Thursday, July 13, 2006

 

Discrimination Charges

In my neverending quest to demonstrate correct usage of statistics, and to of course promote my personal agenda items, I periodically lecture... um, rant and rave about a subject for which the numbers are readily available... if only the people reading them would understand what they mean.

For example, the only objective measurement of something like discrimination (in any form) is the verifiable reporting of the incidents and the accumulation of that data by a reputable agency. I stipulate that the U.S. Equal Employment Opportunity Commission is just such a reputable authority.

The first and main caveat, as should not surprise anyone, is that not all incidents get reported. It is incumbent on the reader of the statistics to keep that in mind, and to view all numbers and analyses under the heading "for the data reported". Social scientists can and do gather auxiliary data that can reasonably indicate the proportion of reported to unreported. If this is available, it may be validly mentioned, along with the statistically rigorous application of the error factors concerning how the reported data informs the unreported data.

So, some tidbits from the EEOC data.

Total reported cases have remained essentially flat over the 14 years worth of data displayed. The annual average for the period was 81,143, bracketed by 72,302 in the first year and 75,428 in the last year. I would like to see an analysis of the large peak in 1993-95, and the lesser peak from 2000-04. Was it sunspots, or did enforcement change?

Except for what may be an anomalous peak in 1992, racial charges have remained virtually flat around the mean of 35.5%. Can this be attributed to a last bastion of racial bigotry embedded in some corner of our society? If so, why isn't the federal govt focusing on and cracking down on them? Or it is just the lowest we can go, there being always and forever the chance that a person will find racial discrimination?

Charges of religious discrimination, the lowest rate of incidence with disability, increased from 1.9% in 1992 to 3.1% in 2005, but the part that caught my eye was the steady annual increase starting in 1996, with a 0.4 point increase in 2002, breaking the 0.1 point per year trend. I bet we could get Muslims to comment on that.

Let's dig further the religion charges, and see how that total number in 2005 breaks down.

Receipts: 2,340
Resolutions: 2,352 (we have to keep in mind that there is carryover from one calendar year to the next.)

By type of resolution.
Settlements: 227
Withdrawals w/Benefits: 98
Administrative Closures: 384**
No Reasonable Cause: 1,442**
Reasonable Cause: 201*
Successful Conciliations: 36
Unsuccessful Conciliations: 165
Merit Resolutions: 526*
* These two categories are not part of the total resolution count; merit resolution is an internal subtotal.

I leave it to you to read the definitions of these terms. I will offer a personal observation.

1,826 [**] out of 2,352 resolutions were not in favor of the person(s) bringing the charge. That's 77.6%. No reasonable cause was found in 61.3% of the charges. This tells me (IMO, of course) that the vast majority of workplaces can be expected to be reasonable about employing religious minorities, and make an effort to prevent discriminatory practices from occuring. Someone else might say that this really means that too many employers are getting away with it. What do you think is more reasonable to assume?

Food for thought...

Comments:
I wonder how much of the "no reasonable charge" items were brought on by someone with a... bee in his bonnet... (the nicest wording I can think of) and how much money and time was wasted dealing with it.

I admit I've had several bones to pick with the whole discrimination issue, best explained by an "are you a racist" test that claimed that I was one... Not because I had any trouble linking the word "beautiful" or "cute" or "honest" with the image of a black, Hispanic, Asian, whatever face, on which I showed no hesitation at all, but because I was not able to readily link the words "ugly" or "evil" with the image of a European-descent face and my response times lengthened.
 
Post a Comment



<< Home

This page is powered by Blogger. Isn't yours?