IML Statistics by Region

IMsL 2008 Hobbit put together this really fascinating analysis of of the ratio of judges to contestants in IML.  Love me some statistics! 

International Mr Leather contestant to judge statistics by region

Why I did this:

I have heard many people complain that their region is always underrepresented on the IML judging panel. I decided to "run the numbers" and generate statistics to prove or disprove this.

Hypothesis:

the percentage of judges for International Mr Leather from the Eastern United States is overrepresented on the judging panel relative to the percentage of contestants from the same region.

Summary

I found that for the Eastern United States for the 5 years that I gathered data the average ratio of judges to contestants was 2.5 judges for each 1 contestant (2.5:1). The greatest disparity year was 2005 when 44% of the judges were from the Eastern United States and only 11% of the contestants were from the Eastern United States. The least disparity year was 2006, when 11% of the judges and 9% of the contestants were from the Eastern United States. Thus showing that the number of judges from the Eastern region is consistently over-represented relative to the number of contestants from the Eastern region.

There are some other interesting statistics to note:

• Each year the majority of the contestants are from the Western Region.

• The Mid-West region is very underrepresented - the average percentage ratio is 1 judge to 2.5 contestants.

• The West region average ratio is underrepresented with 1 judge to 1.25 contestants.

• The South region average ratio is slightly underrepresented with 1 judge to 1.15 contestants.

• The Non-US region slightly over-represents judges, with an average ratio of 1.25 judges to 1 contestant.

• The Eastern region had the fewest total number of contestants at 34 (12.8%)

• The West region had the greatest total number of contestants at 74 (28%)

• The Eastern region had the greatest total number of judges at 15 (33%)

• The Mid-West had the fewest total number of judges at 5 (11%)

• The total number of contestants was 264

• The total number of judges was 45

To establish my raw data I used the following criteria:

Regions: I used Wikipedia’s definition of the 4 regions of the United States. Wikipedia is not considered an authoritative source, however I used it because it represents popular opinion of what the American public thinks.

East:

Delaware
Pennsylvania
New Jersey
New York
Maine
New Hampshire
Vermont
Massachusetts
Rhode Island
Connecticut
Maryland
Washington DC

Mid West:

Illinois
Indiana
Iowa
Kansas
Michigan
Minnesota
Missouri
Nebraska
North Dakota
Ohio
South Dakota
Wisconsin

South:

Florida
Georgia
North Carolina
South Carolina
Virginia
West Virginia
Alabama
Kentucky
Mississippi
Tennessee
Arkansas
Louisiana
Oklahoma
Texas

West:

California
Oregon
Washington
Arizona
Colorado
Idaho
Montana
Nevada
New Mexico
Utah
Wyoming
Alaska
Hawaii

Contestants and judges

• I used the past five years of contestant/judge panels. I was unable to find complete information for 2009 so I used 2004 through 2008. Partial data for 2009 is provided below, but not included in my overall statistical data or my summary.

• The contestant list was obtained from the official International Mr Leather website.

• The judges list was obtained from what was most readily available on-line.

• To determine where the judge was from, I used the residence that they were listed as being from at the time that they were a judge. So if a judge lived in Germany when they were a judge, but originally lived in California and later moved to Washington DC, the judge would be listed as a non-US judge because the judge did not live in the US at the time that they judged.

2008 region	judges	contestants
East	33.3%	17.6%
South	0	13.7%
Mid-West	22.2%	23.5%
West	22.2%	27.4%
Non-US	22.2%	17.6%

2007 Region	Judges	Contestants
East	55.5%	16.9%
south	11.1%	15%
Mid-west	0	32%
West	22.2%	22.6%
Non-US	11.1%	13.2%

2006 Region	Judges	Contestants
East	11.1%	9.8%
South	33.3%	15.6%
Mid-West	22.2%	29.4%
West	11.1%	27.4%
Non-US	22.2%	17.6%

