|
I didn’t read about the story because it simply discussed what was already obvious, but what was the methods used? You have to be real careful to get accurate results and it’s real easy to go wrong (see that study about 655,000 dead Iraqis). |
| |
Written By:
abw
URL:
http://abw.mee.nu
|
|
lol, another example |
| |
Written By:
abw
URL:
http://abw.mee.nu
|
IA MATTERS: Kurtz claimed "a lot of journalists" are giving to Dems — but number giving at all is tiny percentage of whole
MEDIA MATTERS: Kurtz again cited report on journalists’ donations without noting that only tiny fraction gave at all I’m pretty sure that they’ve condemmed FNC on flimsier evidence than that in the past... |
| |
Written By:
shark
URL:
http://
|
Yes, if only there was some method of making inferences about a population based on data from a smaller sample. this made my morning.
Keep up the good work. |
| |
Written By:
Joel
URL:
http://
|
|
Jon, Jon, don’t you know that 8 out of 5 people are ignorant of statistics? ;) |
| |
Written By:
Wacky Hermit
URL:
http://organicbabyfarm.blogspot.com
|
In this particular instance I’d have to say that Media Matters has it right. As far as I can tell this isn’t a sample of journalists, this is a tally of records.
This article is talking about looking at FEC records of journalists donatiions and finding 143 records. I would be interested in finding out
1. how exactly they defined journalist in this case, and what the overall population of people who fit this definition is. And 2. how many of the FEC records they studied, was it a sample? (If so I’d have to think more about the validity of this study because I do not know if you can draw conclusions about the population of journalists by looking at a sample of FEC records.)
Assuming that they looked through all of them, finding 143 FEC records of donations is not really that informative about journalists as a whole. NOw we know that 143 donated to political causes and are therefore political. It would then be interesting to know what proportion of the journalistic community this constituted. If we knew that we could then compare it to the rest of the country and see if journalists donate more or less frequently than non journalists. (I would hope less, a lot less. But alas, we do not know.)
As far as I can tell, they didn’t do any of that analysis, so we can’t really draw that many conclusions about how many journalists donated beyond that there were 143. We can say that those who donate to political causes are more likely to support liberal causes than conservative ones.
It is important to note that you cannot blow up this "sample" of 143 journalists who donated to determine the political leanings of the whole. This is not valid because this sample is inherently biased, it is a sample only of people who care enough about their political leanings to donate money. (perhaps in violation of company policy?) It is a bad assumption to assume that the rest of the population of journalists would look the same. (No matter how much everyone wants to confirm their belief that Journalists are all liberals.)
As far as the Media Matter v. Kurtz, the information this article gives us is that 143 "Journalists" out of X "Journalists" donated to political causes. I don’t really see how we can know if this qualifies as "a lot" without having at least an estimate of the value of X. |
| |
Written By:
Shinobi
URL:
http://liesandstatistics.blogspot.com
|
