The Truth about Teambuilding>
The Personality Scam

Contents:


The Shaky Foundation of Myers-Briggs

I thought I had finally found it: scientific evidence that the Myers-Briggs Type Indicator (MBTI) is worth the time and money for teambuilding. You see, in all of my research, not once have I seen such evidence. I have heard people talk about how they improved their communications with one or two other people on their team after an MBTI exercise, and I believe them. Others have said it was “pretty accurate,” to use the most common phrase (not “completely accurate”). But the only unabashed support I have heard comes from the people who sell MBTI services, and managers who invested in them. Neither are very objective sources, and no one had return-on-investment (ROI) figures.

So I went looking through two databases of peer-reviewed journals. “Peer reviewed” means the article was reviewed by anonymous experts in the field. One database focused on business and the other on psychology and related sciences. Together they cover hundreds of journals going back decades. The number of articles I found referring to MBTI in a team setting: 2.

When I came across the first, I thought we had a winner: “Myers-Briggs Type Indicator: A bridge between counseling and consulting.” But then I came across a second from the same journal which turned out to be a response to the first. It turns out the author of the first article, Mary H. McCaulley, co-founded the Center for Applications of Psychological Type with Isabel Briggs Myers, co-creator of the test. None of the references in her articles were tests of the effectiveness of the test.

McCaulley wrote a description of how the MBTI was created; the theories of psychology pioneer Carl Jung upon which the test is based; and its model of personality. “The gist of the theory is that much apparently random… human behavior is actually quite orderly and consistent, being due to certain basic differences in the way people prefer to use perception and judgment,” McCaulley wrote. She says the MBTI tries to determine how a person prefers to become aware “of things, or people or occurrences or ideas,” and come to “conclusions about what has been perceived.” This results in classification of people according to four types. She goes on to point out that people who become top managers tend to be very different types from people who become counselors, and offers suggestions to help counselors who want to become business consultants on how to bridge the gap. She warns against some “misuses of type” including stereotyping and the use of results in hiring decisions. But, McCaulley concludes, “Counselors who wish to expand their counseling skills into consulting will find the MBTI a valuable tool to help the organizations and the people within them gain greater respect for their differences, work together more productively, and develop as individuals.”

David Pittenger responded in his article, “Although the MBTI is an extremely popular measure of personality, I believe that the available data warrant extreme caution in its application as a counseling tool, especially as consultants use it in various business settings.” Pittenger was a psychology professor at the Univ. of Tennessee at Chattanooga.

Pittenger says, “I suggest that many of the uses of the MBTI, as endorsed by McCaulley (2000), lack empirical support, and that consulting psychologists should consider these facts before using the instrument.” His first criticism is a point acknowledged by McCaulley, that MBTI suggests people can be grouped by clusters of personality traits in ways that makes them similar as a group and different from other groups.

But this requires us to fall into “either/or” categories. More scientific personality tests, in contrast, view personality traits as a continuous range. In other words, you may be an introvert, or you may be an extrovert, or you may be somewhere in the middle. The MBTI agreed with this initially, according to Pittenger: “In an earlier version of the MBTI, individuals who scored at the center of one of the scales received an x rather than a letter code… The developers of the MBTI subsequently abandoned that practice and replaced it with a tie-breaking procedure to avoid assigning intermediate scores.” In other words, the test forces you into a category even if you don’t really fit one. Pittenger notes that McCaulley’s data about what types tend to get into certain jobs does not tell you the degree to which those people differ, because it does not use continuous scales.

To illustrate, if you created a scale from 0 to 100 measuring how much someone likes bananas, the scoring method used by the MBTI would say you don’t like them if you scored a 49 and you do if you scored a 51. But two people who scored a 2 and a 98 are a lot more different than two who scored a 49 and a 51. Scientists would say the latter two are identical, while the MBTI would say they are different.

For mathematical reasons Pittenger details, this practice of the MBTI “may imply… significant personality differences where none exists” and means you cannot predict someone’s behavior based on the test as well as you otherwise could. Another research team found it removes about 30% of the information the test could provide about someone. He quotes another researcher as saying, “‘it would probably be better not to make these interpretations, or at least not rely on them heavily, with respect to important decisions about individuals.'”

Pittenger reports on several studies showing the results of the tests vary over short periods when retaken by the same people. One of the MBTI manuals reports a 35% change in type over 4 weeks, and another study reported 50% over 5 weeks. Since you can’t be sure whether the first test or second is closer to the truth, you would have to take the test a minimum of three times, and preferably more, to nail down an accurate type reading.

Also, at least nine studies have found the scales don’t really produce the type groupings MBTI supporters claim, Pittenger says. That is, when you group the results based on math alone, the groups don’t clearly match the groups the MBTI claims. Other sources, including an MBTI manual, show the test scales are linked (correlated). But they should not be related to each other if they measure separate parts of personality.

