What if you could ask a job applicant a few questions that would allow you to accurately predict whether he or she would be a good employee? Is this person likely to be absent frequently, falsify expense reports and claim credit for work done by other team members? Or, on the plus side, will he or she cheerfully volunteer for extra assignments, deal patiently with difficult clients, mentor junior associates and be a good leader?
The fact is, there is a reliable way to tell, but it is often neglected by employers and hiring managers: ask questions that elicit information about character. Too often employers focus narrowly on competence and prior experience. Yet evidence from recent psychological studies shows that people work harder and perform better at their jobs when they have high standing on three key character traits: honesty-humility, conscientiousness and guilt proneness.
Honesty-humility and conscientiousness are two of the six broad dimensions of personality summarized by the HEXACO model of personality structure. (The other four are emotionality, extraversion, agreeableness and openness to experience.) This six-factor model has begun to replace the famous “big five” model of personality, as mounting evidence bears out its validity. Honesty-humility, a factor not included in the old model, captures the motivational aspect of moral character—the desire to do good and avoid wrongdoing. Conscientiousness captures the willpower element of moral character—the facility and determination to do the right thing.
Guilt proneness is not specifically described by the HEXACO model, although it is related. This trait captures thoughts and feelings that are a blend of honesty-humility and conscientiousness. Guilt proneness specifies the extent to which a person would feel bad if he or she did something wrong, even if no one knew about it. It reflects a deep sense of responsibility for others, which helps to explain why recent studies from my team and others have found that a person's guilt proneness is an excellent predictor of job performance and leadership skills. My colleagues and I are now working to create interview questions that can elicit information about this trait in ways that are useful to employers. As more and more companies lean on personality testing to help get a clearer picture of job applicants, our results are revealing the benefits of focusing on character in addition to competence.
From Five Factors to Six
Through the 1990s the big five model of personality dominated psychology, based on a wealth of studies going back to the 1960s. This model describes five groups of traits, sometimes abbreviated with the acronym “OCEAN”: openness to experience, conscientiousness, extraversion, agreeableness and neuroticism. The model remains very popular, and assessments of the big five are still widely used by personality and organizational psychologists in research and practice. We now know, however, that this older model fails to capture critical elements of personality related to moral character and ethical behavior.
So how did the big five model go wrong? It emerged from empirical research that examined relations among a list of personality-related adjectives from the dictionary, such as “friendly,” “diligent” or “easygoing.” Research participants would rate themselves and other people on the extent to which each adjective described that person. The researchers then would examine the associations among the items through a statistical technique called factor analysis to determine the minimum number of different categories (factors) that were needed to reliably summarize the information. In this case, the adjectives clumped into five groups. This lexical approach to examining personality began in the 1930s but gained in popularity from the 1960s through the 1990s as statistical computing power improved.
One major limitation of the lexical studies that led to the big five is that they were based on limited lists of words in only the English language. In 1997 psychologists Kibeom Lee and Michael C. Ashton, then graduate students at the University of Western Ontario, decided to conduct a lexical study at a university in Seoul examining about 400 common Korean adjectives that could describe a person. By using this long non-English language word set, they found that there were six, not five, groups of traits. Three of these groups were basically the same as three from the big five model: conscientiousness, extraversion and openness to experience.
Emotionality (which is similar to the big five's neuroticism) and agreeableness also emerged in the new six-factor model, but they differed somewhat from their big five counterparts (some adjectives that were part of neuroticism in the big five model were part of agreeableness in HEXACO, and some adjectives that had been part of agreeableness now belonged to emotionality). Most interesting, a new group of traits emerged in this study that was not well represented in the older model: honesty-humility. This new factor contained adjectives such as ethical, honest, modest, loyal, sly, deceitful, greedy, pretentious and pompous.
The results from this study in Seoul led Ashton and Lee to wonder whether their findings were related to something specific to the Korean language or whether the six broad categories of traits they found were, in fact, universal. Was it possible that all the prior English-language studies—decades of research—had missed the honesty-humility dimension? Keeping in mind that the earlier studies all relied on relatively small lists of words because of historical limitations in computing power, Ashton and Lee ran more studies in English and other languages. They also looked back at old English-language data collected by other researchers in the past that could now be analyzed with more powerful computers. The six-factor HEXACO model emerged every time.