2005 Region	Judges	Contestants
East	44.4%	11.5%
South	0	15.3%
Mid-West	11.1%	26.9%
West	22.2%	32.6%
Non-US	22.2%	13.4%

2004 Region	Judges	Contestants
East	22.2%	8.7%
South	22.2%	17.5%
Mid-West	0	26.3%
West	22.2%	29.8%
Non-US	33.3%	17.5%

2009

54 Contestants I was unable to find a list of 2009 contestants

East:

South:

Mid-west:

West:

Non-US:

Judges

East: 2 Judges 22.2%

South: 1 judge 11.1%

Mid-West: 1 judge 11.1%

West: 3 judges 33.3%

Non US: 2 judges 22.2%

Region	Judges	Contestants
East	22.2%
South	11.1%
Mid-west	11.1%
West	33.3%
Non-US	22.2%

2008

51 contestants

East: 9 contestants representing 17.6%

South: 7 contestants representing 13.7%

Mid-West: 12 contestants representing 23.5%

West: 14 contestants representing 27.4%

Non-us 9: contestants representing 17.6%

Judges

East: 3 judges 33.3%

South: 0 judges

Mid West: 2 judges 22.2%

West: 2 judges 22.2%

Non-us: 2 judges 22.2%

2008 region	judges	contestants
East	33.3%	17.6%
South	0	13.7%
Mid-West	22.2%	23.5%
West	22.2%	27.4%
Non-US	22.2%	17.6%

2007

53 Contestants

East: 9 contestants 16.9%

South: 8 Contestants 15%

Mid-West: 17 contestants 32%

West: 12 contestants 22.6%

Non-US: 7 contestants 13.2%

Judges

East: 5 judges 55.5%

South: 1 judge 11.1%

Mid-West: 0

West: 2 judges 22.2%

Non-US: 1 judge 11.1%

2007 Region	Judges	Contestants
East	55.5%	16.9%
south	11.1%	15%
Mid-west	0	32%
West	22.2%	22.6%
Non-US	11.1%	13.2%

2006

51 contestants

East: 5 contestants 9.8%

South: 8 contestants 15.6%

Mid-West: 15 contestants 29.4%

West: 14 contestants 27.4%

Non-US: 9 contestants 17.6%

Judges

East: 1 judge 11.1%

South: 3 judges 33.3%

Mid-West: 2 judges 22.2%

West: 1 judge 11.1%

Non-US: 2 Judges 22.2%

2006 Region	Judges	Contestants
East	11.1%	9.8%
South	33.3%	15.6%
Mid-West	22.2%	29.4%
West	11.1%	27.4%
Non-US	22.2%	17.6%

2005

52 Contestants

East: 6 contestants 11.5%

South: 8 contestants 15.3%

Mid-West: 14 contestants26.9%

West: 17 contestants 32.6%

Non-US: 7 contestants 13.4%

Judges:

East: 4 judges 44.4%

South: 0

Mid West: 1 judge 11.1%

West: 2 judges 22.2%

Non-US: 2 judges 22.2%

2005 Region	Judges	Contestants
East	44.4%	11.5%
South	0	15.3%
Mid-West	11.1%	26.9%
West	22.2%	32.6%
Non-US	22.2%	13.4%

2004

57 Contestants

East: 5 contestants 8.7%

South: 10 contestants 17.5%

Mid-West: 15 contestants 26.3%

West: 17 contestants 29.8%

Non-US: 10 contestants 17.5%

Judges

East: 2 judges 22.2%

South: 2 judges 22.2%

Mid-West: 0

West: 3 judge 33.3%

Non-US: 2 judges 22.2%

2004 Region	Judges	Contestants
East	22.2%	8.7%
South	22.2%	17.5%
Mid-West	0	26.3%
West	33.3%	29.8%
Non-US	22.2%	17.5%

The following are responses to questions I originally received when I posted on the titleholder list.

I was asked by someone to run the numbers on how many of each region made the top 20. I don’t have access to the lists of who made the top 20 for each year. If anybody has that information and sends it to me, I’d be happy to do that.

In order to make sure that the statistical data accurately reflected regional representation I set up my parameters before I started doing the math

I used the following criteria:

I wanted 5 years of contestant/judge data. I originally wanted to use 2005 – 2009, but was unable to find a contestant list for 2009, so I went back one year.

