Dennis is Not More Likely to be a Dentist

by on March 28, 2011 at 7:32 am in Data Source, Economics | Permalink

Pelham, Mirenberg and Jones (2002) found that the names Jerry, Dennis and Walter were the 39th, 40th, and 41st most frequent male names in the 1990 census (moreover the absolute frequency of (Jerry+Walter)/2 was almost identical to that of Dennis). But in a nationwide search they found 482 dentists named Dennis but just 257 named Walter, and 270 named Jerry, a highly statistical significant difference. Hence the meme was born, “Dennis is more like to be a Dentist.”

The expected number of dentists named Dennis, however, depends not on the frequency of Dennis in the 1990 Census but on the entire stock of people named Dennis over the past ~70 years and similarly for Walter and Jerry. If, for example, no one was ever named Jerry prior to 1989 but in 1990 the name skyrocketed to prominence following the appearance of Seinfeld then there would be no dentists named Jerry despite Jerry being a popular name in the 1990 census.

Following this logic, Uri Simonsohn proposes that instead of comparing the number of dentists named Dennis  to those named Jerry or Walter we compare the number of dentists named Dennis to the number of lawyers named Dennis. Making this comparison, Simonsohn finds that Dennis’s are just as overrepresented among lawyers as among dentists, thus the Dennis is a dentist finding is most likely due to a spurious cohort effect.

In addition, to testing the name-profession link Simonsohn reexamines many of the classics of the implicit egoism literature and finds many of them (not all and he does not challenge the experimental results) wanting. Virginia is not more likely to move to Virginia, for example. The Simonsohn paper is impressive and a great resource for anyone wanting to teach the difficulties of doing causal statistical research.

Hat tip to Andrew Gelman who comments here and here.

Bill March 28, 2011 at 9:49 am

Names are also correlated to parents social economic status and to ethnicity. SES can determine a Childs educational attainment.

How many blacks or Hispanics do you know named Dennis?

Rahul March 28, 2011 at 9:54 am

Based on a brute-force google search:

dentist lawywer
dennis 2,330,000 13,100,000
walter 2,330,000 13,000,000
jerry 2,780,000 10,700,000

GE March 28, 2011 at 10:12 am

This is my favorite name related study:
http://aler.oxfordjournals.org/content/11/1/112.abstract

K. Vroooman March 28, 2011 at 10:33 am

Using Wolfram alpha ( which uses US census data, US birth records, SSA records, and mortality tables)

Intersting graphs are % history of US births and Estimated current age distributution

Dennis frequency peaked in terms of % of the US population in the 1945-1955 period. Stands to reason given the post WWII increase in living standards, realted increase for services and more available education from the GI Bill that this age cohort would over-represent professional degrees in comparision to the other two names.

http://www.wolframalpha.com/input/?i=walter+dennis+jerry

Rahul March 28, 2011 at 10:17 am

The observation seems pretty rational if you dig up some more data.

http://www.wolframalpha.com/input/?i=Jerry%2C+Dennis+and+Walter

Pehlam’s paper shows Dennis is about 1.9 times as common as Walter. ( 482 named Dennis but just 257 named Walter ) Now here is the 2010 population distribution of names:

http://www4b.wolframalpha.com/Calculate/MSP/MSP88819f0hc7gbi73b3ib000033c2ag5a5f8f201b?MSPStoreType=image/gif&s=6&w=500&h=235

Bingo! For most ages in the age group of 40 to 60 years, the relative ratio of Dennis: Walter is about the same, i.e. 1.9x

Further circa 1950 the popularity of Dennis seems to have peaked. On the other hand Walter is monotonically going out of vogue right since the late 1800′s:

http://www4b.wolframalpha.com/Calculate/MSP/MSP17319f0hd7aagi2904a00004807bc9h00ifcebe?MSPStoreType=image/gif&s=28&w=496&h=228

From 1940 to 1960 Dennis is more than twice as popular than Walter. And this is the baby boomer generation. I’d expect lot of circa 2002 practicing dentists to be born in this era.

Overall there doesn’t seem much of an anomaly at all. The Dennis versus Jerry effect does not have an equally strong explanation in data but the same trend holds: Historically, simply many more Dennis’ have been born in the birth-years of relevance.

Rahul March 28, 2011 at 10:21 am

The observation seems pretty rational if you dig up some more data.