By 2007 the HEXACO model of personality structure had been confirmed in a dozen different languages. No more than six categories of traits have ever been reliably obtained in lexical studies in any language that has been researched so far. Today, 20 years since their initial study began, the “big six” model that Ashton and Lee termed HEXACO continues to be supported, with new evidence published regularly in top-tier psychology journals.
Predicting Job Performance
In the years since honesty-humility was revealed as a distinct element of personality, psychological studies have linked it to a variety of traits and behaviors. One of the most robust lines of inquiry has to do with behavior on the job—specifically, workplace deviance and organizational citizenship. Workplace deviance such as theft, sabotage and withdrawal by employees costs organizations billions of dollars a year and has profoundly harmful effects on employees in organizations where such abuses occur. On the flip side are helpful work behaviors—known as organizational citizenship behaviors, or OCBs for short. Examples include volunteering to mentor co-workers, switching schedules or working extra hours to accommodate others' needs, coming in early or working late to ensure projects get done, and lending a compassionate ear to colleagues having problems.
To examine the relation between character and job performance, my colleagues and I conducted a three-month study of approximately 1,500 employees in hundreds of different organizations and different jobs across the U.S. In 2011 and 2012 we surveyed employees weekly about their workplace deviance (anything from stealing pens to faking illness) and organizational citizenship. Prior to finishing the weekly surveys, they completed an initial questionnaire about their personality and the organization where they worked. We also surveyed their co-workers about the employees' job performance over the course of one month.
The results, which we published in 2014, revealed that honesty-humility, conscientiousness and guilt proneness each predicted workplace deviance and organizational citizenship. Both self-reports and co-worker reports revealed many more misdeeds for those low in these traits [see graphs below]. The fact that co-worker reports aligned with the self-reports rules out the possibility that some people are simply more willing than others to admit to wrongdoing.
The relations between organizational citizenship and these character traits were less clear but still suggestive of more helpful behavior among those high in honesty-humility, conscientiousness and guilt proneness. For example, employees with relatively low guilt proneness scores reported that they committed an average of 14 OCBs per week, whereas those with relatively high guilt proneness scores reported an average of 17. We found no significant differences in self-reported citizenship behavior between those low and high in honesty-humility or conscientiousness, but their co-workers did see significant differences. For instance, people low in guilt proneness were seen committing an average of 22 citizenship acts in a month, whereas those high in guilt proneness had an average of 37. This pattern was the same for honesty-humility and conscientiousness.
We do not know why stronger patterns emerged in the co-worker reports of good deeds than in the self-reports. One possibility is that employees who score high in honesty-humility, conscientiousness and guilt proneness are less aware of the helpful acts that others spot. That is, their humility and task focus are so strong that they are not fully conscious of their own altruistic behaviors. Similarly, it is possible that those who have low levels of these traits hold an overly generous view of their own positive behavior.
Subsequent studies by my team and others have borne out the importance of these character traits for job performance. For example, a 2016 study my colleagues and I conducted of law-enforcement job applicants in Colorado found that nearly one third of those with relatively low guilt proneness scores (the bottom quartile of the sample) had been fired from a previous job. Moreover, many of the job applicants with very low guilt proneness scores (the bottom 10 percent) were deemed unsuitable for hire by a psychologist retained by the law-enforcement agencies to assess job candidates.
Our study of workplace deviance and citizenship provides clear evidence of the damage that employees with low levels of honesty-humility, conscientiousness and guilt proneness do to their colleagues and the organizations where they work. But are they also worse leaders?
Although research on the HEXACO traits and leadership is limited, there are dozens of studies on the big five personality traits that point to the importance of conscientiousness for leadership emergence and effectiveness. The dependability and achievement focus of individuals high in this trait help them rise to the top and be effective in positions of power. Strong evidence for this relation comes from a meta-analysis by Timothy Judge, then at the University of Florida, and his colleagues. His team conducted a quantitative and qualitative review of more than 70 studies of the big five traits and leadership. Among all the big five traits, conscientiousness was second only to extraversion in its impact on leadership. Extraversion is characterized by high levels of dominance and sociability, so its strong relation to leadership is to be expected. But the effect of conscientiousness was almost the same magnitude.
