Talk:Lake Wobegon effect

not understanding average

In a similar way, a large majority of people claim – impossibly – to be above average;

Actually, almost everyone has an above average number of legs or fingers. Perhaps what is intended is median or mode? It seems quite reasonable for most people to believe that they are above average, all it takes is some really bad samples; and having seen a few car drivers doing the most remarkably dumb things this may in fact be the case. ;) njh 07:08, 19 January 2006 (UTC)

They could also be rating themselves based on different criteria. "Driving skill" could be viewed as containing a large number of factors that different people view differently. For example, some people might care more about how quickly they reach their destinations, or how effectively they react to dangerous situations, or how well they avoid situations that could become dangerous. Based on their own idea of how a good driver drives, the majority of people can claim to be better than the median and still be right. - CronoDAS 17:18, 12 July 2006 (UTC)

My challenge to this argument is that the fact that one weighs one factor more than another factor (and, not coincidentally, give greater weight to the things that they are good at) is completely consistent with the Effect. Of course, this challenge falls apart if people weigh things independently of their own performance, but this seems unlikely to me.

- unsigned comment

In the psychological tests that the article talks about, it is the median, not the mean that is meant by average. Here for example is an extract from the instructions to subjects in the Swenson 1981 driving skills study:

"We want you to compare your own skill to the skills of the other people in this experiment. By definition, there is a least safe and a most safe driver in this room. We want you to indicate your own estimated position in this experimental group (and not, e.g., Eugene, Oregon or in the U.S.)"

MartinPoulter (talk) 17:06, 11 December 2007 (UTC)

Good points there. I noticed a sentence that says that almost everybody thinks that they have some attribute above average (mean or median). :-) Well, I'd certainly hope so, coz it would be awful to suck at everything! ;-) It's in fact easily feasible that there is something where practically everybody really is better than most others. Be it knitting, tying neat knots or sleeping - but I'm pretty sure there is something above average about everyone... --84.250.188.136 (talk) 01:06, 18 June 2008 (UTC)

how commonly is it used?

I'm just wondering, I can't imagine that it's a widely used term. But then again, english is not my first language. I'm just curious.

• It has some currency, though it's not too common and it's usually explained. Searching the archives, the New York Times has used it in four articles in the last 12 months. Publications whose readers are less likely to be familiar with American public radio would probably use it less.--Pharos 18:36, 9 June 2006 (UTC)

A year later, I agree with Pharos. One does read confident reference to "Lake Wobegon" constantly, and it refers to this, um... "effect" is as good a noun as any to couple with it. I tell my students that, by contrast to cases of low-esteem (the unsupported belief in one's inferiority) they're more often suffering from a case of "high self-esteem." Profhum 16:48, 15 July 2007 (UTC)Profhum

I disagree. Lake Wobegon effect, egocentric bias and actor-observer bias are closely related. Distinguishing (or merging) them may require expert advise. Peace01234 (talk) 03:57, 25 November 2007 (UTC)

I don't understand

From the looks of it, there is absolutely no such thing as the Lake Wobegon effect. Someone explain this to me. 60% of all HIGH SCHOOLERS, tested by the COLLEGE board(IE, people who are interested in college, or people who are taking AP exams) say that they are in the top 25% of EVERYONE? What is impossible about that? Chances are, the people who are taking AP classes in highschool are significantly smarter than those who are not. Therefore, 60% of people being tested by the college board(in this case, less than one million) could quite easily rate themselves higher than the other(in america) 300,000,000 people who were not tested. For example. 100% of all people in my AP Calculus class said they were above average in math. All 30 of those people were infact above average in math. Liquidtenmillion 00:25, 3 September 2006 (UTC)

The idea of the Effect is that people are comparing themselves to people in their own cohort - i.e. that this group of high schoolers would be asked to compare themselves against other high schoolers. If it were specific to college-bound students, then I expect that they would similarly be asked to compare themselves against other college-bound students.

Actual values? or an estimation?

"In 1987, John Cannell completed a study that reported the statistically impossible finding that all states claimed average student test scores above the national norm."

The wording seems to implicate that the states were giving actual values. Perhaps ambiguity could be reduced by verifying this, or saying that the states *estimated* above average scores. And of course, if some states estimated, and some gave actual numbers, then not all of the states *claimed* above average test scores, they estimated. I was about to use this example in an assignment, but then I realized how ridiculous it is.

You imply that the article implies actual values. It doesn’t. A claim is a claim. For example, I am fabulously good-looking. Is that not a claim I just made? Did I offer values and hard facts? — Chameleon 04:30, 26 August 2009 (UTC)

average of a binary series lies between the two values

I was thinking of an example in which a test was graded from 1 to 10. Multiple students only received a score of 4, while 1 student only received a score of 1. Thus, even though all students failed the test, most students are still above average. I wanted to see how many students had to fail in order to keep the average less than 4. I started with "(4(100-x)+x)/100<4" and ended with "400<400+3x", which is true for all values of x greater than 0. I realized that, if students could only receive one of two possible scores, in a binary system or a a pass/fail test (which all high school exit exams that I know of are), every student who passed the test would be above average (assuming that one student failed the test). Banaticus 22:52, 2 March 2007 (UTC)

Another example

How many times have you watched a television game show where the contestant says that he has a "beautiful wife" and "two wonderful children"? I'm still waiting to see the guy with the "ugly wife" and a "bunch of bratty kids". Do they ever let those guys on the shows? Mamarazzi (talk) 07:00, 7 January 2008 (UTC)

This brings up a point. Beauty is one of those things that's notorious for being subjective. Given that many people do fall in love based at least in part on looks, and different people are attracted to different looks, doesn't it make sense that most men's wives are "above average beauty", judged by their own subjective standards? This isn't error or poor assessment on their part -- most men probably are married to women who are above average in appearance (based on the criteria they each find important to themselves).

One of the examples in the article is police officers. Each police officer probably has his own opinions of what makes a good officer, which attributes are the most important, etc. Doesn't it make sense that they would try to excel at those attributes they find important (and pay less attention to those they don't)? If this is the case, it becomes quite statistically likely that each police officer actually is above average, based on their personal criteria for what makes a good police officer.

That's not self-aggrandizement, it's a logical consequence of the fact that when there's no objective standard to judge something, and different people have different ideas about what makes something "better", that those who value particular thing are going to strive for it, and thus be truly above average at those particular parts of the equation they consider to be the important ones. And the majority of the attributes people are cited as rating themselves on in this article are subjective ones. So not only do most people see themselves as above average on those things, but if judged on the standards they believe are relevant, they're probably right to think so. --68.187.147.210 (talk) 04:45, 28 April 2008 (UTC)

I suppose it's interesting, but it's also irrelevant.

Example: Five students are asked to rate their looks on a scale of 1 to 10. Four say they rate their looks as a 9 and the last one says their looks rate as a 4. The mean of that series would be 8, making 80% of the students of above 'average' looks.

This mean is taken from the stated self-perception of the participants, rather than from an objective measure which contradicts the self-aggrandizing. Therefore it is irrelevant to this article, and demonstrates nothing relative to the Lake Wobegon effect. It also contradicts the sense of Lake Wobegon, because it is not true that everyone is above average. Furthermore, five people is insufficient for a reasonable normal distribution of attributes. Whoever wrote this probably should have read the article before they added it. --75.63.48.18 (talk) 03:56, 20 January 2008 (UTC)

That's not irrelevant to the article. Most of the examples of the Lake Wobegon effect involve self-assessment of subjective traits. And the "Lake Wobegon Effect" names the tendency of people to rate themselves or their group as "above average", not everyone. And finally, five people is a great number to demonstrate the mathematical basis for the argument -- the author was simply demonstrating in an easy to follow example how it's mathematically possible for the majority of a sample to be above average. --68.187.147.210 (talk) 04:30, 28 April 2008 (UTC)

The Lake Wobegon Effect refers to people believing themselves to be above the median of a group in some trait, not the mean. By definition only half of a population can be above the median. Thus a mathematical example using the mean is in fact irrelevant. The fact that the word "average" is used to denote both median and mean is the source of this confusion.

This article needs drastic surgery

"Lake Wobegon Effect" is used in the literature (e.g. Rudiger Pohl (Ed.) 2004. Cognitive Illusions Psychology Press) as a synonym for what is also called Illusory Superiority. The article at present has some material about this, but much more about means and medians, which doesn't have much to do with the psychological experiments. The psychologists are well aware of the difference between median and mean and other mentioned problems. There are already WP articles on the Overconfidence effect and Dunning-Kruger effect where there is a heavy overlap with this article. "Lake Wobegon Effect" does seem to be used in the literature, but it's a more culture-bound term than "Illusory Superiority", "Better-than-average effect" or "Superiority Bias", which are less opaque. I'll do some further work on this on Tuesday 27 May.MartinPoulter (talk) 14:16, 24 May 2008 (UTC)

Agreed, this is a mess. It would be better to get all the psychology material into a separate article and just have a link to it from here - "Lake Wobegon Effect" is almost never used in the academic literature on these effects (only 8 references in Web of Knowledge compared with far more talking about "better than average effect" or "above average effect" - though those are quite tricky to sift out from the mass of articles where those words are used without technical intent. The trouble is that the psychology literature is also a bit of a mess, with a number of different names for what may or may not be a single phenomenon - partly because cognitive and social psychologists have both come on the same issues but from different directions. This isn't an easy fix. seglea (talk) 22:44, 18 August 2008 (UTC)

Yes, agreed. Illusory superiority article now exists, by the way. Perhaps psychological material could go there? MartinPoulter (talk) 14:47, 19 August 2008 (UTC)