I'm not really interested in that right now …’,” Inf25), did not know the applicant’s resumé (e.g., “I think he didn’t even look at my resumé and application,” Inf20), imputed behavior or opinions (e.g., “that the interviewer implied that their company is better than another ... and therefore they can understand ... that I had applied to them,” Inf9), talked for a long time (e.g., “So, this person talked extremely much,” Inf33), or disagreed with the applicant (“So, we got into quite a bit of a tussle,” Inf25, also see Table 2 for the description of all concepts).

With regard to distance, Wilhelmy et al. (2016) also found a corresponding IM intention on the part of the interviewer, which they referred to as “signaling distance in terms of superiority” (p. 318). In doing so, the interviewers had the goal of demonstrating power or of triggering uncertainty in applicants (Wilhelmy et al., 2016). Until now, however, it was unknown to which psychological processes and behaviors it leads on the part of the applicants. We found a pattern that distance seemed to lead to cognitions in applicants that the interviewer was not interested in them as a person. We refer to this as “impersonal assessment” (e.g., “my interviewer didn’t look at my resumé ... I definitely found it extremely negative ... because sometimes I really had the feeling they didn’t know which applicant was sitting in front of them,” Inf5). Alternatively—or additionally—distance was associated with the concern of leaving a negative impression (e.g., “One questions to what extent one’s own application is actually promising,” Inf5). Applicants reacted to distance on the part of the interviewer with honest assertive as well as with deceptive IM.

In addition, it should be mentioned that when the interviewer talked for a long time, some of the informants also perceived this behavior positively (e.g., “I was much more relaxed than during any questions that I had to answer,” Inf14) and thus, this was not always followed by negative affect. Hence, similar to closeness, distance did not always affect applicants’ IM.

Additionally, it was particularly interesting that, for one informant, implying a behavior or opinion led to a conclusion about the prevailing working methods, which resulted in rejecting the potential offer (e.g., “Yes, a bit along the lines of: ‘You can’t trust the employees’ ... . I didn’t take up the position because of that,” Inf19). This example is particularly relevant because it illustrates that deceptive IM may not be the only possible consequence of this behavior.

Concerning the next category of situational cues related to the interviewer, we found that when the interviewer was a potential direct supervisor, honest IM was reported (e.g., "But with the head of department, the direct supervisor, ... , there, one made a ... greater effort," Inf28; also see Bourdage et al., 2018, for a comparable result).

Furthermore, several informants reported deceptive IM when an interviewer arrived late. A possible reason for using deceptive IM in such a situation was provided by one informant who explained: “And then we first waited five minutes for the direct supervisor. ... My first thought was ‘Oh my God. There’s a second applicant here’,” Inf20). This finding goes well beyond past findings and indicates that late arrival might be associated with deceptive IM.

Interview Content Cues

Given that we found more situational cues concerning the interview content than we could fit into one figure, we decided to illustrate the cues in two different models in order to make the visualization clearer (cf. Figs. 3 and 4). As can be seen in Fig. 3, we found that the opportunity for applicants to talk freely about themselves (self-introduction; e.g., “I was supposed to talk about myself,” Inf5) was associated with their IM. Interestingly, we found two different patterns in our data regarding how this opportunity to present themselves was processed.

Situational Cues with Regard to the Interview Content. Note. IM = impression management

Full size image

Situational Cues with Regard to the Interview Content. Note. IM = impression management

Full size image

In the first pattern, a self-introduction was perceived as positive and was mainly followed by honest assertive IM (e.g., “when I had to introduce myself, I focused more on that [my practical skills],” Inf15). In the second pattern, a self-introduction was perceived as negative and was not only followed by honest assertive IM but also by deceptive IM. Within this pattern, we repeatedly found concerns about leaving a negative impression (e.g., “I didn’t want to work clinically from the beginning, but I expressed it like this ... . Because I felt I couldn’t say that it wasn’t my first priority,” Inf14). This situational cue has not received any attention in the existing literature and also illustrates very well that psychological processing can be crucial in determining what kind of IM applicants use in response to the cues that they perceived during the interview.

