In response to a comment left in a previous post on names and outcome, I actually tracked down the paper, Moniker Maladies, and read it. As often happens, the real story is much more interesting.
Actually, the paper itself is a nice demonstration of the scientific method. It takes some interesting observations made by others, forms a hypothesis and then tests that hypothesis. The testing leads to new questions that they then examine.
An important aspect of research is to examine things we know to determine if they are in fact true. That is part of how we fight confirmation bias. Simply because the work demonstrates that they are in fact true does not make the research irrelevant or useless.
Same with this paper, which does a nice job demonstrating that people whose initials match a designation of failure, fail more often than those initials lack any such congruence. And that this failure is due to something unconscious since the failure happens even when they like their initials.
This report actually does a fairly comprehensive examination of something that has been know about since the lat 80s– that people like things that remind them of their name, especially if the first letters match. This is am empirical observation that was called the name-letter effect – NLE.
Now, it was not possible to tell if the positive NLE was due to unconscious connections or “do a few people named Jack deliberately move to Jacksonville for its Jack-resembling appeal”? To answer this, the authors of the paper asked an interesting question: If positive NLE exists, does a negative NLE where initials correlate with failure? As no one would purposefully choose to fail, this might clear up the question of unconscious drive or not.
So, this paper examines the negative aspects. Was there still a correlation even for things which people would not consciously choose? A batter does not want to strike out. Turns out there is, which seems to shore up the hypothesis that NLE comes from unconscious desires rather than conscious.
By applying some straightforward logic to a puzzle, they have helped unravel the correlation and identified something that we did not know or understand before. While the effects may not be huge, they do appear to be real.
This study does not really prove why this happens. I mean, a priori, why would having the first name Karl mean you strike out more? Is it simply an unconscious fear of failure that produces the results or something else?
They looked at the statistics for almost 6400 players from 1913 to 2006. Those guys whose names started with K struck out about 1.5 more times every 100 at bats than anyone else. This may not sound like a lot but it is statistically quite robust. an attempts to remove possible confounding effects, such as age of the batter, etc. did not remove the correlation.
They looked at grades, 15,000 of them. Names from E to Z were scored as ‘Other’, while they examined people with A, B, C or D grades. What they found people whose initials were ‘As’ or ‘Bs’ had similar GPAs as ‘Other’ but people with initials that were ‘C’ and ‘D’ had significantly lower grades.
But, one obvious problem here is that it is not possible to separate out some confounding factors. Perhaps umpires do not like people with an initial K and call them out more. Or maybe teachers give people with C or D initials lower grades. That is, their performance is also determined by others.
So their next study was ‘Do students achieve initial-congruent grades, or do teachers assign initial-congruent grades?’ They essentially asked a large group of students to rank each letter of the alphabet – thus giving insight into how they ranked their initials with all the other letters. They then examined the correlation of GPA with initials.
What they found was that those who had A or B as an initial, and liked those initials, had high GPAs. If they did not like their A or B initials, they had lower GPAs. This indicates that there can be a conscious negative effect. The grades were determined by how the student felt.
Interestingly, for C or D individuals, those with the lowest GPA were those who actually liked their initials suggesting an unconscious effect.
They also looked at law school admissions to get an idea of real world consequences and found something similar. As they state:
The first four studies suggest that people whose initials match objectively undesirable performance outcomes perform worse than people with other initials. Moreover, these studies dem- onstrate this effect in real-world situations that have important consequences: Strikeouts, grades, and graduate schools can affect salaries, status, and careers.
All their work so far was mainly archival instead of creating their own data. So, the fifth study was one they did themselves. They gave a group of people a task to complete. If they succeeded at the task, they could click a named button and get a chance at a prize of more than $100 – big prize button. If they failed, they could click another named button and get a chance at a prize of less that $100 – small prize button.
The two buttons could have irrelevant names (i.e. X and Y), positive-name congruence – the big prize button could match the initial of the subject (i.e. T for Toby), or negative-name congruence – the small prize button could match the initial of the subject (i.e. T for Toby),
The trick was that the task was impossible to perform, so nobody should pick the big prize button.
When the buttons were named with irrelevant initials (i.e. X and Y), the participants finished about 5.3 of the 10 tasks. When there was positive-name congruence, they finished the same number . But when there was negative-name congruence, the number of completed tasks dropped significantly.
As the authors state:
Thus, participants performed especially poorly on the anagram task when the prize for failure matched participants’ first initial.
So across 5 quick studies, these authors were able to demonstrate several instances where there were negative consequences passed upon initials. The effect is not large but it is statistically significant.
Why the self-fulfilling prophecy? This paper shows it to be unconscious but an unconscious need to fail? Their conclusion starts:
The five studies reported here suggest that name liking guides the pursuit of initial-resembling performance outcomes, even when those outcomes are explicitly negative.
This is an interesting phenomenon. Even when they consciously like their initials the outcome was similar.