Pehlam’s paper shows Dennis is about 1.9 times as common as Walter. ( 482 named Dennis but just 257 named Walter ) Now here is the 2010 population distribution of names:

Bingo! For most ages in the age group of 40 to 60 years, the relative ratio of Dennis: Walter is about the same, i.e. 1.9x

Further circa 1950 the popularity of Dennis seems to have peaked. On the other hand Walter is monotonically going out of vogue right since the late 1800′s:

From 1940 to 1960 Dennis is more than twice as popular than Walter. And this is the baby boomer generation. I’d expect lot of circa 2002 practicing dentists to be born in this era.

Overall there doesn’t seem much of an anomaly at all. The Dennis versus Jerry effect does not have an equally strong explanation in data but the same trend holds: Historically, simply many more Dennis’ have been born in the birth-years of relevance.

JasonL March 28, 2011 at 10:27 am

I’m glad our great minds and institutions are engaged in such important research. Maybe if we find just the right names for our children, we can overcome the great stagnation. A nation of Scott the Scientists or Edward the Engineers, perhaps.

Jay March 28, 2011 at 10:52 am

“thus the Dennis is a dentist finding is most likely due to a spurious”

Alex, you just lost every left-wing nut with the use of the word spurious. I mean come on, Y=mx+b where Y is GDP growth and X is 1 for democratic president and 0 for republican president explains everything!

Popeye March 28, 2011 at 11:28 am

It certainly doesn’t explain why you think Republican Presidents are better for the economy than Democratic Presidents.

Andrew Edwards March 28, 2011 at 11:46 am

It’s too bad that this is so hard to test – which party is better for the economy is one of those things where we’re all so hopelessly biased that the only chance we have is to look at the data. And it would be interesting to know the right answer to, and would help us make practical decisions, and all kind sof good stuff.

But the testability of the claim is almost nil – before 1960ish the parties were quite different, and so your sample is hopelessly small. E.g. the Dems benefit hugely from Bill Clinton’s time in office. It’s very tempting (especaily, I imagine, for republicans) to treat him as an outlier, but if you do that you’ve thrown away basically a third of your sample of years of Democratic presidents since 1960.

Dean Sayers March 28, 2011 at 1:59 pm

They’re both bad in different ways. They both try to engorge the same groups via graft and to accumulate wealth into the few hands that can afford to invest in government bias.

I think the above comment by Jay is not in keeping with good faith. That kind of prejudice betrays one’s own low-level of discourse, by trying to paint a general tendency as universally ignorant of a topic.

The neo-Keynesians have some severe shortcomings, but I think this has a lot to do with a narrow field of inquiry – not their apparent bias towards Democratic presidents.

Andrew Edwards March 28, 2011 at 5:19 pm

Definitely agree with the second part. This is part of a new hobby of responding rationally and judiciously to trolls.

anonygoat March 28, 2011 at 11:34 am

How about a menace?

Todd March 28, 2011 at 12:07 pm

“Spurious cohort effect” would be a perfect way to describe my last trip to the dentist. No Dennis, though.

Jim March 28, 2011 at 12:18 pm

I just got the entire nation of China to flip the same penny 1000 times.

Most people got heads between 450-550 times.

But this one guy got it to come up heads 804 times!!

WHAT COULD IT MEAN?!?!?!?!

Dean Sayers March 28, 2011 at 1:18 pm

This seems to point a lot more to cultural divergence along class lines than it does to mystical notions about names-begetting-careers. Some kind of normalization along income brackets might shed more light on the issue, if it is possible.

Lord March 28, 2011 at 2:13 pm

In the weird statistics department, anyone notice Political Calculations division of 56709 assaults in a population of 31624963 is not 770 but 179, way lower in Canada than the US?

Richard March 28, 2011 at 3:35 pm

I once had a barber named Jerry Barber. And our hometown has a proctologist named Dr. Butt.

dearieme March 28, 2011 at 3:53 pm

I’ve met only two people called Dean. 50% was an academic.

Charles Collom March 28, 2011 at 11:23 pm

If you’re using US Census data, isn’t that population data and not sample data? Therefore, isn’t “statistical significance” overkill since it is a tool used to make inferences about the population and you already know the population data?

Joanne Jacobs March 29, 2011 at 5:05 pm

“Dennis the Menace” began running in 1951. As the cartoon became more popular, I’d guess fewer parents chose “Dennis.”

Comments on this entry are closed.

Previous post:

Next post: