About
Content
Store
Forum

Rebirth of Reason
War
People
Archives
Objectivism

Post to this threadMark all messages in this thread as readMark all messages in this thread as unreadBack one pagePage 0Page 1Page 2Page 3Page 4Page 5Page 6Page 7Page 8Page 9Page 11Page 1Forward one pageLast Page


Post 220

Tuesday, January 19, 2010 - 11:15amSanction this postReply
Bookmark
Link
Edit
Hi Merlin,

I would add that the term "subjectivist," as applied to Bayesian probability, needn't entail the "subjectivism" that Objectivists abhor. Rather than referring to arbitrary preference or whim, "subjectivist" in this context is better viewed as referring to the "personal." Under Objectivism, it may still be considered "objective" in that it still may appropriately relate conciousness with existence.

Jordan


Post 221

Tuesday, January 19, 2010 - 11:35amSanction this postReply
Bookmark
Link
Edit
"allegedly on epistemology" - Aha! Robert has figured out that ITOE is secretly just a recipe for chopped liver.

[Edit: p.s., that was sarcasm - Robert, it really isn't about chopped liver.]
(Edited by Steve Wolfer on 1/19, 11:37am)


Sanction: 6, No Sanction: 0
Sanction: 6, No Sanction: 0
Post 222

Tuesday, January 19, 2010 - 12:02pmSanction this postReply
Bookmark
Link
Edit
Robert,

Where does Objectivism say reason is a piece of cake?

Jordan

Post 223

Tuesday, January 19, 2010 - 1:34pmSanction this postReply
Bookmark
Link
Edit
Steve,

ITOE is about concept-formation and definition. If you will notice, concept-formation and epistemology are different but related fields.

(Edit: sarcasm noted.)
(Edited by Robert Keele on 1/19, 1:35pm)


Post 224

Tuesday, January 19, 2010 - 1:37pmSanction this postReply
Bookmark
Link
Edit
Jordan,

You asked: Where does Objectivism say reason is a piece of cake?

As a mental faculty, reasoning is, according to Objectivism, an active process of differentiation and integration. So as I said, if only it were that simple.

Post 225

Tuesday, January 19, 2010 - 1:43pmSanction this postReply
Bookmark
Link
Edit
Robert,

Under Objectivism, "Reason is the faculty that identifies and integrates the material provided by man’s senses." The process of differentiation and integrating you mention refers *just* to Objectivism's notion of concept formation. The two are not identical.

Jordan
(Edited by Jordan on 1/19, 3:03pm)


Post 226

Tuesday, January 19, 2010 - 1:45pmSanction this postReply
Bookmark
Link
Edit
I tried the breast cancer story problem out on another "victim" today. When I printed it off the internet and showed it to him the first thing he said was, "I've been to college, I took statistics and I know all about it." I simply suggested he take a closer look at this particular problem.

After a minute, he declared that the answer was really simple and that all he needed was a calculator, but that the answer was somewhere in the 70% range.

So I showed him the four-square graph, the Bayes formula, then showed him how they coincide to give the same answer. He smiled at my efforts but wasn't 100% convinced because 7.76% is not close enough to 80% which is the test's rate of success on women who have breast cancer.

What is it about this problem that throws off most people?



Post 227

Tuesday, January 19, 2010 - 1:48pmSanction this postReply
Bookmark
Link
Edit
Jordan,

"Identification and integration" doesn't sound very complicated.



Post 228

Tuesday, January 19, 2010 - 2:34pmSanction this postReply
Bookmark
Link
Edit
What is it about this problem that throws off most people?
The false positive rate of 9.6%.  If it were much smaller the Bayes' formula result would be much larger.  For example, 0.1% would yield 89%.


Post 229

Tuesday, January 19, 2010 - 2:44pmSanction this postReply
Bookmark
Link
Edit
Hi Robert,
What is it about this problem that throws off most people?
Some heuristics are hard to shake. You might be interested in a somewhat related thread.
"Identification and integration" doesn't sound very complicated.
It sounds broad enough to include oh-so-complicated Bayes' Theorem. In some sense, all the Theorem does is identify and integrate material provided by our senses.

Jordan


Post 230

Tuesday, January 19, 2010 - 3:21pmSanction this postReply
Bookmark
Link
Edit
Merlin,

Actually, I'm finding that the 80% figure throws them off. 80% accuracy on true positives integrates to a high probability of a positive being true.

But it doesn't correlate that easily. In the formula this is represented as A|X versus X|A. In other words, people are comparing apples to donuts.

I can only speak from my personal experience in that I had a great algebra teacher through 3 years of high school who taught us, among other things, how to make these story problems our bitches instead of letting them make us their bitches.

But notice that those who have no method - degrees and diplomas coming out the ying-yang are not a substitute for method - will always tend to fall back on intuition which hardly amounts to more than mere guesswork and reliance on past experiences.

Here's the catch: past experiences may have led to negative results, yet people will still insist on using their intuition, they still rely on guesswork, common-sense intuition, and past experiences even if the results were always negative. In other words, people will always, no matter how hard we may try to convince them otherwise, continue to rely on mediocre methods, and get mediocre results again and again.

I have observed that Objectivism only feeds this tendency toward intellectual mediocrity.




Post 231

Tuesday, January 19, 2010 - 3:33pmSanction this postReply
Bookmark
Link
Edit
Jordan,

You wrote: It sounds broad enough to include oh-so-complicated Bayes' Theorem.

Sounds to me like you used intuition to decide that. I have definitely noticed an Objectivist bias toward trying to sound pro-math, pro-science, pro-physics - while at the same time disagreeing and even vehemently denouncing some of the results because they are disagreeable to Objectivist premises.

That is not being pro-math or pro-science, it is only brushing shoulders with math and science in hopes that some of their respectability will transfer over. And it is only putting on a veneer of respectability hard-won by disciplines that have been around for centuries longer than Objectivism.



Post 232

Tuesday, January 19, 2010 - 7:24pmSanction this postReply
Bookmark
Link
Edit
"Identification and integration" doesn't sound very complicated.
................

When ye dealing with chaos theory, which pertains to real life in the real world, it becomes complicated - if it were easy and not complicated, there'd be nothing to it, and all would be doing it...

Post 233

Tuesday, January 19, 2010 - 7:34pmSanction this postReply
Bookmark
Link
Edit
Robert,

I don't know Objectivism's take on chaos theory. I was specifically referring to QM, and I now have seen objections here to Bayes' Theorem. Objectivism's "what you see is what you get" or "A is A" premise doesn't work in the real world.

Post 234

Tuesday, January 19, 2010 - 7:54pmSanction this postReply
Bookmark
Link
Edit
Robert K,

I took Robert M's point as similar to mine:  "Identification and integration," the crux of Objectivism's take on reason, can be complicated.
Objectivism's "what you see is what you get" or "A is A" premise doesn't work in the real world.
I see this in Objectivists, not in Objectivism, per se.
Sounds to me like you used intuition to decide that.
No no. Listen closely.

Jordan


Post 235

Tuesday, January 19, 2010 - 8:38pmSanction this postReply
Bookmark
Link
Edit
Jordan,

Your link didn't work. Try it and see.

Post 236

Tuesday, January 19, 2010 - 9:13pmSanction this postReply
Bookmark
Link
Edit
No. :-)

Jordan

Post 237

Wednesday, January 20, 2010 - 5:21amSanction this postReply
Bookmark
Link
Edit
I said the false positive rate throws people off, to which Robert Keele replied:

Actually, I'm finding that the 80% figure throws them off. 80% accuracy on true positives integrates to a high probability of a positive being true.
Yes, but it doesn't integrate the probability of a positive test being false. I don't believe our claims are that incompatible. I could say the 80% lures them, but the 9.6% throws them off  because they don't recognize its significance.

Here is the calculation:  p1 = Pr{cancer} = .01  p2 = Pr{positive test given cancer} = .8
p3 = Pr{positive test given no cancer} = .096
p4 = Pr{cancer given positive test} = p1*p2/(p1*p2 + (1-p1)*p3) = .01*.8/(.01*.8 + .99*.096) = .07764 or 7.8%
Play with the input numbers and calculate the results all you want. I said it was the 9.6% because p4 becomes sensitive as p3 is ratcheted down.  For example, p3 = .003 makes p4 an "intuitive" 72.9%. Setting p3 = .001 makes p4 a counter-intuitive 89%.
Here's the catch: past experiences may have led to negative results, yet people will still insist on using their intuition, they still rely on guesswork, common-sense intuition, and past experiences even if the results were always negative.
 I have observed that Objectivism only feeds this tendency toward intellectual mediocrity.
Perhaps the first explains the second.  :-)

(Edited by Merlin Jetton on 1/20, 9:54am)


Post 238

Wednesday, January 20, 2010 - 12:11pmSanction this postReply
Bookmark
Link
Edit
Merlin,

People will ignore the fact that 1% have cancer, thus they will also ignore the fact that 99% don't have cancer. The 99% figure throws the false positives sky-high. In this case the false positive rate is 92.24%. If 50% of the women in the study have cancer, then the false positive rate plummets to 10.7%

Post 239

Wednesday, January 20, 2010 - 2:15pmSanction this postReply
Bookmark
Link
Edit
People will ignore the fact that 1% have cancer, thus they will also ignore the fact that 99% don't have cancer.
The question is about breast cancer, and I believe doctors would expect to hear a reasonable probability.
In this case the false positive rate is 92.24%.
What case?
If 50% of the women in the study have cancer, then the false positive rate plummets to 10.7%
Isn't assuming 50% of women have breast cancer unreasonable?

 


Post to this threadBack one pagePage 0Page 1Page 2Page 3Page 4Page 5Page 6Page 7Page 8Page 9Page 11Page 1Forward one pageLast Page


User ID Password or create a free account.