Subject:
|
Re: The value of reading (was: If you could leave any book on Kjeld's nightstand...)
|
Newsgroups:
|
lugnet.off-topic.debate
|
Date:
|
Sat, 30 Mar 2002 13:57:30 GMT
|
Viewed:
|
1462 times
|
| |
|
|
In lugnet.off-topic.debate, Allan Bedford writes:
> But to be honest again, I didn't realize you were researching anything. I
> > A correlation says that when you observe X, you are to some extent likely to
> > ovserve Y as well. Y might be phrased as _not_ Z which means
> > that X and Z are
> > inversely correlated. Level of formal education and childhood socio-economic
> > status are inversely correlated.
> > To demonstrate a correlation between two factors does not at all
> > demonstrate a
> > causal link. I can't look at the data, pronounce the correlation between
> > education and SES, and say that poor families don't have equall access to
> > education. There are tons of other causal and non-causal explanations that
> > have to be explored.
> Chris, I don't mean to harp on this point, but I just don't see what you're
> trying to get at in the two paragraphs above. Can you perhaps use an
> analogy or two for those of us who haven't studied education at a university
> level?
Allan, are you saying you don't get what the difference between correlation
and causality is?
Chris's words seem pretty clear to me... there seems to be a statistical
link (a correlation) between poverty and education level, to wit, people
who are poor seem to have less formal education. But he's caveating that by
pointing out that merely noting that statistical correlation doesn't imply
causality.
That is, it doesn't necessarily follow that either:
"people are poor BECAUSE they have less education"
or:
"people have less education BECAUSE they are poor"
merely because a statistical link (a correlation, in statistical terms) is
observed between the two statistics (education level and socioeconomic
status) in the sample set. More explanation would be necessary. In fact, in
general, statistics hardly ever is useful in proving causative links
(although it is great at suggesting areas where digging around might find
some) by itself.
This is basic statistics. And that's all that Chris seems to be saying in
the paras you are questioning. It has nothing to do with having studied the
theory of education, merely with a knowledge of statistics and logic that
hopefully is being instilled in high school or earlier (my 7th grader knows
it, I just posed her the question to check).
I mostly stay out of .debate these days but this one just jumped out at me.
|
|
Message has 1 Reply:
Message is in Reply To:
59 Messages in This Thread:
  
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