In addition to the opportunity to introduce themselves, our results indicate that previous research (cf. Melchers et al., 2020) has overlooked many types of questions that could have an impact on IM. For instance, we found a pattern that questions about applicants’ motivation had an impact on applicants’ IM (cf. Fig. 3). These questions aim to determine the interviewees’ motivation for their application. They were primarily associated with positive affect and were followed by honest assertive as well as deceptive IM (e.g., “then questions were asked about motivation ... I didn’t say that I believe I have found my passion in the previous areas, although I continue to believe that this is the case. Because I would have ruined all the chances for myself,” Inf8).

Questions about the person were another situational cue of IM (cf. Fig. 3). These questions refer to the applicants’ past behavior, interests, or private matters such as leisure activities (e.g., “interviewers asked about situations … and how I’ve handled those,” Inf25). Two different patterns emerged from our data. In the first pattern, personal questions were perceived as positive and followed by honest assertive IM. Within this pattern, we repeatedly found the cognition that applicants had the feeling the organization wanted to get a picture of them as a person (personal assessment; e.g., “I felt like they really wanted to get to know me,” Inf18). In the second pattern, personal questions were perceived as negative (e.g., “I think that question is somehow inappropriate,” Inf9) and were followed not only to honest but also deceptive IM. Within this pattern, having concern about leaving a negative impression was reported multiple times, and some informants also thought that such questions targeted non-work-related matters.

Interviewer questions about the position also had an impact on applicants’ IM (cf. Fig. 3). These questions cover aspects such as salary expectations, expectations of the job, or how applicants became aware of the position (e.g., “Then the question came what I expect from an employer or from a good employer,” Inf27). Again, we found two different patterns about how applicants reacted. In the first pattern, these questions were perceived as positive and followed by honest assertive IM (e.g., “I thought ... ‘Well, if I manage to get into one of these companies now, then I’ll be absolutely the happiest person on earth’ and I actually just tried to remain myself again, to simply remain honest and open,” Inf2). In the second pattern, they were perceived as negative and were followed by deceptive IM. In this pattern, we found several examples of the concern about leaving a negative impression (e.g., “in the case of salary [expectations], where I simply stated less than I would actually like. … So that I don’t appear too greedy,” Inf30).

In addition, across multiple informants, we found that situational questions had an impact on applicants’ IM (cf. Fig. 4). These questions referred to hypothetical future situations at work (Latham et al., 1980), and asked applicants about their specific behavior in these situations (e.g., “They described situations and then asked me ‘What would you do in this situation?’,” Inf25). Although we only found a small number of cases, our findings suggest that affect in the context of these questions can play a crucial role with regard to the subsequent IM behavior. If these questions were perceived as positive, they were followed by deceptive and honest assertive IM (e.g., “These were all very relevant questions … and I answered them to the best of my knowledge,” Inf25). If, on the other hand, they were perceived as negative, deceptive assertive IM was reported (e.g., “I don’t know if I would always react as calmly as I’ve described. I think I’ll maybe, in some situations probably become a little more explosive. … [however] I think the one who gets loud loses,” Inf21).

We also found a pattern that questions to test applicants’ suitability affected their IM (cf. Fig. 4). These include, for example, questions about their qualifications and skills but also job knowledge questions or work-related tasks (e.g., “I was asked specifically, ‘How do I deal with subordinates?’,” Inf19). This category was associated with negative affect and honest as well as deceptive IM (e.g., “What I find really stupid is simply the question [about] ‘strengths and weaknesses.’ … [regarding the weaknesses] that’s where I had to be a bit careful not to be too honest,” Inf17).

Finally, we identified a pattern that interviewers’ questions about the organization affected applicants’ IM behavior (cf. Fig. 4). These questions were asked to check applicants’ knowledge about the organization (e.g., “my boss then asked ‘Yes, what do you know about the company?’ And I just told him all the information I could remember, that I could learn from the website,” Inf20). For applicants, these questions were associated with negative affect and resulted in honest assertive IM.

With regard to the different question types, our results again indicate that applicants’ affect can be crucial in determining which type of IM is used. Thus, applicants tended to rely mostly to honest IM when they had a positive feeling whereas they exclusively or additionally tended to use deceptive IM when they had a negative feeling. In addition, the associated cognitions provide insights into the reasons for the affect that occurred and the IM that was used. For example, we found that questions about the person can lead to negative affect and to deceptive and honest IM. As a possible reason, we found that the informants thought that these questions were not related to the job. These findings go beyond previous research on the impact of different question types on deceptive and honest IM (e.g., Bill et al., 2023; Bourdage et al., 2018; Ho et al., 2021; Levashina & Campion, 2007). This previous research discussed different question types mainly with regard to the opportunity to fake but did not consider that applicants might psychologically process questions differently which then affects their willingness to fake. In addition, going beyond these past findings (cf. Melchers et al., 2020), we found that applicants’ IM is also influenced by questions about applicants’ motivation, position, and organization, and by personal questions, job knowledge questions, questions to test their suitability, and work-related tasks (cf. Table 2 for definitions).

Apart from different question types, we also found that the presentation of the organization or job previews had an impact on applicants’ IM (cf. Fig. 4). When presenting their organization, interviewers described the organization in more detail and talked about aspects such as culture, structures, or current challenges (e.g., “Then they talked about what the company stands for, what the company does,” Inf23). With regard to job previews, interviewers described the position in more detail and addressed aspects like the working conditions (e.g., “we then relatively quickly started talking about the details of the advertised position,” Inf22).

In our data, we identified two different patterns showing how applicants responded to presentations of the organization or job previews. In the first pattern, these cues were perceived as positive and were followed by honest assertive IM (e.g., “they stated what they stand for and what they want to achieve and what drives them. ... it was part of the strategy that I wanted to tell them ‘Okay. I also have honest [the same] goals and I want to contribute to the purpose that the company has,” Inf11). This pattern is consistent with a finding from Bourdage et al. (2018) that attraction to the organization is positively related to honest IM.

In the second pattern, the presentation of the organization or job previews were perceived as negative and were mainly followed by deceptive defensive IM. In this case, among other things, applicants tried to conceal their negative feelings (e.g., “Once, I had a job [application] and I would have been alone a lot, so I thought, ‘Oh, that's too bad, a small team would be nice’. ... But then I acted as if it would be okay,” Inf17). This finding goes beyond previous research and again demonstrates the crucial role of psychological processing of situational cues. In addition, the data suggested that only the negative evaluation of the job preview was followed by the rejection of a potential offer, whereas a negatively evaluated presentation of the organization was not. This result is in agreement with meta-analytic findings that job characteristics are related to intentions to accept a job (Chapman et al., 2005). In addition, our analyses revealed that the presentation of the organization and the job preview did not have an effect on the behavior of every informant (cf. Figs. 3 and 4).

Additional Results

Although it was not the focus of the present study, we were able to identify different IM tactics that go beyond the previous literature (e.g., Bourdage et al., 2018; Levashina & Campion, 2007). Regarding honest IM, informants told us that they presented their personal values, drive, attitudes, and opinions in order to make the best impression on the interviewer. In addition, they also corrected erroneous information, clarified their added value to the organization, or portrayed their emotions in order to present themselves in the best possible way. In contrast, informants also reported purposefully not expressing their emotions, which we assigned to deceptive defensive IM (cf. Table 2 for definitions and examples for the different strategies).

Summary of Contributions

The present study makes at least five important contributions to the literature: First, it identified several new antecedents concerning interviewer behavior and interview content and helped to understand how these antecedents affect subsequent IM. Specifically, we found that there are many situational cues that affect deceptive and honest IM that have received little or no attention so far, such as closeness on the part of the interviewer or a delay in the arrival of the interviewer. These findings suggest that the interviewer affects applicants’ IM. Furthermore, we also found that interview content seems to affect deceptive and/or honest IM and that previous research (cf. Melchers et al., 2020) has overlooked many relevant types of questions or the presentation of the organization and the job preview.

As a second important contribution, our study stresses that both deceptive and honest IM should be considered together in conceptual models of IM. A joint consideration of these constructs fell short in previous research and represented an important research gap. In the past, there were models related to deceptive IM alone (e.g., Levashina & Campion, 2006; Roulin et al., 2016) or that did not distinguish between honest versus deceptive strategies (Marcus, 2009) but models that focus on honest IM or that integrate both deceptive and honest IM have not previously existed. In contrast to these earlier approaches, our study details when applicants choose honest IM, deceptive IM, or a combination of both.

Third, our findings stress the importance of applicants’ cognitions and especially of affect for subsequent IM behavior. In particular, we found that positive affect was mainly followed by honest assertive IM whereas deceptive IM was usually preceded by negative affect. This finding extends results from the qualitative study by Ho et al. (2021) that provided initial insights into what cognitions may be behind deceptive IM, but that did not offer a holistic view that included affect and cognitions for different cues and considered their effects on both deceptive and honest IM.

Fourth, we identified several IM tactics that were not considered in the past (e.g., Bourdage et al., 2018; Levashina & Campion, 2007). For example, we found that applicants honestly presented their personal drive (e.g., their motivation and willingness to work) to make as good an impression as possible. Even though these additional findings were not the main focus of the present study, it seems worthwhile to consider them in future research.

Finally, with regard to Levashina and Campion’s (2006) faking model and the attempt by Bourdage et al. (2018) to apply this theory to honest IM, our findings mainly provide new insights into antecedents of applicants’ willingness to use IM. Thus, the perception of the different situational cues is associated with different affective reactions and the specific affect then seems to impact applicants’ willingness to use honest and/or deceptive IM. In line with this, Levashina and Campion state “Willingness to fake represents psychological and emotional characteristics that influence the degree to which candidates are inclined to distort their response during an interview” (p. 302). Furthermore, different question types, which were mainly discussed with regard to their effects on the opportunity to use deceptive IM (e.g., Bill et al., 2023; Bourdage et al., 2018; Ho et al., 2021; Levashina & Campion, 2006; Levashina & Campion, 2007), also seem to influence applicants’ willingness to use honest vs. deceptive IM. Only the opportunity for applicants to talk freely about themselves in a job interview turned out to be a previously unmentioned opportunity for IM. In contrast to antecedents of willingness and opportunity, we did not find new insights on aspects related to applicants’ capacity to fake, which is the third factor from the model by Levashina and Campion (2006). This, however, is not surprising given that antecedents of the capacity to fake are mainly related to individual difference variables and not to situational cues (e.g., Buehl et al., 2019; Moon et al., 2024). Thus, they are beyond the scope of the present study.

Implication for Practice

In addition to the different theoretical contributions described above, our results also have several important practical implications that could help to ensure that applicants engage in less deceptive IM in interviews and predominantly use honest IM, given that the latter might “allow interviewers to make more informed decisions by providing accurate, job-related information” (Bourdage et al., 2018, p. 600). First, our results show that interviewer behavior could be an important factor in how applicants present themselves. Given that interviewers are not able to reliably detect deceptive IM (Roulin et al., 2014, 2015) and that most of the previous approaches to prevent deceptive IM have not been effective (e.g., Bill & Melchers, 2023) interviewer behavior might be one of the few things that could help to limit deceptive IM. Thus, first of all, interviewers should make sure that their behavior toward applicants is approachable. Being approachable could include that interviewers showing understanding, talking about themselves, not playing up their position, or giving positive feedback during the interview. Such a behavior could give applicants not only a more positive feeling, but applicants could also be more likely to use honest assertive IM to present themselves favorably in such a situation. To promote approachable behavior on the part of interviewers, they could be specifically trained to exhibit this behavior more. In addition, our results suggest that interviewers should arrive on time for the interview because a late arrival could give the applicants the impression that the interviewer might have been in a conversation with another applicant, which in turn may result in deceptive IM.