Other personality tests do not show the problems the MBTI does, Pittenger says. In particular, a type called  “Big 5” tests is drawn mostly from the results of many studies over decades, whereas the MBTI is based primarily on Jung’s theories developed decades ago and tries to fit test results to those. Pittenger adds, though, that no published study shows any personality test is better “for predicting job performance or other work-related behavior.” As long ago as 1996, Pittenger says, a review of studies on the MBTI in work situations showed that efforts to show simple links “‘between type preferences and managerial effectiveness have been disappointing.'” Personality appears to only account for about 20% of the difference between people in work performance, he reports.

Pittenger warns that acting upon what you know about someone’s type ignores how situations can change someone’s thoughts and behaviors. Studies show that personality is only one factor in someone’s decision style at a given moment. No one who has seen me do a presentation or training would think me an introvert, but that’s what I am. I treat my public work as a performance to make it interesting for attendees.

Pittenger writes from personal experience, “a wholesale application of the MBTI in an academic setting… resulted in the dissemination of dubious advice to students regarding their selection of courses and majors.”

In contrast to McCaulley’s 12 sources cited, almost all MBTI materials like manuals, Pittenger cites more than 50 sources, most of them studies from peer-reviewed journals unrelated to MBTI.

Ms. Briggs was not a psychologist: she had only a bachelor’s degree, in political science. And the MBTI was based almost exclusively on Jung’s theories. Though an influential theorist right up there with Freud, and the guy who came up with the terms “introvert” and “extrovert,” he is controversial within the scientific community and much of his theory work remains untested. Personality tests preferred by researchers are based on evidence guided by theory, not theory alone.

Even if the MBTI consistently helps most takers get along better with two teammates (the highest number I hear people say), it would be aiding less than half of the total number of communication channels within a team. Within roughly the same time taken by the typical MBTI exercise, and for free, you could draft:

  • a communication plan covering who needs what information from whom by when, and in what form, and
  • a set of team-created rules for how you will talk to each other, and what to do if someone breaks a rule.

These two tools, if backed by team and manager self-discipline, will eliminate the majority of your team’s communications issues. Well-supported by research, they also meet the standards of adult learning by providing concrete, relevant information. With MBTI exercises, you end up with abstract type information and general suggestions that users must try to project into the future, instead of clear behavior-based guidelines.

In the movie “Jerry MacGuire,” a client of the title character makes him scream into the phone, “Show me the money!” He meant, of course, show me the financial results for me from your efforts. I would say the same thing to the next MBTI practitioner who claims the test is a good way to spend your money for teambuilding.

Sources:

  • McCaulley, M. (00), “Myers-Briggs Type Indicator: A bridge between counseling and consulting,” Consulting Psychology Journal 52(2):117.
  • Pittenger, D. (05), “Cautionary comments regarding the Myers-Briggs Type Indicator,” Consulting Psychology Journal 57(3):210.

Slipped DiSC: A Popular Profile Fails the Test

If you have done a DiSC® assessment and gotten something out of it, I believe you. But after reading a book from the company that created the commercial version, I don’t believe it happened for the reasons you were told. I also do not believe DiSC is marketed and applied correctly by most sellers, or is a good use of time and money for a group trying to get along better.

I’m Stuck, You’re Stuck, was written by the chief learning officer of Inscape Publishing, Tom Ritchey, and business writer Alan Axelrod. It says DiSC was developed by Dr. John Geier of the Univ. of Minnesota sometime prior to the 1970s. He based it on a model of human behavior “developed by psychologist William Moulton Marston in the 1920s.” (Marston also invented the Wonder Woman comic book character.) Officially the assessment is called the Personal Profile System (PPS). “DiSC” stands for the four quadrants on which people are scored: Dominance, Influencing, Steadiness, and Conscientiousness. The lower-case “i” was a typo they decided to keep for marketing purposes.

We’ll start with the test itself. The number of quadrants is a problem. In 1992, educational researchers conducted a validity check called “factor analysis,” in which the computer looks at how responses clump together. For example, if you like parties and talking to people and goofing around in public (three different responses), a valid test would say you are an extrovert. The 1992 study found only one factor that fully matched a DiSC factor, and another that somewhat did. A third looked like two of the DiSC factors combined, and a fourth did not look like any of them. The book claims Inscape’s factor analysis supports the DiSC quadrants, but they provide no data to support this in the book or on the company Web site. The 1992 study was in a peer-reviewed journal, meaning it was reviewed by anonymous scientists before publication.

On the positive side, a 1982 study tested for something known as the Barnum Effect, under which people presented with their assessment results by an authority figure often call the results accurate even when those results were faked. The abstract says that problem did not show up with DiSC.

As I partially did for the previous topic, the Myers-Briggs test, this time I searched three business or psychology databases with thousands of peer-reviewed studies, using various terms. Only the two discussed above mention DiSC. The reason, simply, is that scientists use other, more valid personality and behavior assessments developed in public (unlike Inscape’s) with scientists checking and rechecking each others’ work. I also looked at the tables of contents and indexes of eight scientific books about behavioral or personality assessment, another for behavior analysts, and one on organizational assessment. None mentioned DiSC, PPS, or related terms.

