Personnel selection is among the most important human resource activities in an organization, which, however, is threatened by frequent occurrences of faking (Donovan et al., 2014; Douglas et al., 1996). Faking in noncognitive assessment procedures can negatively impact the validity of these procedures (Douglas et al., 1996), the utility of hiring decisions through false positives, and job performance (e.g., Law et al., 2016; O'Neill et al., 2017). Despite many years of research in the area of faking and its predictors, there is still a considerable lack of knowledge about the antecedents of applicant and employee faking during selection. Therefore, the main purpose of the present study was to provide empirical evidence for understanding the faking behavior of employees working in an organizational context based on the antecedent’s interplay. Particularly, an interaction of two suggested antecedents (Roulin et al., 2016) - personal integrity and competition, which were separately examined in past research (e.g., Buehl & Melchers, 2018; O’Neill et al., 2013), − is considered in the frame of the signaling theory (Bangerter et al., 2012).

Theoretical models identify several predictors of faking. These include characteristics on the part of the respondent, such as personal integrity or interpersonal skills (e.g., O’Neill et al., 2013), as well as situational variables serving as environmental signals, such as the competitive organization of the selection process (Canagasuriam & Roulin, 2021; Roulin & Krings, 2020). Since empirical testing of the predictors is time-consuming and technically challenging, previous research has delineated directions for further research for the models of faking to be empirically tested (e.g., Goffin & Boyd, 2009; Levashina & Campion, 2007; McFarland & Ryan, 2006; Roulin et al., 2016). Consistent with the model of faking based on the signaling theory proposed by Roulin et al. (2016), we tested the interactions among the antecedents of faking in the current study, including personal integrity and situationally signaled competitiveness. We followed this model because of its emphasis on the interplay between situational and dispositional factors. In particular, it has been suggested that if personality matches with a relevant situation-specific cue (e.g., competition), faking intention can be triggered more (Bill et al., 2020).

Furthermore, Roulin et al. (2016) noted that research examining the link between perceived competition and faking is limited, which was followed by studies reflecting the predictive power of competitiveness in faking on the organizational level and in interviews (e.g., Canagasuriam & Roulin, 2021; Ho et al., 2020; Roulin & Krings, 2020). Therefore, this study aimed to test the moderating role of perceived level of competition on the relationship between personal integrity and faking behavior. We conducted a between-subject experiment where all participants were randomly assigned either to competition or non-competition conditions. Participants in both conditions provided data on personal integrity and social desirability, which was used as the measure of faking. We then tested whether participants who scored lower on integrity and were exposed to the competition condition (compared to participants in the non-competition condition) scored higher on social desirability or faking.

Faking

The tendency of respondents to intentionally distort and influence the results of non-cognitive measures is closely related to the use of personality questionnaires and other assessment methods in organizational practice. Therefore, it is critical to understand and recognize such behavior in all diagnostic contracts involving decisions affecting the ability to meet the understandable needs of those being assessed. In particular, when applicants are selected for desired positions or employees are evaluated for their performance, the distortion of their answers in questionnaires, called faking, has specific sources and dynamics.

Various terms are used in the literature to refer to faking, such as social desirability, impression management, self-enhancement, and response distortion (McFarland & Ryan, 2006; Ones & Viswesvaran, 1998). Faking can be defined as the tendency of respondents to intentionally provide inaccurate answers that they believe will increase the chances of achieving a valued outcome (Goffin & Boyd, 2009). Several models have attempted to explain organizational faking behavior (Goffin & Boyd, 2009; Levashina & Campion, 2007; McFarland & Ryan, 2006; Roulin et al., 2016). Models converge in identifying dispositional and situational factors for faking behavior. We identify three main factors that lead to faking: a) the ability to fake, b) the perceived opportunity in a situation for such an act, and c) personality traits associated with faking (Goffin & Boyd, 2009; McFarland & Ryan, 2000; Roulin et al., 2016). The ability to fake indicates the extent to which a person succeeds in twisting her/his answers in the desired direction to increase person-job/organization fit (Levashina et al., 2009; McFarland & Ryan, 2006). The ability to fake can be considered a prerequisite for faking well but does not automatically lead to faking, as the respondent can simply choose not to fake.

