Friday, May 27, 2005
The Numbers Game
I hate it when people use bad stats to push a story or cause.
The most recent one here in Des Moines centered around an article for Bike to Work Week. The Des Moines Register was reporting on people coming into work on their bicycles and at some point commented that it was kind of dangerous because drivers of cars do not treat them as their equal, or in some cases don't even see them. Some idiot wrote an editorial saying the stats do not support that lie! He quoted that only 4 bicyclists died in traffic accidents last year in Iowa compared to over a hundred motorists. Therefore bicycling is actually safer! OK, I hope I do not have to point out the stupidity here. When comparing 2 things that have a dissimiliar number pool, you have to take percentages. When you use his numbers to create percentages, by looking at fatalities per hour on the road, I am sure you will see how much safer you are in a car than a bicycle (or motorcycle for that matter).
The one that burns me the most that I see all the time is the following rant made by liberals, "Company ABC is not diverse because only Y% of their engineers are minorities compared with Z% of the population." This stat, while probably true, is totally misused time and again. They should not be comparing the diversity of their employees to the population. What they should be doing is comparing the diverstiy of their employees to the diversity of engineers graduating. Even that is misleading as you have to not only consider the last graduating class, but historically as well since most of the companies employees were hired in the past. So, then the question becomes, "Why are only X% of engineers a minority compared with Z% of the population?" This is good, because now we are trying to get to the root of the problem. Rather than accuse Company ABC of something they are not, we are trying to find the cause. Is it the fault of colleges, high schools? All we would have to do is follow the percentages down the line to see. It might even go beyond the fault of any school, but we don't know because they are too busy using the race card.
The other one that bugs me is women in management. Now, I will admit there might be a problem somewhere, but we will never know because once again it is assumed that if management is not 50% women, something is wrong. There is so much to consider before making this assumption. For example, upper management is almost exclusively men. However, these men have had years of moving up the ladder and they started when women were unfairly shut out. Now that women are not being shut out like they were in the past, we cannot expect to have all these women CO's when they are still working up through the ranks. Also, I would love to see the 2 stats on this. What percentage of women are working professionally? I bet it's below 50% which means once again we should not be using the 50% stat. Second, what percentage of women vs men want to be in management? I bet you will also find that men lead, probably only slightly.
I am not saying there aren't racist or bigot hiring practices going on, but I am saying we need the whole story to know what is truly going on. Not statistics being used poorly.
The most recent one here in Des Moines centered around an article for Bike to Work Week. The Des Moines Register was reporting on people coming into work on their bicycles and at some point commented that it was kind of dangerous because drivers of cars do not treat them as their equal, or in some cases don't even see them. Some idiot wrote an editorial saying the stats do not support that lie! He quoted that only 4 bicyclists died in traffic accidents last year in Iowa compared to over a hundred motorists. Therefore bicycling is actually safer! OK, I hope I do not have to point out the stupidity here. When comparing 2 things that have a dissimiliar number pool, you have to take percentages. When you use his numbers to create percentages, by looking at fatalities per hour on the road, I am sure you will see how much safer you are in a car than a bicycle (or motorcycle for that matter).
The one that burns me the most that I see all the time is the following rant made by liberals, "Company ABC is not diverse because only Y% of their engineers are minorities compared with Z% of the population." This stat, while probably true, is totally misused time and again. They should not be comparing the diversity of their employees to the population. What they should be doing is comparing the diverstiy of their employees to the diversity of engineers graduating. Even that is misleading as you have to not only consider the last graduating class, but historically as well since most of the companies employees were hired in the past. So, then the question becomes, "Why are only X% of engineers a minority compared with Z% of the population?" This is good, because now we are trying to get to the root of the problem. Rather than accuse Company ABC of something they are not, we are trying to find the cause. Is it the fault of colleges, high schools? All we would have to do is follow the percentages down the line to see. It might even go beyond the fault of any school, but we don't know because they are too busy using the race card.
The other one that bugs me is women in management. Now, I will admit there might be a problem somewhere, but we will never know because once again it is assumed that if management is not 50% women, something is wrong. There is so much to consider before making this assumption. For example, upper management is almost exclusively men. However, these men have had years of moving up the ladder and they started when women were unfairly shut out. Now that women are not being shut out like they were in the past, we cannot expect to have all these women CO's when they are still working up through the ranks. Also, I would love to see the 2 stats on this. What percentage of women are working professionally? I bet it's below 50% which means once again we should not be using the 50% stat. Second, what percentage of women vs men want to be in management? I bet you will also find that men lead, probably only slightly.
I am not saying there aren't racist or bigot hiring practices going on, but I am saying we need the whole story to know what is truly going on. Not statistics being used poorly.
# posted by chroi @ 15:33 