The much larger problem is how the test is being used in the modern business world. Run a search on the Internet, and many of the hits will call it a “personality” assessment. The book says it is a “behavior” assessment. So, many certified resellers of the test don’t even know what they are purveying. That said, of the 48 items in the book’s version of the assessment, at least 14 (29%) are purely thoughts or feelings, not behaviors (like “happy most of the time”). Many others, like “willing to help,” are debatable. At least one leading scientist defines personality solely in terms of behavior.

Maybe the creator of the test doesn’t know what they are purveying. If so, that’s another problem with the quadrants, since the consensus among psychologists is that personality has five factors (hence the term “Big-5”). This consensus jelled in the 1990s, after DiSC was created.

People regularly tell me their profile: “I’m a high ‘D, low I,’” they’ll say. To quote the book, however, “You’re not a D, I, S, or C. You are responding to a situation in a D, I, S, or C way.” This book from DiSC’s earlier publisher stresses that the assessment is for one particular situation. Facilitating a meeting is different from attending one in which you disagree with the group, according to a case study from the book. Mathematically, to claim you tend to be a “high D,” you would have to take the test dozens of times. DiSC trainers who don’t make that clear don’t understand what they are selling.

This means companies that encourage workers to post their profiles on their cubicles have completely missed the point. To accurately reflect the assessment results, you would have to post multiple results labeled with their specific situations. “If you assume that others will always react the same way, you may become prejudiced about them,” the book warns. “You can start pigeonholing people or limiting their responsibilities or careers.”

You also see DiSC extolled as a way to improve group performance and personal relationships. However, the book states, “The main purpose of the instrument is to assist individuals on their own path of human growth and discovery.” Ritchey and Axelrod point out that researchers “have not tested it on the organizational level.” Despite this admission, they go on to say, “Organizational style may not have a research basis, but when you work someplace, you can recognize it.” This kind of rationalization is how the many myths about teamwork persist. The chapter on organizational impact is pure guesswork presented as fact, and mars an otherwise respectable effort.

Why did DiSC work for you or your team, if you feel it did? “Perhaps the greatest benefit is a common language for talking about feelings, thoughts, and behaviors,” the I’m Stuck book says. In one study described in the next topic, discussions based on a completely made-up (and free) set of types created measurable performance improvement. We know scientifically that teams that have open discussions perform better. DiSC helped you because it gave you a safe way to discuss your differences with your teammates using jargon all would understand. Too bad you were not given a more accurate view of yourselves and each other in the process.

Sources:

  • Aamodt, M., and W. Kimbrough (1982), “Validity Considerations in Acceptance of Personal Profile System Interpretations,” Educational and Psychological Measurement 42(2):625.
  • Henkel, T., and J. Wilmoth (1992), “Factor Analysis of the Personal Profile System,” Journal of Experimental Education 60(3):271.
  • Ritchey, T., and A. Axelrod (2002), “I’m Stuck, You’re Stuck: Break Through to Better Work Relationships and Results by Discovering Your DiSC Behavioral Style,” Berrett-Koehler Publishers, Inc.: San Francisco.
  • Books mentioned in paragraph 5—All that were on the shelves in the relevant section at Davis Library, Univ. of North Carolina at Chapel Hill.

Escape “The Cult of Personality Testing”

When I first questioned the value of personality assessments for teams, I assumed no one outside of academia had done so given the tests’ popularity. But back in 2004, someone more objective than I was (since I was a consultant at the time) and with more science-writing credibility came to the same conclusion. Yale Univ. and Columbia School of Journalism graduate Annie Murphy Paul used to be a senior editor at Psychology Today and has written for major magazines. She put her conclusion in the title of her book: The Cult of Personality Testing.

Some telling themes emerge as she sketches the lives of creators of some influential tests and the tests themselves. I did not realize there are hundreds of these tests. Each of them has adherents that believe their preferred assessment is absolutely correct and powerful—and in a sense I’ll explain, they do all “work” for their supporters. The backers’ arguments are often based on belief in the founder and the founder’s writings. Though some may marshal scientific-sounding arguments in support of their assessment, they ignore the consensus of actual scientists against its fundamental elements. Glaring flaws in how the test was created or is used are denied despite overwhelming evidence. Supporters sometimes claim the popularity of the test is a sign of its value, ignoring the large number of alternatives other people find equally valuable and the alternative explanations for why their choice provides value to themselves. Even people who should know better use the assessment, though the realists among them admit the problems with it and use it guardedly.

Now replace the words “test” or “assessment” with “religion” in the previous paragraph. I follow a spiritual path, so this is not a criticism of religion. But we all know faith can be contorted by fanatics into something harmful, and I think that is the case with this approach to hiring, promotions, and teambuilding.

Paul covers some tests you have may have heard of, such as Myers-Briggs, Rorschach blots, and the “Big-5” tests. All of them “work” in some ways. Most people think the test has captured them correctly. But this is partially “the Barnum effect,” in which, studies have shown, many people believe the results handed us whether they are our results or someone else’s. It also is because we tend to answer the questions the way we want to see ourselves, and/or interpret the findings through our personal filters. Each test is partially correct. There are obvious human tendencies some people have a lot of and others do not, like people who are extroverted or introverted in most situations—unlike most of us who are a mix. The tests can be better than nothing for team- or self-improvement, given that a systematic approach to a problem is always more effective than no system. Every medical breakthrough, business success, and genocide in history was achieved through a systematic approach. That doesn’t make the system moral or cost-effective.