As for honesty-humility, high levels of this trait have been shown to predict ethical leadership. In 2012 Reinout de Vries of the VU University Amsterdam found that business leaders from a variety of small and large companies who had higher levels of honesty-humility were rated as more ethical by their subordinates. Consistent with the idea that it is the combination of high honesty-humility and conscientiousness that makes for a particularly ethical person, conscientiousness was also found to correlate with ethical leadership.
Guilt proneness has been found to predict leader effectiveness. Among the studies showing this is one published in 2012 by Rebecca Schaumberg, now at New York University, and Francis Flynn of Stanford University. The researchers surveyed supervisors, colleagues and direct reports of 139 students entering a master of business administration (MBA) program at an elite U.S. business school. The surveys contained questions about various facets of leader effectiveness: the ability to achieve results, manage others, manage oneself and communicate effectively. The students themselves were surveyed about their guilt proneness and sense of responsibility for others, as well as the big five traits. Their intelligence was examined by collecting their scores on the Graduate Management Admission Test (GMAT).
MBA students who tested as having higher levels of guilt proneness reported a greater sense of responsibility for others and were rated as more effective leaders by their former supervisors, direct reports, peers and clients. Positive relations with leadership effectiveness also emerged for conscientiousness and extraversion, but these correlations were weaker.
Although plenty of research indicates intelligence is important for many aspects of job performance, Schaumberg and Flynn found no relation between leader effectiveness and GMAT scores. GMAT scores were also unrelated to guilt proneness, conscientiousness and sense of responsibility for others (and GMAT scores were negatively related to extraversion), suggesting that selecting employees based exclusively on standardized test scores or other indicators of intelligence would be a mistake. Instead the results of this study suggest that organizations seeking effective leaders should consider guilt proneness in their hiring and promotion decisions. Employees who anticipate that they would feel bad about their behavior after doing something wrong are better able to get along and get results.
In thinking about these findings, it is important to keep in mind that not all organizations desire an ethical and responsible leader, though when asked, most will say that they do. There are indeed certain tasks and roles in organizations—such as cost-cutting and downsizing— that require a cold, self-focused approach. For those types of jobs, a Machiavellian narcissist would likely do just fine—too much empathy and concern for other people would be a liability. That said, such tasks and roles are relatively rare as compared with ones that involve getting along and getting ahead by working with and leading others. Insensitive and unethical behavior may, in some circumstances, bring positive business results in the short term, but there is a real risk of long-term damage for the organization and society if individuals with low levels of moral character are appointed to high-level leadership positions.
One final note: Organizations must be cautious in appointing leaders who say they would not feel uncomfortable about a decision that harms other people. Sometimes leadership requires such decisions, but a person with this level of indifference is likely to choose the selfish path when deciding between options that lead to personal advancement and those that benefit the organization at large.
Character Assessment on the Job
Psychologists have developed and validated self-report and peer-report questionnaires for measuring honesty-humility, conscientiousness and guilt proneness. These assessments do a very good job of measuring character in anonymous research settings, but how do they fare in high-stakes settings, such as hiring?
As you might imagine, character assessment becomes more difficult when individuals are highly motivated to make a positive impression. This is not to say that personality tests do not have value in such settings—indeed, they are widely used by organizations, and their validity has been substantiated with solid evidence from psychologists Robert Hogan and Joyce Hogan of Hogan Assessment Systems, as well as many others over the past several decades.
Still, the century-old argument that “faking” invalidates personality assessment has not been entirely debunked, and the debate continues. Self-reports and peer reports do have value in high-stakes settings, but they are not infallible, and their predictive power is far from 100 percent. Therefore, it is critical that organizations complement personality testing with other methods.
Behavior-based interview questions are one option. In my current research with Yeonjeong Kim of Carnegie Mellon University and Abigail Panter of the University of North Carolina at Chapel Hill, we are exploring whether valid information about a person's character can be gleaned from such questions. Here is one that seems to work well:
“Please describe a time when you made a mistake at work. How did you feel when this occurred? What did you do? What, if anything, did you learn from this experience?”