To gain deeper information regarding situational factors, we employed signaling theory as a framework because of its emphasis on the information exchange between the participants of the job market with different interests (Bangerter et al., 2012) which has been extensively used in studies examining counterproductive work behavior and, in particular, faking (Canagasuriam & Roulin, 2021; Roth et al., 2021, 2023; Roulin & Krings, 2020). According to the framework of Bangerter et al. (2012) based on the signaling theory, organizations and applicants can both benefit from exchanging honest signals to accurately evaluate the prospective fit. However, these parties can have conflicts of interest and incentives for faking, which is considered an adaptive response to changing environmental cues in that framework. That is the case of competitiveness, the widely examined antecedent of faking, especially in job interviews. Organizations may express a low acceptance ratio or manifest competitive clime and/or culture as signals towards applicants (see Roulin et al., 2016). In the response, applicants may adapt their signals towards organizations in a striving to succeed and thus establish the faking dynamic (Ho et al., 2020; Roulin & Krings, 2020). Following this view, faking was conceptualized as an adaptive response to an environmental cue implying competition in this study.

Finally, we focus on personal integrity in terms of individual differences, which has been identified as a potential source for faking in a model adopted in the current study (Roulin et al., 2016). The interest in personality traits underlying unethical, dishonest, counterproductive work behavior or faking broadly attracted the attention of researchers in the last decades. Personal integrity plays a crucial role among the examined traits (for a comprehensive review see Giordano et al., 2020).

Methods of measuring faking in organizational research

In the field of intentional deception research, we most often encounter a repeated-measures within-subjects design in which respondents are administered the same method under different conditions with a specified time interval between measurements. The degree of faking is then operationalized as the difference between the measured values (McFarland & Ryan, 2000). These studies are typically conducted in an experimental setting, usually with college students. The need for repeated measures is technically challenging in a real-world setting. Although a within-subjects design has more statistical power and internal validity, it is not practical to use in employee samples because a within-subjects design does not represent the face validity of situations that occurs only once, and usually inhibits natural reactions. Given that there should be a certain amount of time between the manipulation (competition vs. non-competition), drop-outs would also be highly possible. In terms of resembling the natural work setting, the amount of time between the competition vs. non-competition manipulation can also be tricky to determine. Another problem would be a carryover effect, especially for participants who receive competition manipulation first. An alternative is a between-subjects study design which has less power but is more aligned with the real-life situation in that study. In this case, two independent groups randomly assigned to the conditions are compared and differences are examined (Hogan et al., 2007).

Currently, social desirability scales constitute one of the common methods of detecting faking. Zerbe and Paulhus define social desirability as “the tendency of individuals to present themselves positively in relation to current social norms and standards” (Zerbe & Paulhus, 1987, 250). In general, we can identify two main factors of social desirability: impression management (IM) and self-deception (Paulhus, 1984). The self-deception scale involves the unconscious tendency to put oneself in a better light. This is a type of presentation in which a person positively overestimates his or her socially desirable characteristics and truly believes that he or she is telling the truth. Individuals who score high on this scale have low self-knowledge (Preiss & Mačudová, 2013). This scale is relatively closely related to personality and is therefore less susceptible to distortion due to the reasons already mentioned. The impression management scale, on the other hand, measures the extent to which a good impression is sought. Individuals who score high on this scale are attempting to make a desirable positive response (Paulhus, 1988). This scale is more influenced by situational factors and individual motives (Paulhus, 1991). Impression management is considered the only measure that is susceptible to deliberate bias (de Vries et al., 2014). Therefore, we used the impression management scale in the present study as a measure of faking.

Social desirability scales are highly susceptible to deliberate distortion, as it is relatively easy to identify the desired direction of the response. Viswesvaran and Ones (1999) found that a “fake good” instruction increased the score by more than one standard deviation compared to an honest administration. For this reason, the use of social desirability scales to detect faking has been criticized. The degree of distortion on the social desirability scales is not the same as the bias on other content scales. Therefore, distortion on other scales cannot be corrected based on the social desirability scale (Hough, 1998). Social desirability also follows a response style that refers to the content of individual items rather than their social desirability (Kuncel et al., 2012). However, social desirability items often contain assertions that are unlikely to have been committed, such as “I have never littered on the street.” A person who agrees with such an assertion is likely to be misrepresented, even though the claim may be true (Griffith & Peterson, 2008). Therefore, the use of social desirability scales could be beneficial for identifying faking because the desired response is relatively easy to identify, whereas the faking itself can only be assessed by comparing the results of between-subjects designs.

Personal integrity

Based on signaling theory, people are sensitive to high stake situations and competitiveness signaling (Bangerter et al., 2012; Ho et al., 2020; Roulin et al., 2016; Roulin & Krings, 2020), which points to the importance of competitive dynamics on faking. Providing accurate information about themselves to organizations can be risky for employees especially when this is not to their advantage (e.g., when highly competitive cues are in the environment). Then the question arises: Who will fake information to strengthen their position when competitiveness is signalized in high stake situations? One proposed variable is integrity. Personal integrity is proposed to be one of the personality traits that significantly influence the degree of motivation to fake (Goffin & Boyd, 2009; O’Neill et al., 2013). In terms of individual differences, personal integrity includes reliability, honesty, and morality (Noelliste, 2013). The topic is of interest not only to researchers, but also to organizations, as any organization prefers honest, reliable, and moral members. In addition, integrity test scores seem to have additional value in evaluating members or candidates for certain positions. It has been found that individuals with low scores in integrity, honesty-humility, and morality are more likely to perform response distortion on personality tests, to be hired, and engage in deviant behaviors in their current job because they have an increased tendency to deceive (O’Neill et al., 2013). In addition, Ones et al.’s (1993) comprehensive meta-analysis showed that integrity tests play a significant role in predicting overall job performance and counterproductive work behavior (CWB). A more recent meta-analysis found a somewhat smaller but still moderately large association between integrity tests and overall work performance and CWB (Van Iddekinge et al., 2012). Thus, previous research has demonstrated the incremental validity and practical utility of using an integrity test along with tests of cognitive ability in personnel selection procedures (e.g., Giordano et al., 2020; Marcus et al., 2016; Ones & Viswesvaran, 1998a; Schmidt et al., 2016).

Since there are several different traditions from which the concept of integrity comes, there is no single definition of integrity. The two most commonly used traditions are the philosophical perspective and the empirical perspective, which distinguishes between overt and personality-based tests of integrity. In the present study, we adopted the approach proposed by Dunn (2009) and further developed by Schlenker (2008) and Seitl et al. (2022), which is rooted in the philosophical tradition. According to this view, integrity is defined as a commitment to moral standards that encourages individuals to act in accordance with a set of moral values that emerge from cultural or social norms that have become evident over time and in different social contexts (Dunn, 2009; Schlenker, 2008). To assess integrity in the present study, we used the Occupational Integrity Scale (OIS; Seitl et al., 2022), which was developed following the philosophical tradition. The instrument has three factors. Briefly, the first factor Reliability refers to the image of a “good employee” in relation to work and others. The second factor adherence to principles concerns whether a person identifies with larger, more universal ideas that transcend the workplace. The third factor, moral sense, which represents applied moral principles, typically involves anonymous application in social contexts beyond the work environment. It also reflects attitudes or experiences of integrity concerning specific issues. These factors are consistent with the theoretical background of integrity (Seitl et al., 2022).