So yes, if a team has muddled along without any communication assistance, any of the tests on the market followed by a discussion might provide benefits. But the discussion, using the common vocabulary from the test, does the work. It could have been generated better ways. Meanwhile, people have been misinformed about themselves and their teammates.

In a study mentioned in the previous topic, college students put through exercises with their class teams in which they discussed their personality differences at the start of a semester earned higher project and individual grades in the class than those who didn’t. They also were more satisfied with their teamwork experiences. The exercises did not use personality tests, and took only 30 minutes. Students in two marketing classes were randomly assigned to teams of four or five students, and the teams were assigned by a lottery to do the exercises or not. The exercise leader showed slides of eight humorous team personality types based on political roles and gave brief explanations of related team member behaviors, such as:

  • The Communist: Gets upset if each member contribution is not exactly equal.
  • The Dictator: “Do you think anyone will notice if I don’t say anything, but just start bossing people around?”
  • The Anarchist: “Let’s just all do our own thing and see what happens.”
  • The Capitalist: Everything is a competition between group members or other groups, “we WILL win”

Students were told to “write down any reflection of their own behavior in teams…” according to the journal article in Journal of Marketing Education. They were also asked to write “how their personality can sometimes inhibit team progress and, if and when that happens, how they would prefer their teammates deal with them.” In team meetings, the students then shared what they wrote, and the other members gave them feedback. Halfway through the project, students rated themselves on how they were doing as team members and wrote what others could do to help them do better. These again were discussed with their teams.

The researchers, Matthew Lancellotti and Thomas Boyd of California State Univ., Fullerton, had been using the exercise for eight years, though this was the first scientific test of whether it worked. They provided some “lessons learned” based on their experience. “When teams meet with the instructor they openly joke about their different styles, perhaps because their baggage is already out in the open,” the researchers write. “Instead of complaining about each other, they speak in terms of how they can accommodate each other’s styles.

“We have also found that trying to match personality types that would seem to appropriately fit together does not work…” they continue. For example, putting a “natural leader” with people who prefer to follow “often results in the leader simply doing most of the work or delegating all of the work.”

Notice that these “personality types” were fictions made up by a pair of marketing professors, yet had profound impacts on measurable team performance. Clearly the types were not the reason. The discussions using a common vocabulary did the job.

In 1968, Paul writes, Stanford Univ. psychologist Walter Mischel wrote a book called Personality and Assessment that exposed the weak underbelly of the tests. She says Mischel “simply points out that personality tests don’t do a very good job of predicting how humans will act.” The link between personality and behavior was around 0.30, a figure repeatedly verified since then using the most scientifically accepted tests. “This meant that less than 10 percent of the variance in a person’s behavior was explained by personality as measured by personality tests,” Paul explains.

That is because, Mischel wrote, the assessments “’have ignored the individual’s actual behavior in real life situations.’”

Paul summarizes, “We adjust our behavior according to our role (worker, parent, friend), to the occasion (a meeting, a family outing, a party), and to a thousand other details of our ever-changing environments.” One reason test supporters ignore this is because it is impossible to test us in each of those moments. And accepting the impact of the moment means you cannot sell a test to a company or create a simple method for saving the world—as some test creators openly wanted, to their credit.

The potential harm is great. Paul provides cases where children have been shunted into substandard education based on these tests. Kids have been removed from their parents. Innocent, rational people have been sent to jail or mental institutions. People have been refused jobs or promotions based on personality assessments, and companies have wasted millions in search of a simple answer to the complex problem of increasing team performance without mistreating workers. Instead, Paul says:

“(A)ssessments of workers and students should be concerned with their specific abilities, not with overarching judgments of their personalities. The most effective evaluations are made by observing the individual in a situation as close as possible to the one in which he or she will be expected to perform. If personality tests must be used, they should be chosen carefully—free of invasive questions, fair to all groups, proven scientifically valid and reliable—and interpreted cautiously, with an acute awareness of their limitations. Their results should be kept strictly confidential.”

This is not one woman’s opinion. They match the guidelines produced separately by a test publisher’s group and the American Psychological Association, guidelines routinely ignored by members of both.

What can you do as a team member? Resist, Paul says. You will have to decide for yourself how far you are willing to take this, meaning, whether you are willing to risk your job. What a shame the cult of personality has been allowed to force any of us to face such a choice.

When scheduled for a test, look it up in the Mental Measurements Yearbook from the Univ. of Nebraska-Lincoln at your local university library or online for a few bucks.[1] This will give you ammunition to argue against taking it, and if that fails, possibly to “game” the test to your advantage. “If you go ahead with the test,” Paul advises, “inquire about how its results will be used, ask for feedback once it is scored, and request an assurance that your answers will be kept confidential.”