Second, when interviewers ask questions about the applicant, they should make sure that the questions are perceived as work-related, otherwise this probably could not only lead to negative affect, but also to deceptive IM in addition to honest IM. Here, it is certainly advisable to conduct a job analysis, for example, using critical incident technique (Flanagan, 1954), prior to the interview to ensure that the questions are specifically related to the job.

Third, we found that applicants’ cognition to fit to the organization and to the job probably leads to honest assertive IM. Here, we recommend making the job as well as the organization’s goals and values transparent before the interview so that applicants can get a picture of them before the interview and, if applicable, can withdraw their application if they think they are not a good fit for the organization or the job.

Limitations and Propositions for Future Research

Although our study provides important insights into which cues affect deceptive and honest IM, it has some limitations. First, given that we have chosen a Grounded Theory approach (Glaser & Strauss, 1967) to explore the range of situational cues, the underlying psychological processes, and their effects on deceptive and honest IM, no conclusion can be drawn on causality and as to how strong these effects are because this is beyond the scope of such a qualitative approach. Thus, the identified cues and the influence of psychological processing on deceptive and honest IM should be evaluated in further quantitative studies to determine causality and effect size, for example, by using experimental study designs.

Second, another limitation concerns our inclusion criterion for participants. This criterion required that individuals had participated in a selection interview during the last 6 months. Although we had already reduced the time from 12 to 6 months, gave them time to remember the interview situations, asked them about general salient memories of past selection interviews in the beginning of the interview to prime their memories, and had the impression that informants could remember these interviews very well, it may still be that some informants could not always fully remember every detail of the respective interviews. Thus, it might be valuable to collect information from informants in future studies more immediately after they have completed an employment interview.

A third limitation concerns that it might be that despite all measures, not all informants were completely honest about their deceptive IM behavior in interviews. To address this potential issue, we started slowly into the topic, used softer wording, turned off the cameras during the interview to make informants feel less observed, avoided judgmental language, ensured confidentiality, and emphasized that it is okay if informants were not able to answer questions. There was also no incentive in the sense of a reward for informants if they presented themselves as predominantly honest. In addition, we found numerous cases in our data where informants reported using deceptive IM in the interview.

A final limitation is that we focused only on cues that occur within the employment interview situation. However, our interviews with informants revealed that contextual cues (i.e., the general conditions under which an interview takes place) may also be critical in determining what type and extent of IM is exhibited. For example, we found that the high demand for labor in the health sector was associated with a lower perceived need for using IM. Also, owing to the sparse research on contextual antecedents of deceptive IM (Melchers et al., 2020), and also of honest IM so far (Bourdage et al., 2018), qualitative research in this area could be very insightful.

In addition to these limitations, the different models that emerged from the data offer numerous opportunities for future research. Since the present study is a qualitative study, further quantitative studies should be conducted to substantiate our results. In the following, we would like to give examples of possible future studies that we believe can be carried out on the basis of the data.

First, we were able to identify the behavior of the interviewer as the first major influencing factor. Here we found that closeness and distance on the part of the interviewer can be decisive in determining whether applicants answer predominantly honestly or also dishonestly. Future experimental studies could address these issues by specifically varying the interviewer’s behavior to more closely examine its influence on applicants’ exhibited IM. Similarly, studies could ask applicants about the IM tactics they used and about the behavior of the interviewer or might even consider the natural occurrence of IM as a response to situational cues in observational studies or on the basis of interview transcripts (e.g., Wilhelmy et al., 2021).

Second, informants told us of many other interview questions that influenced their IM behavior and for which there has been no prior research, such as questions about applicant’s motivation, questions to test their suitability, questions about the person, position, and the organization. Thus, future research could use quantitative methods to compare these questions in terms of the IM shown. In addition, especially for questions about the person and position we found that positive affect was predominantly followed by honest assertive IM whereas negative affect was followed either by deceptive IM alone or by both deceptive and honest IM. Therefore, future studies should further investigate this finding and take affect into account.

Finally, our data suggest that affect may be fundamentally important when it comes to how applicants present themselves. Therefore, the relationship between affect and behavior should be investigated in more detail. For this purpose, applicants could be asked retrospectively about their IM in selection interviews and the associated affect.