Recently, previous research has shown that IM is positively associated with virtue-related personality traits such as honesty/humility/integrity and honest behavior (de Vries et al., 2014; Müller & Moshagen, 2019; Příhodová et al., 2021; Zettler et al., 2015). The impression management is primarily related to the Honesty-Humility dimension of HEXACO due to its emphasis on honesty and integrity (de Vries et al., 2014). It has been suggested that IM functions similarly to any other personality trait, particularly in low-demand situations, and assesses true virtues (Zettler et al., 2015). Thus, it has been suggested that IM captures both true virtues and self-presentation. The interpretation of high scores in IM does not necessarily indicate the presence of self-serving response biases that may undermine the validity of the assessment (Müller & Moshagen, 2019). Based on this evidence, we expect that all subscales of OIS would be positively associated with IM. More specifically, since the reliability subscale of OIS overtly measures what it is supposed to measure and reliability concerns the image of a good employee in the eyes of others (Seitl et al., 2022), we expect a positive association between reliability and IM (H1a). Similarly, we also expect a positive association between adherence to principles and IM since adherence is about individuals’ identification with universal principles beyond the workplace, which reflects true virtues (H1b). Finally, we hypothesized that there would be a positive association between IM and morals sense which refers to applied moral principles entailing the anonymous application beyond work context (H1c).

Competitiveness

According to Ho et al., 2020, Roulin and Krings (2020), Roulin et al. (2016), and Tett and Simonet (2011), competitiveness is another predictor of faking. Kohn (1992) distinguishes between two basic types of competition: structural and intentional competitiveness. Structural competitiveness is tied to a specific situation and strongly focused on the outcome in terms of profits. This type of competition is used primarily in situations where an individual’s gain means that others lose. Structural competition can also be present in situations where the individual competitors do not interact with each other, just the awareness of their existence can lead to the activation of competition. An illustrative example would be telling trainees in a company that only half of them will receive a permanent job offer. Another type of competition is called intentional and involves the individual’s inner need to be the best. In this case, it is the individual’s own proactivity to be better than others. Compared to structural competition, here there is no need for rewards, wins, or knowledge of other competitors (Kohn, 1992). Roulin et al. (2016) consider structural competitiveness as a type that is more strongly associated with the willingness to fake.

Roulin et al. (2016) offered a model on the dynamic aspects of faking rather than a static approach focusing only on the applicant characteristics and selection tools. Therefore, we adopted the perspective of Roulin et al. (2016) stemming from the signaling theory because of the dynamic representation of faking to better understand faking behavior in a competitive situation. According to this model, the intensity of competition signals leads applicants to consider faking as an adaptive strategy to outperform their competitors and get the job. Moreover, Roulin and Krings (2020) have shown that competitiveness is one of the key elements of corporate culture and that applicants adapt their answers in personality tests to present an ideal profile that corresponds to these elements. In the present study, we, therefore, adopted the perspective of structural competition.

Previous research has shown that perceptions of competitiveness can be activated through various methods, such as informing applicants that selection rates are low (Tett & Simonet, 2011) or instructing participants to read texts about the competitive nature of an organization (Roulin & Krings, 2020). In the present study, employees were given a text informing them that their scores would be compared with employees in other professions. It should be noted that in the present study, employees already working at a company were given competitive instruction, not candidates applying for a new job at the company. Since faking includes a dynamic rather than a static thought process (Goffin & Boyd, 2009), it can also be observed in employees who have been working for a while. We already proposed that reliability, adherence, and moral sense would be positively associated with IM. Based on the propositions of signaling theory, these associations will be moderated by competitiveness. More specifically, we hypothesized that respondents with lower scores on reliability (H2a), adherence (H2b), and moral sense (H2c) and exposed to competition-inducing instruction will score higher on IM than respondents in the control group (non-competition condition) since providing honest information wouldn’t be to employees‘benefit and can be risky in a competitive environment. Thus, we expect that having lower scores on personal integrity will lead to higher scores on IM for the participants receiving a competitive cue.

The present study

Although this is a dynamic area of inquiry since organizations have been actively screening employees and applicants, there have been few empirical studies on the interplay between integrity and faking over recent years. Faking has been estimated to happen in 30–50% of job applications, resulting in changes in applicant rankings (Griffith et al., 2007), and more than 90% of candidates are faking in interviews (Levashina & Campion, 2007). These findings suggest that faking is a widespread problem that requires investigation with novel instruments and methods. Based on the model offered by Roulin et al. (2016), and the following results supporting the model (see Canagasuriam & Roulin, 2021; Ho et al., 2020; Roulin & Krings, 2020), the interaction between antecedents of faking was tested. The present study contributes to the literature by examining the moderating role of competition in the link between personal integrity and faking. In addition, it has been pointed out that overreliance on student samples may lead to an inaccurate representation of actual applicant faking (Griffith et al., 2007). For this reason, we used a between-subjects design that included employees already working in different organizations. Overall, this study aims to expand the current understanding of when employees fake by highlighting the role of competition as a situational factor.

Methods

Participants

The data of the present study were collected as part of a larger project. Initially, 7367 primary school teachers from 6 Czech districts (about 25% of the population) were contacted by e-mail and asked to participate in the study. Nine hundred and twenty-two of them completed the online experiment (12% men). Participants were given the immediate opportunity to delete their responses after reading the debriefing form at the end of the experiment if they felt uncomfortable with being measured in integrity. Forty-three participants withdrew from the study. Moreover, 3 participants who gave an unrealistic age were also excluded. We performed an a priori power analysis using G*Power to determine the required sample size for achieving a power of .95. According to power analyses, 98 participants were required to detect a medium effect at a significance level of 0.05. Cohen’s (1988) criteria for effect size indicate that f2 values of 0.02, 0.15, and 0.35 or higher represent small, medium, and large effect sizes, respectively. Given the effect sizes for situational factors (e.g., competitiveness) on faking were small to medium in previous research (e.g., Ho et al., 2019), we used .20 as the effect size. Since participating in an online experiment can be demanding for employees, we reached as many employees as possible considering the dropout rates. For the present study, we narrowed the sample to 470 female elementary school teachers of productive age between 25 and 50 years, 229 in the experimental group (M = 38.86, SD = 7.46) and 241 in the control group (M = 39.11, SD = 7.12). In selecting participants for the narrow sample, we aimed for homogeneity. Therefore, all same-sex participants in young and middle adulthood with work experience were selected. The mean age of all participants was 38.99 years (SD = 7.28). Teachers perform a similar job in a similar setting. The study was approved by the Ethics Committee of the Department of Psychology at Palacky University.

Materials and procedures

To test the proposed relationships, we conducted an online experiment using a between-subjects design. The online platform, developed for both the experimental and control groups, included the informed consent form on the first page. The following page contained the Occupational Integrity Scale. The third page of the online platform differed by the presence of the experimental variable for the experimental group. Randomization of respondents was ensured by a defined script after they clicked on the provided link. Figure 1 illustrates the steps of the online experimental procedure.

Fig. 1
figure 1

Details of the experimental procedure

Full size image

Participants in the experimental group were presented with the following paragraph, the participants in the control group were not shown any paragraph.

“On the next page, you will find a second questionnaire, the results of which will be compared with the results of other professions. The aim is to find out the position of teaching staff among representatives of other monitored professions in socially desirable characteristics.”

After presenting this text to participants in the experimental group, the rest of the procedure was the same for both groups again. The fourth page contained the social desirability scale. Finally, at the last page, we offered a clear description of the study’s aims with the direct option to delete participants’ records and quit the study in case of unpleasant feelings. The contacts for the study authors were repeated.

Measures

Personal integrity was measured with the Occupational Integrity Scale (OIS Schneiderová, 2017; Seitl et al., 2022). Items were generated based on Dunn’s (2009) definition of integrity. The monitored characteristics include belief in general rules, reliability, credibility, unwillingness to violate one’s own policies, and selected examples of counterproductive work behavior, which were added in the later development of the scale. OIS consists of 23 items corresponding to 3 main factors: reliability, adherence to principles, and moral sense; the residual invariance model shows a good fit to data, χ2 (487) = 1025.382, CFI = 0.96, RMSEA = 0.05. Briefly, the reliability subscale describes the characteristics to honor the promise (e.g., It’s important to be at work on time), adherence to principles describes the characteristic of generally standing by its principles (e.g., Stand by your principles even if it is disadvantageous), and moral sense describes the tendency to act in accordance with moral values and all the items in this subscale are reverse (e.g., A justified theft can be excused) (Seitl et al., 2022). Items were rated on a 1 (totally disagree) to 5 (totally agree) scale. Cronbach’s α reliability coefficient of the reliability subscale was .73, .for moral sense .64, and for adherence .68 in the present study.

Social Desirability was measured with the Czech version of Paulhus’s Balanced Inventory of Desirable Responses from 1988 (BIDR-CZ), which was translated into Czech by Preiss and Mačudová (2013). This is a translated sixth version of the questionnaire containing 40 items spread over two scales, a scale of self-deceptive enhancement (SDE) and a scale of impression management (IM). The first 20 items measure SDE, and the remaining 20 items measure IM. Briefly, BIDR stresses exaggerated assertions regarding positive cognitive traits (e.g., overconfidence in one’s judgments and rationality). As already mentioned, we only used IM for this study. IM refers to the presenting self to an audience on purpose. Participants rated the items on a seven-point scale ranging from 1 (totally disagree) to 7 (totally agree). Example items from the scale: “I never take things that don’t belong to me” (IM). Cronbach’s α reliability coefficient is .75 for IM in the present study.

Data analysis

Prior to the main analyses, the normality assumption was checked by calculating the values for skewness and kurtosis for IM. The results showed that the values for skewness ranged from .04 to −.36 and for kurtosis from .14 to .16. According to the standards described by Curran et al. (1996), IM showed sufficient normality. Preliminary analyzes were then performed with respect to descriptive statistics and correlation among the main variables. Then, IM was predicted based on the subscales of the OIS via hierarchical regression analyses. Age was controlled for in these analyses. Finally, to examine the moderating effect of competition (i.e., 1 = competition, 2 = no competition) on the relationship between integrity and IM, Process macro for SPSS was used with Model 1 (Hayes, 2013).

Results

Descriptive statistics and correlations between study variables were displayed in Table 1. IM was correlated positively with the subscales of OIS.

Table 1 Correlations between study variables and descriptive statistics
Full size table

Hypothesis testing

IM was predicted in hierarchical regression analysis. IM was predicted by all OIS subscales positively, which supported hypotheses H1a, H1b, and H1c. This model accounted for 34% of the variance in IM, [F (4, 469) = 60.18, p < .001] (Table 2).

Table 2 Predictors of IM
Full size table

Finally, the moderating role of competition on the relationship between OIS and IM was tested. The subscales of OIS were used separately in the analysis. A total of 3 different moderating relationships predicting IM were tested (i.e., reliability x group, adherence x group, moral sense x group). The analyses revealed that 2 of these interactions were significant in predicting IM. The interaction between reliability and competition (H2a) was significant (B = .44, SE = .19, p < .05, 95% CI = [.07, .81]). The moderating relationship was such that respondents with lower levels of reliability (− SD below the mean) in the experimental group scored, when exposed to competition-inducing instruction, higher in IM than respondents with lower levels of reliability in the control group (non-competition). According to simple slope analysis, this relationship is significant in both groups but it weakens with increasing reliability (see Fig. 2).

Fig. 2
figure 2

The group as a moderator of the relationship between IM and reliability. The moderating effect is graphed for: 1 = experimental group and 2 = control group

Full size image

The second significant interaction effect was observed between the competition and moral sense (H2c) in predicting IM (B = .19, SE = .10, p < .05, 95% CI = [.001, .39]) (see Fig. 3). A similar trend was observed with the previous moderation. The respondents with lower levels of moral sense (− SD below the mean) in the experimental group scored, when exposed to competition-inducing instruction, higher in IM than respondents with higher levels of moral sense in the control group (non-competition). According to simple slope analysis, this relationship is significant in both groups but it weakens with increasing moral sense (see Fig. 3). All in all, the findings revealed that our propositions were supported except for H2b. The perceived competition did not moderate the relationship between adherence to principles and IM.

Fig. 3
figure 3

The group as a moderator of the relationship between IM and moral sense. The moderating effect is graphed for: 1 = experimental group and 2 = control group

Full size image

Discussion

The main objective of this study was to investigate the moderating role of perceived competition in the relationship between integrity and faking in a model of faking proposed by Roulin et al. (2016) based on the signaling theory. Signaling theory describes faking behavior as a dynamic process in response to competitive signals in the environment (Bangerter et al., 2012). Thus, competition manipulation can increase the tendency of employees to fake to increase their fit into the organization. The predictors of IM were also examined using the OIS, a recently developed scale to measure integrity in the work setting. The results were consistent with expectations. IM was positively predicted by reliability (H1a), adherence to principles (H1b), and moral sense (H1c). More importantly, competition moderated the relationship between IM and reliability (H2a) and IM and moral sense (H2c). As expected, the difference between the competition condition and the non-competition condition with respect to IM was greater for participants who scored lower on reliability and moral sense (i.e., − SD below the mean). More specifically, participants with lower reliability and moral sense (− SD below the mean) in the experimental group, when exposed to instruction that encouraged competition, scored higher on IM than participants with lower reliability and moral sense in the control group (non-competition).

The predictors of faking

As expected, reliability, adherence to principles, and moral sense were positive predictors of IM. In other words, high expression of these integrity traits is associated with increased IM. These results contrast with the meta-analysis of Ones and Viswesvaran (1998b), who found a correlation of r = 0.08 between integrity and social desirability in a sample of 3973 respondents. The data come mainly from selection situations - the so-called “high stakes” situations in which we would expect motivation to fake to be even higher. The problem could arise from the nature of the impression management scale in the BIDR itself, as Müller and Moshagen (2019) suggested and Příhodová et al. (2021) recently supported. The impression management scale captures both true virtues and self-presentation, which, as mentioned earlier, could lead to the observed effect. Strong correlations were particularly found between the Honesty-Humility trait of the HEXACO personality model (de Vries et al., 2014) and could therefore also explain a strong relationship with integrity. For our study, this positive relationship between integrity and IM, as similarly described by Müller and Moshagen (2019) and Příhodová et al. (2021), has no impact on our main results due to the chosen study design.

The moderating role of perceived competition

Regarding competition manipulation, the role of the competition-inducing variable was operationalized in line with Kohn’s (1992) concept of structural competitiveness, and application of the signaling theory in an organizational environment (Bangerter et al., 2012). Structural competitiveness was considered to be more strongly related to the propensity to fake (Roulin et al., 2016), as it is linked to the dynamic circumstances of the environment (Kohn, 1992). Therefore, in the present study, we adopted the structural competition perspective instead of the intentional competition perspective, which focuses on attitude or personality traits (Kohn, 1992). Our results (H2a and H2c) were supportive of the role of competition as a buffer against excessive score inflation in the form of faking (IM) by individuals with low scores on integrity-related traits (reliability and moral sense). Namely, respondents with lower scores on reliability and moral sense and exposed to competition-inducing instruction scored significantly higher on IM than respondents in the control group (non-competition condition). That is, respondents with higher reliability and moral sense were not willing to fake to the same extent as the rest of the group. The results of the present study suggest that measuring integrity and competition-inducing instruction can be a good discriminator of employees willing to fake. Surprisingly, competition-inducing instruction did not moderate the relationship between adherence to principles and IM, indicating that H2b was not supported. Adherence to principles focuses on commitment to principles without concrete content (one’s own principles) that have probably nothing to do with perceived competition. Therefore, the relationship between a lack of commitment to one’s principles and IM was not influenced by competitiveness.

Our results are consistent with previous literature suggesting that situational factors signaling competition-inducing cues increase people’s intention to fake (e.g., Canagasuriam & Roulin, 2021; Ho et al., 2019; Ho et al., 2020). In addition, our study showed that competition is a relevant process among employees who are already working in a company. Although the hiring process is considered highly competitive (Roulin et al., 2016), the post-hiring process also contains competitive cues for employees. This is because some organizations have a more competitive work environment than others (Fletcher et al., 2008), which may influence employees’ desire to fake (Roulin et al., 2016). To gain a better understanding of faking behavior in competitive work environments, future research could consider intentional competition through the use of longitudinal studies. Although our findings do not directly relate to the selection process, they could also be applied to faking during the hiring process, which contains competitive elements (Roulin et al., 2016). However, it should be noted that the dynamics of a selection process and our experiment may differ. This may be an avenue for further research.

Among the many approaches to measure faking, we chose a between-subjects design. The Czech version of the Balanced Inventory of Desirable Responding BIDR-CZ was used to measure faking. The BIDR-CZ questionnaire was chosen despite the criticism of identifying intentional bias through social desirability scales due to the concept of the implemented design. The presented design does not use the social desirability scores to correct the scores of other methods, the scores themselves are compared in BIDR-CZ between the experimental and control groups. The reason for choosing BIDR-CZ was mainly the high susceptibility of the Impression Management Scale to intentional distortion (Lönnqvist et al., 2007). In addition, the Occupational Integrity Scale (OIS) was used to measure the construct of integrity, which is one of the few overt integrity tests available in the Czech Republic. Based on the work of Schneiderová (2017), Seitl et al. (2022), and the results of the present study, the scale appears to be a promising instrument for measuring the integrity of personality.

One clear managerial implication derived from the findings of this study is that competitive cues should be considered critical characteristics of an organizational climate, which may be incorporated into the decision-making process. The findings also point to the impotence of recognizing the interplay between personality and competitiveness, as a dynamic characteristic of an organizational climate. In this way, strategies to mitigate faking can be tailored effectively. Finally, it should also be noted that supporting a competitive environment can encourage faking inadvertently. Thus, strategic manipulation of competitiveness can be more helpful instead of encouraging a completely competitive climate in the work setting.

Limitations and suggestions for further research

Despite its strengths, this study has some caveats that we must mention. First, participation was voluntary, which may have resulted in enthusiastic teachers responding in a socially desirable manner. However, it should be noted that the method we used enabled us to compare the results of the experimental and control groups in terms of faking. Second, some of the insignificant results might be related to the fact that we used only the structural competition approach. This assumption, therefore, needs to be explored in more detail in future research. Using a real “high stake” situation and an alternative definition of a competition-inducing variable in line with Kohn’s (1992) individual-level structural competitiveness seems to be an appropriate design. Another approach may be to measure motivation for competition. Measuring the motivation behind competition may be a promising perspective to understand the source of motivation for this behavior. It can be either extrinsic or intrinsic, and this approach can provide a complementary perspective to structural vs. intentional competition. Finally, the present study used an overt test to measure integrity. Both overt and personality-based integrity tests can be used together to detect faking and provide more accurate conclusions in future research.

Conclusion

In summary, a strong positive relationship was found between the dimensions of integrity and IM, indicating the similarity of the constructs and providing additional support that IM also captures true virtues, as suggested by Müller and Moshagen (2019) and confirmed by Příhodová et al., 2021. Finally, the present study provided support for the model of faking by Roulin et al. (2016) based on signaling theory. Our findings suggest that workers may use strategies to cope with a highly competitive work environment, just as in the hiring process. Thus, detecting faking is critical not only during the hiring process but also afterward. Accordingly, we provided evidence of the effective role of competition manipulation against excessive score inflation in the form of IM by individuals with low integrity-related characteristics (reliability and moral sense). To conclude, this study shows that administering integrity tests and competition manipulation together appears to be a promising method for detecting IM in the work environment.