Sources:

  • Lancellotti, M., and T. Boyd (2008), “The Effects of Team Personality Awareness Exercises on Team Satisfaction and Performance,” Journal of Marketing Education 30(3):244.
  • Paul, A. (2004), The Cult of Personality Testing. Free Press: New York.

The Mystery of Personality and Performance

The trend toward pre-employment personality testing concerns me for various reasons I won’t rant about now, but one of the biggest is the spurious basis for the judgments drawn from these tests. Personality and performance are both so complex, any claim that you can predict the best personality for a given job is, at best, shaky.

First, you don’t just have to fit the tasks of the job, but also the personalities of the team (current personalities, that is), manager, and anyone else the worker will deal with regularly. As discussed in earlier topics, personality is situational: though I come up in tests as an introvert, correctly, I love making a living immersed in groups of people. Also, four researchers argued in the journal Group Dynamics, “there are specific, lower-level facets within the higher-level… traits that may have differing and even contradictory effects on team performance.” (I bet the same is true for individual performance.) For example, do extroverts help teams because they are assertive or because they are social? Both are subtraits of extroversion. If you were hiring for a team that was too aggressive and thus not getting along well with other teams, an extrovert might or might not be a good addition depending on that person’s specific subtraits. By the same token, there are many different ways to measure performance, the researchers point out. To use the test results accurately, you would have to match the specific subtraits to the specific type of performance you were looking for in a specific team within a specific environment—any piece of which could change the next day.

The article dates from 2006, but I found it while researching my topic on the Myers-Briggs test and think it has a lot for hiring managers and HR directors to consider. I am intrigued by the broad background of the researchers, with representatives from academia (Univ. of Central Florida), consulting, a nonprofit research organization, and the U.S. Army. They reviewed the results of studies linking personality to team performance and predict how specific traits will relate to various aspects of that performance.

By the way, there was a nearly 40-year gap in the research from 1959 to 1997. My point is, the personality/performance link does not have multiple decades of studies supporting it like other aspects of teamwork do.

Here are the parts of performance they found in the literature and how they defined them (all are direct quotes):

  • Adaptability—Team members use information from the task environment to adjust strategies through the use of flexibility, (changing) behavior, and reallocation of resources.
  • Shared situational awareness—Team members develop shared knowledge of the team’s internal and external environment.
  • Performance monitoring and feedback—Team members give, seek, and receive task-clarifying feedback.
  • Team management—Team members direct and coordinate task activities, assign tasks, plan and organize, and motivate other team members.
  • Interpersonal relations—Team members optimize interpersonal interactions by resolving conflicts, use of cooperation, and building morale.
  • Coordination—Team members organize team resources, activities, and responses to ensure complete and timely completion of tasks.
  • Communication—Team members exchange information efficiently.
  • Decision making—Team members integrate or pool information, identify alternatives, select solutions, and evaluate consequences.

Out of 12 personality traits, the authors think only two will be positive for every aspect of performance: emotional stability and flexibility. When you understand that the two subtraits of the first one are “adjustment” and “self-esteem,” this makes sense. As the researchers say, “Given that those low on adjustment are prone to be distressed, upset, hostile, irritable, and nervous, they are not likely to excel in interpersonal or team settings.” As for flexibility, a team that resists changing to new conditions clearly won’t perform as well as one that easily adapts. However, I think you can have too much of a good thing. Some teams are too quick to chase the latest gig or trend. I worked for a company that essentially failed because it leapt to meet every new customer request and spread itself too thin. And in a highly regulated industry like drug development, any flexing needs to be grounded within the restrictions placed on the team. In either case, having at least one “stick-in-the-mud” to serve as devil’s advocate would be helpful.

The authors think someone with a strong desire to achieve will also be good for a team, with the exception that a high achiever may not have the best interpersonal relations. (What if that person wanted to achieve at any cost?) No trait is absolutely negative, they think; dominance comes closest, with a mix of negative or neutral marks. A team with lots of dominant people would probably have difficulty making decisions, but those same people might monitor the team’s performance carefully, which could be a good thing.

The rest of the traits, the authors think to be a mixed bag. You might think people with high levels of affiliation (desire for personal connection) would be good for a team, and it probably would improve communication and interpersonal relations. But that trait could hurt the ability to lead, coordinate, or make decisions with team members, because those people can be more interested in socializing than getting work done.

All of this is only educated guessing, mind you. The critical point? These are predictions, results these very educated guessers think will prove true if studies are done on them. They are saying nobody knows what personality traits will help specific parts of team performance, which means nobody knows better than a team’s leaders and members what traits will help that team. Even then, being human, they may go for the traits they prefer rather than the ones they really need to perform better.

Source: Driskell, J., et al. (2006), “What Makes a Good Team Player? Personality and Team Effectiveness,” Group Dynamics 10(4):249.

In Hiring, Too, Behavior Matters More than Personality

Given the trend toward testing of job applicants mentioned earlier, you would think there is a ton of hard evidence that you can predict job performance based on personality tests.

Think again. No hard evidence. Little evidence at all. A team of later researchers quoted a 2007 study in writing that scientists, “disappointed at the seemingly low validity of personality in predicting job performance, have suggested that ‛self-report personality tests should probably not be used for personnel selection.’”

Do I think this inconvenient truth is going to stop the headlong rush into hiring new team members based on invalid reasons? Of course not. Maybe, though, I can get people who use such tests to think more broadly about the results. From their recent journal article, I think that was part of the reason Huy Le of Trident Univ. International, with a team from three other universities, ACT Inc., and the Iowa Dept. of Administrative Services (DAS), conducted two studies into a possible explanation for why the assessments don’t work. Study 1 was in the DAS, while Study 2 used data from 25 organizations in different industries from across the U.S. They focused on two of the five main factors of personality per the Big-5 test, “conscientiousness” and “emotional stability.” (The others are “openness,” “agreeableness,” and “neuroticism,” with OCEAN thus the mnemonic.) The researchers compared personality test results from individual workers with their leader-reported job performance ratings. Note that the leaders did not know the workers’ personality results.

“Conscientiousness refers to the extent to which persons are dependable, persistent, organized, and goal directed…” the study article explains. “Emotional stability indicates the extent to which people are calm, steady under pressure, and less likely to experience negative emotional states, including anxiety, depression, and anger…” It seems like both would be positive for job performance, and in both studies, they were on average. But the correlations were fairly small, around +0.20 (on a scale of -1.0 to +1.0, with 0 meaning no relationship).

A possible explanation is that you can have “Too Much of a Good Thing,” to quote the article title. The researchers’ theory is that performance rises with “good” trait ratings only up to a point. Then performance begins to drop again. In the team’s first study, higher levels of conscientiousness or emotional stability did not raise job performance in a straight line, as assumed by people who use personality tests in hiring, the team suggests. Instead the lines curve like a hill. Somewhere above the average score for all workers in the study, higher ratings were linked to lower job performance. On a scale of -3 to +3, the top of the hill was between 0 and +1.

Someone who is too conscientious “can be considered rigid, inflexible, and compulsive perfectionists,” the article says. “Such persons may pay too much attention to small details and overlook more important goals required on the job…” In the same way, there was a point past which higher emotional stability provided no added value. The authors say another researcher has pointed out that higher neuroticism (the opposite of emotional stability) can help job performance, because neurotics anticipate future problems.

There was a another complication in the data, however, as the team had theorized. The curve differed based on how complex the job was. For low-complexity jobs like truck driving, the curve was much sharper than for high-complexity jobs like accounting. For emotional stability, low ratings lowered performance on simpler jobs more than on complex jobs. Also, in simpler jobs high ratings lowered performance as well. In complex jobs, on the high end performance was basically the same whether the person was somewhat above average or far above average.

For conscientiousness, the curves for simple and complex jobs were similar through the mid-point (0 on the scale). But being too conscientious appeared to hurt performance on simpler jobs after the mid-point, and on complex jobs starting around a 2.

Adding further to the murkiness, the evidence in the second study was not strong enough for the team to say it supported their “curve” theory. The team writes this may be due to Study 2 using a sample from various organizations, such that differences between the organizations masked the curve effect. On the other hand, the Study 1 sample might be unusual. As usual, more research is needed.

More useful to team leaders, perhaps, is what the studies discovered about behaviors. The researchers also measured:

  • Organizational citizenship behaviors (OCB), helping actions that are “not directly task related but contribute to the goals of the organization,” and
  • Counterproductive work behaviors (CWB), actions “that potentially harm the well-being of the organization…”

OCB examples from the leader surveys included:

  • “Assists other employees with their work when they have been absent.”
  • “Takes times to listen to coworkers’ problems and worries.”

Again, in Study 1 the links between each of these sets and each of the personality traits were curved. Higher conscientiousness lowered helping behaviors after a point, for example. The behaviors had a higher impact, at least on the high ends, for simple jobs. However, Study 2 only found curves and complexity effects for emotional stability (not conscientiousness) with OCB, and found a curve but no complexity effects for that trait with CWB.

This brings me to the most telling comparison: personality traits versus OCB and CWB as predictors of job performance. As noted above, the relationship between personality and performance was only around 0.2 on a -1.0 to +1.0 scale. The link between OCB and performance, however, was +0.8 and that for CWB and performance was -0.63. (The minus sign means higher CWBs were related to lower job ratings, as you would expect.) In short, behavior had three to four times more impact on job ratings than personality.

Taken together, these results mildly support but by no means prove the curve theory. The team admits the findings do not add as much to the debate over personality testing in hiring as they had hoped. This is especially true since, as they point out, job applicants are more likely to skew their reported traits upward than comfortably employed people taking the tests anonymously, which the study subjects were.

In short, the evidence is that behavior is what matters, not personality. Personality is about your preference in the ideal situation; behavior is about your chosen action given a specific situation, environment, and set of people. Which kind of situation happens in your organization? If you give too much weight to a test result showing someone is very emotionally stable, you could very well cut out your best candidate.

Source: Le, H., et al. (2011), “Too Much of a Good Thing: Curvilinear Relationships Between Personality Traits and Job Performance,” Journal of Applied Psychology 96(1):113.

