Childbirth death rates (of mothers) is on the rise, we're told. "US women are dying from childbirth at the highest rate in decades, new government figures show."
We're given various reasons for this "increase", including new reporting measures, maternal obesity, and the increased age of the mother.
Is this what the data tells us? The numbers are clear-cut, aren't they? How could there be any problem reporting these numbers, calculating rates, graphing results, and viewing the conclusion?
An Alternative Headline:
Childbirth death rates continue to decline for all ages of mothers.
Is it possible this headline describes the "same reality" as the headline above?
Yes.
Monday, August 27, 2007
The Nature of Bias
Image doctoring has been in the media lately, as the media has been taking hits for "touching up" photos. Of course, doctoring a photo by definition distorts reality, but an incorrect inference is the original photo DESCRIBES reality.
A larger part of the reporting process, however, is why the particular image was shot in the first place. Of all images a reporter could submit, why focus on the French President in a boat (to use the recent photo-doctoring incident as an example)? Does submitting the undoctored image "move us closer to the truth"? Why did the reporter claim this to be news in the first place? War coverage is a further example. Iraq is a big country. Reporters cover "what's happening" and submit images. Undoctored images don't necessarily mean "truth", as this is but a small part of "the truth, the whole truth, and nothing but the truth".
What is one to do, then?
Walter Shewhart's first rule for the presentation of data, taken from Donald Wheeler's book, Understanding Variation, assists in this regard:
"Data should always be prsented in such a way that preserves the evidence in the data for all the predictions that might be made from the data"
This is a start - a great start.
A larger part of the reporting process, however, is why the particular image was shot in the first place. Of all images a reporter could submit, why focus on the French President in a boat (to use the recent photo-doctoring incident as an example)? Does submitting the undoctored image "move us closer to the truth"? Why did the reporter claim this to be news in the first place? War coverage is a further example. Iraq is a big country. Reporters cover "what's happening" and submit images. Undoctored images don't necessarily mean "truth", as this is but a small part of "the truth, the whole truth, and nothing but the truth".
What is one to do, then?
Walter Shewhart's first rule for the presentation of data, taken from Donald Wheeler's book, Understanding Variation, assists in this regard:
"Data should always be prsented in such a way that preserves the evidence in the data for all the predictions that might be made from the data"
This is a start - a great start.
Tuesday, August 21, 2007
The (Im)Morality of Political Endorsements
It seems all so natural. The political season is in full swing, and the politicians rush to secure ... what? The endorsements of other politicians? Other leaders? It's become so natural, it seems unnatural to ask the question, "Is this good?"
Let's see.
Let's see.
Subscribe to:
Posts (Atom)