What the Dog Saw: And Other Adventures

In many fields, it is difficult to distinguish which promising candidates will actually do the job well. This was the case for Dan Shonka, a football scout who evaluated 800 to 1200 candidates per year to choose which college players would make National Football League (NFL) teams. Even promising quarterback Chase Daniel, who completed an impressive 78% of his passes, was scrutinized against all the eventualities that may occur in a game. Shonka recalled that when he worked for the Philadelphia Eagles in 1999, only one of five quarterbacks who appeared as talented as Daniel lived up to their promise. This is because the professional game is faster and more complex than at college level, and the athletes may feel they are playing a completely new game.

The “quarterback problem,” where little evidence exists to predict a candidate’s performance, also afflicts the teaching profession (317). Studies have shown that quality of teachers more than quality of school has the biggest impact on how a student will learn and perform.

Bob Pianta, at the University of Virginia’s School of Education, attempted to predict early-career teacher performance by monitoring student-teacher interaction and student engagement. The better teachers responded to the cues the students were giving them, rather than ignoring them or growing snappy. It turned out that the traditional emphasis on a teacher’s academic performance prior to teaching had little correlation to their performance in the classroom. What was needed was the elusive quality of “withitness” (330), and that was difficult to predict before entering the classroom.

The field of financial advice has been the most successful at tackling the quarterback problem. Recruiters know that relying on prerequisites is futile; instead, a stringent process of continual assessment is needed. Ed Deutschlander, co-president of North Star Research Group, oversaw a process in which candidates who passed the interview stage attended a training camp that simulated the real-life job of being a financial advisor. If they met the metric of obtaining 10 official clients and 10 meetings a week, Deutschlander knew they had the requisite speed for the profession. They would then work as an apprentice advisor for three to four years—the time it would take to determine whether a person could make it in the job. Gladwell feels that teachers too should be judged by their performance rather than their prerequisites and that selectiveness should come after classroom experience. If such a system were to be implemented, it would require more flexibility in teachers’ pay scales and the willingness of taxpayers to fund trying four teachers to find a good one. Gladwell thinks that it is unflattering for US society to devote more care to those who handle wealth than to those who handle children.

Authorities and detectives have always looked for tricks to make criminal profiling easier. Gladwell turns to the example of the “Mad Bomber,” who deposited bombs around New York City in the 1940s and 1950s. The New York City Police Department enlisted the help of James Brussel, a Freudian psychiatrist, to identify the bomber. Brussel used clues such as the fact that the “W”s in the “Mad Bomber’s” handwriting looked like a pair of breasts to intuit that the man had never developed beyond the Oedipal stage, was unmarried, and lived with a maternal figure. Brussel also determined that his use of a bomb-and-knife combination was particularly Eastern European. When Brussel seemingly randomly made the prediction that the “Mad Bomber” would be wearing a buttoned double-breasted suit, the police department was skeptical. However, within a month, an Eastern European in such attire, who had changed his name from George Milauskas to George Metesky and had been in the employment of Con Edison, one of his targets, was arrested.

FBI agent John Douglas thought that Brussel had invented an essential new science in criminal profiling and came up with a similarly detailed profile for the Bind, Torture, Kill serial killer, or BTK, in Wichita in the 1970s. Ironically, his profile details were similar to those of Beverly Newton, a psychic who had been asked about the case. Douglas was flattered by the comparison, stating that he tried to enter the mindset and emotional state of the offender. Douglas further developed a profile of two different types of serial killer: the organized, who is a charismatic, socially able individual who has likely been following their victim, and the disorganized, who is a social outcast and whose crimes tend to be spontaneous and take place in high-risk environments, like a busy place at midday.

However, there are also numerous occasions where FBI agents have created the wrong profile for a killer, and this has led them astray. As Gladwell clarifies, “a profile isn’t a test, where you pass if you get most of the answers right. It’s a portrait, and all the details have to cohere in some way if the image is to be helpful” (348). A group of psychologists at the University of Liverpool found the FBI’s assumptions about organized and disorganized serial killers were unreliable, as it seems that crimes are nearly always a combination of organized and disorganized traits. Also, two offenders might display the same behavior for markedly different reasons, making the creation of a personality profile more difficult.

Gladwell reveals that Brussel made just as many inaccurate as accurate predictions about the “Mad Bomber.” The person who found Metesky was not Brussel but a woman named Alice Kelly who noticed that after a dispute between Metesky and his employers following an accident, Metesky had threatened to take justice into his own hands. This phrase had appeared in the “Mad Bomber’s” letters, and so Kelly was able to make the identifications. As for Brussel, Gladwell considers that he understood that offering a plethora of predictions would ensure that some of them would be correct.

Gladwell examines big business’s obsession with talent, despite abundant evidence that high intelligence does not translate to great performance. This was the case in the 1990s with McKinsey & Company launch of a War for Talent. The consulting firm was obsessed with talent, recruiting nonstop and rewarded star performers disproportionately. This approach has been copied, as big companies stalk elite business schools for their talent and nurture a hierarchical system where the best rewards are at the top.

The problems of this approach were evident in McKinsey’s partnership with Enron, a talent-obsessed firm that went bankrupt and saw its top executive, Jeffrey Skilling, go to jail. The highly competitive structure was known as the “rank and yank” policy (360), where the highest-ranked got bonuses that were 66% higher than those of the 30% below them while those at the bottom risked being fired. However, the process of ranking was cavalier and little discussed. Gladwell evaluates their assessment tactics: Ranking according to IQ runs the risk of missing out on how good someone is at interacting with others or performing common-sense tasks, both of which make for good managers. Another option is to test tacit knowledge, which entails being able to make decisions and manage others. Enron wanted its managers to produce detailed reports on employees, but this proved difficult when the people moved from post to post too quickly to be assessed. This was because Enron was a company where stars got to dictate the terms of their jobs. As a result, performance evaluations were not based on performance.

Enron ultimately failed because a good proportion of its managers were narcissists who had immense presence and energy but overcredited themselves for their successes and avoided responsibility for failure. The absence of disciplined management and overemphasis on talent led to a lack of systemic integration. Psychologist Carol Dweck’s investigation into the difference between those who think talent is innate and fixed and those who believe it to be acquired and mutable shows that those who are burdened with the label of being gifted flounder when they make mistakes and would rather lie about them than correct them. While the talent myth assumes “that people make organizations smart […] more often than not, it’s the other way round” (371). Instead of being on the hunt for stars who think outside of the box, Enron should have considered improving its box. In this vein, Gladwell notes the high performance of Southwest Airlines compared to its competitors, given its prioritization of efficient organization over star treatment.

Gladwell considers how little interviews tell us about a candidate’s future performance in a role by considering the case study of Harvard student Nolan Myers, a calmly charismatic programming enthusiast who averaged B and B+ grades. Gladwell, Hadi Partovi from Silicon Valley startup Tellme, and Microsoft CEO Steve Ballmer, all of whom interviewed the candidate only briefly, were convinced that he would thrive in their company. All three had a strong conviction of Myers’s likeability, competence, and high self-esteem. Gladwell, however, maintains that the brevity of their interaction meant that none of them could truly know Myers at all. While the job interview is a stalwart for recruiting people, the lifeblood of any company, it is difficult to discern a true profile of someone after such a short period of time.

These interviews often rely more on nonverbal cues than on acceptable answers. Experimental psychologist Nalini Ambady investigated the nonverbal components involved in teaching. She used muted videotapes of teaching fellows for her research and asked outsiders to rate the effectiveness of the teachers according to the body cues they could see. Even when they were shown as little as two seconds of footage, the observers were able to rate the teachers in 15 categories relating to personality. Additionally, the reactions of observers were shown to be very similar to those of students when presented with the same material. Working in tandem with a University of Toledo psychology experiment, grad student Tricia Prickett extended Ambady’s research by showing how the 15 seconds during which an interviewee walks through the door and shakes an interviewer’s hand influence the outcome of the interview. This explains why Gladwell, Partovi, and Ballmer were all able to make a snap decision about Myers’s capacity to do a job without knowing him.

First impressions have a cumulative effect, as they color how we interpret data that comes in later, owing to confirmation bias. For example, the same statement may appear self-assured or arrogant depending on how much interviewers like the candidate in the first place. While our intuition assumes that people exhibit the same behaviors and characteristics in all situations, this is not the case. Thus, a person who is shy in some circumstances may be extroverted in others.

Human-resources consultant Justin Menkes notes that the standard interview questions are structured so that the right or most flattering answer is obvious. However, we can gauge a truer impression of a candidate’s characteristics when the questions are adjusted to make conflict inevitable. For example, if you were to ask how a candidate would behave if a boss publicly humiliated them, their hypothetical actions would need to come down on the side of either stoicism or confrontation, which would reveal more about their character. Such structured interviewing attempts to minimize the role of intuition and is thus more predictive of performance. However, Gladwell acknowledges that many interviewers reject structured methods in favor of testing their chemistry with prospective employees.

Nolan Myers eventually went to work for Tellme, noting that the start-up environment best fit his personality. Gladwell judges that this idea of a “good fit” is a modern phenomenon in stark contrast with, for example, the IBM of the 1960s, where Myers would have been merely required to fit in and be an efficient employee. Gladwell worries, however, that new companies will be too focused on personality when it comes to the hiring process and that “all we will have done is replace the old-boy network, where you hired your nephew, with the new-boy network, where you hire whoever impressed you most when you shook his hand” (392). Social progress will require us replacing such arbitrary factors with less arbitrary ones.

In Gladwell’s view, pit bulls have much to teach us about crime. This breed has been banned in Ontario, Canada, after violent attacks on civilians. It is true that pit bulls, which are descended from the bulldog, have been bred to have low inhibitions when it comes to aggression. While most dogs fight as a last resort, a pit bull is far more easily triggered. Statistically, however, most pit bulls do not bite, while Dobermans and rottweilers are frequent biters. The breeds listed as the most dangerous have actually changed, over the years, according to the breeds most commonly picked by aggressive and neglectful dog owners. While it would be better to conduct further studies into which dogs do the most damage, this is often impractical in day-to-day life, and we end up stereotyping or generalizing.

Investigating the inefficacy of stereotyping, Gladwell looks at how, after the London bombings in July 2005, the New York City Police Department, headed by Raymond Kelly, decided to perform a series of random searches to hunt for potential terrorists. At the time the terrorists they were looking for were young Arab or Pakistani men, matching the demographic employed by Al Qaeda, but Kelly emphasized that racial profiling was ineffective, citing how the 9/11 hijackers did their best to not look like the media-image Islamic terrorists, deliberately blending in by shaving and even attending topless bars. In a former job as the head of the US Customs Service, Kelly narrowed the profile of drug smugglers to specific behaviors, such as nervousness or the raised alarm of a drug-sniffing dog. These more specific criteria meant that while the number of searches enforced by the customs office went down by 75%, the number of successful arrests rose by 25%.

Gladwell makes the point that, just as it is difficult to categorize which “Middle Eastern-looking” person is an Islamic terrorist, it is difficult to gauge which pit bull is aggressive. The Ontario ban tries to simplify this by accounting for any dog that has an appearance and physical characteristics that are “substantially similar” to a pit bull (399). This necessarily relies upon a generalization that includes any dog that humans might perceive as dangerous. However, pit bulls were originally bred to fight other dogs rather than humans. The American Temperament Test Society has tested 25,000 dogs for their reactions to stimuli as a measure of their aggressiveness. They have found that of the 1,000 pit-bull-style dogs they have tested, only one was disqualifiable for not having a safe temperament. Thus, aggressiveness is not native to the breed, but has been bred in by the trainer or exacerbated by the owner, who may be abusive, be neglectful, or provide a chaotic environment. Often fatal dog bites are a “perfect storm of bad human-canine interactions—the wrong dog, the wrong background, the wrong history in the hands of the wrong person in the wrong environmental situation” (408).