Introverts are Normal, Too

There’s that person at your work who doesn’t say much. Hold a brainstorming session, and their mouth stays calm. Ask for their ideas, and they seem to have none, until you receive an e-mail two days later. Invite them to the bar with the rest of the gang, and you’ll have one less drink to buy. Invite them for a one-on-one, though, and they might accept.

This most likely is an introvert, and if you aren’t one, you may well wonder what is wrong with them. “‘Don’t you like people?’ is a common remark introverts hear, says Marti Laney, a psychologist and the author of The Introvert Advantage.” I quote an article from Psychology Today by psychologist Laurie Helgoe, Ph.D., based on studies about introverts. “‘Usually we like people fine,'” Laney insists. “‘We just like them in small doses.'”

Helgoe makes some points managers and teammates of introverts who aren’t introverted themselves should keep in mind. For example, “shy” and “introverted” are not synonyms. Helgoe explains, “An introvert and a shy person might be standing against the wall at a party, but the introvert prefers to be there, while the shy individual feels she has no choice.”

The world does not divide cleanly into introverts and extraverts. “There are a few extremely extraverted folk, and a few extreme introverts, while most of us share some extravert and some introvert traits,” Helgoe writes. There’s nothing wrong with introverts, and extraverts are not “most people.” If you are extraverted and try to change an introvert to act more like you, you are practicing a form of discrimination.

The differences are visible in the brain, Helgoe reports. “In a classic series of studies, researchers mapped brain electrical activity in introverts and extraverts. The introverts all had higher levels of electrical activity… whether in a resting state or engaged in challenging cognitive tasks.” That means introverts need to limit outside stimulation to maintain an optimum level of activity, while extraverts need to increase it. In the same way, introverts do not require external rewards to the same degree as extraverts. Studies have also shown that introverts have greater blood flow “in the frontal cortex, responsible for remembering, planning, decision making, and problem solving—the kinds of activities that require inward focus and attention,” and in “a region associated with speech production—likely reflecting the capacity for self-talk,” Helgoe says.

The question of whether the introvert is contributing to your team must rest on the amount and quality of contribution, not the quickness. Helgoe writes, “Introverts like to think before responding—many prefer to think out what they want to say in advance—and seek facts before expressing opinions.” Whether or not their feedback is timely is usually the requestor’s fault, not the introvert’s. Some decisions have to be made instantly, but the vast majority of important business decisions can and should allow time for reflection by everyone on the team.

Helgoe makes one mistake in claiming 50 percent of people in the U.S. are introverts. She bases this on results from the Myers-Briggs Type Indicator, apparently unaware that Myers-Briggs forces people into one category or another. That is, if you are 51 percent introverted on MBTI, it classifies you as an introvert. So with all of us evenly distributed on the continuum from introverted to extraverted, you would expect MBTI to split us 50-50. My guess is that most research psychologists would only call someone an introvert or extravert if they fell into the extreme one-quarter or third on either end. But the point remains that there are just as many of one as the other. The extraverts simply get more attention, because they choose to.

When dealing with someone more introverted than you, I suggest you:

  • Ask one question at a time, per Psychology Today‘s Matthew Hutson at the end of Helgoe’s article: “‘Extraverts think we have answers but just aren’t giving them,’ Laney says. ‘They don’t understand we need time to formulate them’ and often won’t talk until a thought is suitably polished.'”
  • “Don’t interrupt if an introvert does get to talking…” Helgoe says. “Introverts are unlikely to repeat themselves.” In other words, learn and practice active listening skills.
  • Plan meetings to allow time for introverts. Add possible problems as agenda items as soon as they come up; send agendas out at least 24 hours in advance along with background materials; and set action items for people to provide post-meeting feedback before making decisions final.
  • Don’t act like there is something wrong with acting introverted. Hutson says, “Many introverts are happy with the way they are. And if you’re not, that’s your problem.”

The burden of maintaining good working relations does not fall completely on extraverts, of course. If you are more introverted than most:

  • Say so. If a decision is coming too fast for your comfort level, make yourself ask for time to think about it.
  • Push yourself when required. Some decisions do have to be made quickly. Before trying to slow one down, ask why the deadline was chosen and do the best you can if the timing makes sense.
  • Monitor your self-criticism. Helgoe reports that “introverts are more self-critical than others—but also more realistic in their self-assessments.” That’s okay, but bad-mouthing yourself around extraverts could lower their opinions of you unfairly.
  • Control your criticism. It’s as easy to think an extravert “talks too much” or “never thinks anything through” as it is for extraverts to make mean assumptions about you. “Scientists now know that… introverts have no special advantage in intelligence,” Helgoe points out.

I am an introvert who loves sitting in a library going through studies. I am extraverted enough that I enjoy giving energetic, joke-cracking team speeches and facilitating teams. Both sides of my personality add value in the workplace, I believe. Both personality types add value in yours.

Source: Helgoe, L. (2010), “Revenge of the Introvert,” Psychology Today, 09/01/10, http://www.psychologytoday.com/node/46944.