Paul Nickerson asked the following questions:

How do you factor judges who have lived in multiple regions in your statistics? See response above

Is it even possible? Yes, parameters are set up beforehand and apply across all raw data regardless of how it impacts the outcome.

What if they're more partial to a region they've previously lived in and not the current one they represent? Now you are asking a question about who is picked as the winner. This is not an issue that I have addressed. I have addressed only where the judges are from, and where the contestants are from. The judging panel has no say in deciding who runs for IML

And how might one know something like this to figure into these kinds of statistics? See above. Your question is about who is picked as the winner which is not an issue I have addressed, therefore this question is moot.  

Comments

[Hobbit Joost]

I was sent the following questions by a former International Mr Leather.

Here are his questions and my response.

“there were originally 4 judges from outside the US for IML 2004. “

“how does this compare with the number of gay male leather folk living in these regions?”

“As you know the 3 states with the largest gay population are New York, California, and Florida.”

“Who at IML did youi speak with about how they pick judges?”

“When are you releasing statitics about IMsL?”

When I first posted this information to the titleholder list Randall Kinner pointed out that last minute judge change to me. I re-ran the numbers with the updated information. Changing the one judge from inside the US to outside the US altered the breakdown by a statistically insignificant amount, and therefore did not affect the total outcome; that compared to the contestants, the East Coast is over-represented in numbers every year in included in my survey.

The reason I did this was because I would hear many people complaining about how 'their' geographical region always got short shrift when it came to the judges at IML. Bear in mind that 'their' geographical region was always different people from many different regions. So since I work in research with statistics I decided to run the numbers and see what the real truth about the region the judges are from.

(I am aware that there are many other things they look at in judge selection such as the different parts of the community people represent, wanting to have one woman each year etc. This survey was only about geography)

What I found out in doing so was that everybody that complained was right and wrong at the same time. There is one region that is always over-represented, but all other regions were sharing the burden of that over-representation.

Since I work in research I understand that in order to have accurate data you have to set up the parameters for your study *before* you begin to run the numbers. If you start to run numbers, then adjust your parameters, you skew your data. I was not invested in any outcome over another, I just wanted to satisfy my curiosity, but still wanted to do as accurate of a survey as possible.

The comparison I chose was to compare the region the judges were from to the region the contestants were from.

In order to obtain my data I divided up the geographical regions according to what is listed in Wikipedia. Wikipedia is not considered an authoritative source, however it represents what is the most popular thought among the public.

The judges lists were what was obtained readily by searching on google. I was not able to find a consistent source for the judges lists. The only year in question was 2004. The first time I ran the numbers, which is the one you are referencing, the list was incorrect by one judge. Randall Kinner pointed out the error and I ran the numbers again. However the numbers did not have a statistically significant change.

I chose to do this with 5 years of data. I would have gone one year later, 2005 – 2009, however when I did this the contestant list for 2009 was not available yet.

The contestant lists I obtained from the IML web site itself.

I didn’t speak to anybody that works on IML. All of the information I used was obtained on-line without any privilege. In other words, it was information that is accessible to anybody.

I have been asked to do many different comparisons.

I compared the judges to the contestants only.

I did not compare the judges to the total number of gay people or gay leather people living in any given region. The total number of gay leather people living in a region was irrelevant to the data I was looking at which was comparing the number of contestants to the number of judges. Not the total population.

Somebody asked how the judges compared to the number of men placing in the top 20. Also data I didn’t look at. I think it would be interesting to see what those numbers look like, but the data pool would be so small at that point it would start to become statistically insignificant.

I could run the numbers on IMsL but I question the validity at this point in their history. IMsL has been owned by Glenda for only 3 years. Up until 2007 it was owned by Amy Marie Meek and she had her own system. The data on IMsL wouldn’t be indicative of anything because of the change in ownership resulting In the radical change in how the contest is conducted. However, when a few more years pass, it will be interesting to see how things are going under new ownership.

I hope this helps answer your questions. Having been the one conducting this research I know exactly what I mean, but I might not be so clear to others.