When people respond to this set of questions, they often communicate information about their guilt proneness, honesty-humility and conscientiousness, which in turn can be used to predict ethical decision making and behavior. Our research on detecting character via interview questions is still in its early stages, but our results thus far are promising. Some studies suggest that it may be possible to train people to correctly identify personality traits from such open-ended interview questions. For example, a 2016 study by psychologists Deborah Powell of the University of Guelph and Joshua Bourdage of the University of Calgary found that undergraduate business students could be trained to more accurately identify key personality traits from videos of interviews.
Knowing that high levels of guilt proneness, honesty-humility and conscientiousness are hallmarks of an ethical employee gives us strong clues as to what we should look for when making hiring decisions and promoting people to leadership positions. When evaluating a job candidate or future leader, ask yourself:
Would this person feel bad about committing a transgression or making a mistake even if no one knew about what he or she did? Does he or she have a strong sense of responsibility for others? Would this person feel bad about letting others down? Is this person truthful, humble and fair? Is he or she hardworking, careful and thorough when completing tasks?
If the answer to these questions is no, then the individual is unlikely to be an ethical worker and will probably be an ineffective and disreputable leader. Conversely, if the answer is yes, you can bet that he or she will be a good colleague and star performer in the organization—a person who will exhibit decency and integrity when called upon to lead.
Guilt vs. Shame
After realizing you have received too much change at a store, you decide to keep it because the salesclerk doesn't notice. How likely are you to feel uncomfortable about keeping the money?
Researchers use such questions to assess guilt proneness, a personality trait that encompasses aspects of honesty-humility and conscientiousness. With colleagues at the University of North Carolina at Chapel Hill in 2011, I developed the Guilt and Shame Proneness scale—GASP—a now widely used tool to measure people's propensity to experience guilt and shame in response to wrongdoing. Studies using the scale have shown that guilt proneness predicts a variety of positive behaviors inside and outside the workplace, including being an honest and trustworthy negotiator, being a responsible and valued team member, and being a good leader and star performer at work.
Guilt proneness—the personality trait—taps into psychologically healthy guilt, not the unhealthy, debilitating kinds of ruminations that might lead to anxiety, depression or mental illness. The latter types of feelings, I would argue, are not truly guilt but rather shame. Shame is feeling bad about oneself, whereas guilt is feeling bad about one's behavior. The former leads people to hide or lash out at others. But the latter leads to positive behaviors that help rather than harm others, such as apologizing and trying not to err again, and that is why psychologists recognize guilt as the quintessential moral emotion.
To find out your level of guilt proneness, take a quiz at www.guiltproneness.org. —T.R.C
Can a dishonest person change
Honesty-humility, like all the HEXACO personality dimensions, can be thought of as a broad habit—or the way someone behaves most of the time. Yet unlike, say, extraversion, studies show that honesty-humility scores tend to increase substantially across our life span. In other words, a person who is dishonest as a teenager is not doomed to be a dishonest adult. For most of us, honesty-humility and guilt proneness increase dramatically from about the time we enter adult life, around 20 years old, until we are in our 60s.
Another way to conceptualize change in personality is to consider “rank order” stability—that is, the extent to which a person who is relatively high (or low) on a given trait will continue to rank relatively high (or low) over time compared with peers. Rank-order change from young adulthood to older age is, in general, quite minimal. This means that a young person who is more honest and conscientious than his or her classmates will likely continue to be relatively more honest and conscientious than same-aged adults as he or she grows older.
This stability is not predestined—people can and occasionally do change their relative standing. But dispositions and habits emerge very early in life and are based in part on genetic influences. The nature-nurture debate continues to be a lively one, but the emerging consensus from twin studies is that approximately 60 percent of the variation in personality differences can be attributed to genetic influences. This appears to be just as true for moral character traits as it is for other characteristics. —T.R.C
How Ethical Are You?
The psychologists who developed the six-factor HEXACO personality model, Kibeom Lee and Michael C. Ashton, have made their personality test available at http://hexaco.org/hexaco-online. There you can size up your own levels of honesty-humility, along with emotionality, extraversion, agreeableness, conscientiousness and openness to experience.