Reject the Cult of Personality to Change Your Team

The cult of personality in the business world is a useless distraction that becomes an excuse for managers to blame employees’ bad behaviors—or their own—on factors supposedly beyond our control. Phrases like, “That’s just the way he is,” or, “I can’t help it,” rationalize away responsibility. This cult is built on a number of rampant myths about personality:

  • After adolescence, personality does not change (False).
  • Personality is the main driver of behavior (False).
  • We have no control over our personalities (False).
  • A ten-minute test can reveal useful information about your entire personality (Fraud).

If you won’t believe me or psychology researchers, how about economists? Mathilde Almlund, James Heckman, and Tim Kautz of the Univ. of Chicago, and research psychologist Angela Lee Duckworth of the Univ. of Pennsylvania, wrote a textbook chapter titled, “Personality Psychology and Economics.” In correcting the view of many economists that personality has no role in economic outcomes, their review of the psychology literature also blows away those claims above.

Quoting another researcher, they define personality traits as “’the relatively enduring patterns of thoughts, feelings, and behaviors that reflect the tendency to respond in certain ways under certain circumstances.’” The authors decide that personality is a “system of relationships” between various factors and resulting actions, factors including: genes, motives, values, abilities (verbal, spatial, etc.), stories we tell ourselves, scripts we tend to follow, self-identity, and culture. “Behavior depends on incentives created by situations,” they add.

Correlations between personality and behavior rarely surpass 0.3, though the economists mention the same is true for IQ and group psychology. One lab study found a higher correlation of 0.43 over 15 different tasks, but the point for me is that personality does not control behavior. Almlund’s team strikes a balance when it writes, “A rich body of correlational evidence… shows that for many outcomes, measured personality traits are predictive… and situations also matter.”

They explain, “Situations could include physical aspects of the environment in which the agent is located or the network (and other social situations) in which the agent is embodied. The situation can include social factors such as peer effects.” Positive peer pressure is a key element in Full Scale agile™. The economists provide fascinating evidence that work interventions can change personality. A project offering subsidies to welfare recipients to take jobs not only increased their incomes, it changed participants’ view of how much control they had over their lives—the personality trait called “locus of control.”

Far from being genetically set, traits and abilities can be impacted by self-identity, the chapter says. A mechanism called “epigenetics” appears to shape how genes are expressed, the most obvious example being the link between behaviors and diseases. Even if you have the “shy” gene, it is possible to change how shyly you behave. In turn this might cause the shy gene to stop pressing you so hard.

In fact, the chapter provides impressive evidence that personality can change throughout our life. Lifetime stability is 0.74, where 1.0 would mean zero personality change. IQ becomes relatively stable by age 8 on average, but personality does not reach that maximum stability until age 50 or later, as measured by tests retaken after seven years. I found it interesting that after big changes in early childhood, personality slowly hardens into our 30s, then changes more in our 40s. Mid-life crisis, anyone?

Having generated international debate by my comments in Teams Blog on workplace personality tests (some of the topics above), I’ll point out that the team provides a detailed description and critique of various tests, focused on the Five-Factor or “Big-5” Model which has won out over competitors in the scientific community. Most of the economists’ criticisms of the model are issues about the factors, not whether the model itself is the best. Even in the section on “Alternatives to the Big Five,” the popular Myers-Briggs and DiSC tests—both claiming only four factors—aren’t scientifically accurate enough to warrant a mention anywhere in the chapter.

The key question addressed by the economists is, do personality traits predict economic outcomes in life? The team says “yes,” with studies showing the Big-5 trait of “Conscientiousness” is the best predictor. The American Psychological Association defines this as: “The tendency to be organized, responsible, and hardworking.” The chapter says conscientiousness ratings predict college grades as well as the SAT does. Conscientious people do better in school, live longer, and tend to get higher job performance ratings and wages. (However, as reported above, it is possible to be too conscientious on the job, for example when trying to do everything perfectly means you don’t get all your work done.) Mounting evidence suggests people with General Educational Development diplomas or GEDs, which prove equal knowledge to that of high school graduates, do not do as well as those graduates in life, probably because the skills required to persist through the challenges of high-school life are the same required for career success.

Personality leads to preferences, not behaviors. A team leader must reject the cult of personality and focus on how you communicate expectations, identify and reward positive behaviors, confront negative behaviors, and leverage group psychology to the advantage of all. To change an employee’s behavior, provide a measurable goal, training, opportunities to use that training, ongoing reminders (coaching), and rewards for using the learned skills. The chapter says the mechanism for making that change is called “cognitive control” or “executive function,” both phrases used to refer to “the voluntary, effortful blocking of a habitual behavior in order to execute a less familiar behavior.” Use your executive function to change your own habits, and you will make your life as a manager easier regardless of your personality.

Source: Almlund, M., A. Duckworth, J. Heckman, and T. Kautz (2011), “Personality Psychology and Economics,” in Hanushek, E., S. Machin, and L. Woessmann (eds.), Handbook of the Economics of Education, Vol. 4. Elsevier: Waltham, Mass.

Section Menu | ← Back | → Next


[1] http://marketplace.unl.edu/buros/

Tell the world: