|
| 95% of the people polled think Norm hates statistics |
|
| |
This is based off of evidence gathered from people after viewing this post. So let's look at the facts. The dark side: - I did not like the statistics presented in that blog.
- I made the statement that I had a pet peeve and that was "You can prove anything you want with statistics".
The light side:- I have used statistics in my posts before.
- I am a computer programmer who works with data all day. I love crunching numbers and love seeing what they say.
The truth:Statistics are necessary for making certain decisions. However there are a number of factors which cause problems when dealing with them - People who look to statistics to validate their opinion. This goes back to my pet peeve. If you want you can twist numbers to say anything you want. An example found further in the comments is whether the president is doing a good job. People on either side can find ways to make statistics prove their point. The question is "Is this a large enough, accurate enough, consistent enough sample with an unbiased enough demographic to actually prove anything"?
- Bad Data/Bad Data Collection Methods. In the post mentioned above, if you actually look at how data is gathered in the "latest" analysis, I think you would come to the same conclusion I did. What's there isn't a realistic measure if you actually want the truth. On the other hand, look at what it was to dispute. There isn't any indication that there was any sanity check there, which brings me to point 3.
- Percentages automatically bring validity to your point. So people who give numbers (whether "accurate" or not) automatically gain credence especially if used in conjuction with point number 1. The theory here is that if you have numbers, you must have done some level of research and so people take you as an expert. Look at the title of the post. Whether or not, you accepted 95% as the actual number, what you will find is that you have been convinced that most people you ask would say that I don't like statistics. So where did the 95% come from? I made it up. And there's more of that going on then you might like to think.
BTW, if you made it this far congratulations. :) I do not think this is the Ideal blog, nor do I even think it fits in the Best Practices. However, way too often people get scared off by numbers without ever asking "Yeah but...". |
|
| To add a comment to "95% of the people polled think Norm hates statistics" |
 |
|
|
|
|
|
|
| July 08, 2007 |
 |
|
|
| Norm, I must confess...It took me months, but I finally get your humor!! |
|
|
| July 08, 2007 |
|
|
|
| I led with a couple of statistics in the opening paragraph of this blog, and they were very convincing! I've leaned on them occasionally in the past, but I'm always careful to use them in the most objective scientific way possible. (wink) ~mike |
|
|
| July 09, 2007 |
|
|
|
Norm, 85% of the data I use is always 100% in line with the results I need. I always keep about 15% around just to show I "considered other alternatives". ;^)
Point well taken. You are so correct. Actually, there is a whole art - maybe a science - to polling and gathering statistics. I never really understood statistics anyway. Either something is or it isn't. They once asked me...if I had a bag with 30 red balls and 70 white balls what is the probability that the ball I pull out will be white? I said 50%. It will either be white or red. I am 100% sure it won't be blue. I was a trouble maker in that class. |
|
|
| July 09, 2007 |
|
|
|
I used it elsewhere, but one of my favorite sayings is 78% of statistics are made up. BTW, if you do a search trying to find that, don't focus on the 78 part of that or you're going to have trouble finding it if you know what I mean. Sue: You've only been reading my stuff since April! (29th most likely). So it only took.. erm 2 and a half months? (Sorry for the delay on the answer, I ran out of fingers while counting.) That's not bad at all. Now if you could help me figure out where to go next with the conversation in my profile that would be great! I'm not quite sure I'm ready to get to the testimony thing yet and it's starting to feel too long. Mike: And if you remember, I kept asking where you were getting your data from. Voice: Now's not the time to bring up Schrodinger's Cat is it? |
|
|
| July 09, 2007 |
|
|
|
| Schrodinger had problems with the same concept...he was just smart enough to put an experiement together to prove himself...gotta love it... |
|
|
| July 09, 2007 |
|
|
|
| "Either something is or it isn't." --only in hindsight, or from God's point of view. That's why statistics are useful to us! ~mike |
|
|
| July 10, 2007 |
|
|
|
| No, something either is or it isn't. That doesn't change; it has nothing to do with hindsight. Whether we translate the data correctly is the variable. And statistics are only useful if we are looking at the correct data or taking the data for what it is. See Walmart customer poll from link above. At the time, there was a true percentage of people who were or were not pregnant in Walmart. The fact that I was only observing the baby section led to some innaccurate conclusions. |
|
|
| July 11, 2007 |
|
|
|
| My point was stats are very useful for making inferences on what is unknown to us. Including whether something will be or won't be (future tense), not just whether it is or isn't. For example, I could use stats to predict a)whether you'll disagree with this post and b)the approximate degree to which you'll disagree. That would help me to decide whether I should leave this post or delete it now before you ever read it. You are very argumentative Norm. But then, that's not always a bad thing. :-) |
|
|
| July 11, 2007 |
|
|
|
| So am I disagreeing or not? :) |
|
|
| July 11, 2007 |
|
|
|
| And was your comment argumentative with mine? :) |
|
|
| July 11, 2007 |
|
|
|
Well..... when you start a response with "no", that is being disagreeable, isn't it? Was my comment argumentative? Well..... no. It was a clarification, since your previous post obviously missed my intent. :-) |
|
|
| July 11, 2007 |
|
|
|
No. :) OK that was for the funny factor. Was I disagreeing or clarifying? Depends on your point of view. If I missed your intent I was hardly disagreeing. If I understood your intent I was clarifying. Was what I responded to of yours a clarification of Voice's point or arguing he wasn't right? :) My points are this, an argument can't happen without two arguing. It can get started without one or the other (or both) realizing that it's happening, but it can't continue. And what appears to be argumentative to one person can just as easily be perceived as points of clarification to another. For what it's worth, I'm sure I am perceived as argumentative here. I don't like grey areas. In times of grey areas I search to see if I can find (or state) an answer. So I'll see something in a post (probably completely against someone's intent ;) ) that makes me think... Hmmm what about that and off I go. And worse yet, I'm bulldoggish (passionate if you prefer) and completely forget tact at times. I'm working on the last area, but not sure I'm going to touch the first two. :) I also don't do too many "Me too" responses, which leads to the number of "negative" responses to seem high compared to the "positive" ones. |
|
|
| July 11, 2007 |
|
|
|
a.) seems to me this evolving dialogue should be going on here, given how argumentative we've become
b.) what does the last post have to do with statistics (further proof of my assertion in a.)
c.) today it's only supposed to get up to 85 degrees here in MD (yes I realize that should have been posted here, unless you want to 1.argue about it or 2.analyze it with statistics) |
|
|
| July 11, 2007 |
|
|
|
| b) I was clarifying points and adding to the statistical data available on whether I would argue. ;) |
|
|
| July 11, 2007 |
|
|
|
I am having gnarly flashbacks of my psych stats class. R values....T values.....EGADS!
You people are freaking me out!!! The only stats I'm interested in is the percentage of Coke versus Diet Coke fans. And I read that 88.7% of my fellow MyChurch members heartily agree with me.
::ducking and running:: |
|
|
| July 11, 2007 |
|
|
|
| I will chime in on the Coke vs. Diet Coke if you can explain that 54% statistic above (More cola taste). |
|
|
| July 11, 2007 |
|
|
|
| What 54% statistic??? Dang. You made me look! (less filling...oh, wait. that's another beverage). |
|
|
| July 11, 2007 |
|
|
|
That's the problem with editing! Before I hit post on my original item I had a comment about how I hated the commercial that said that in a recent poll 54% of people said that Pepsi had more cola taste than Coke. I hated it because I have no idea what that means. I deleted it because I wasn't willing to check whether it was Pepsi that had 54% or Coke, which I still haven't done. |
|
|
Follow this blog
