well, of course. A lot of psychology can't be empirically validated, that is just the reality of it. That is why insurance companies have endorsed shorter-term, behavioral-based therapies more than longer-term, other kinds of therapies (psychoanalysis, narrative, self psychology, object relations, etc etc). Not to mention short-term therapies are more cost-effective. That doesn't mean other therapies don't work. You can still get significant data from qualitative studies. While quantitative studies seem to show that cog-B or say, solution-focused therapy, are more effective on paper than other kinds of therapies; this should be looked at with caution, as behavioral therapies are just easier to analyze with scientific methodology, due to scales and other measures of behavioral change. That doesn't say much about attitude changes or greater insight. Anyway, this is obviously a very complicated issue... I guess the message I want to get across here is not to throw the baby out with the bathwater.
Hmmm. I guess I was speaking to a general feeling (bias, really) in the mental health field, which shows up at county, state and federal levels in terms of funding especially...that tests or forms of therapy that "prove" things "quantitatively"--known as "evidenced-based practices" these days--are given higher value in terms of whether or not it's "valid". Actually, I've noticed this showing up in other scientific communities and skeptic circles as well. If a study is qualitative, it's seen as less significant than a quantitative study in psychology, even though that doesn't totally make sense. I know intellectually that they are both empirically validated- but that's not the general attitude towards it (this is totally based on what I've personally observed in the field in general terms, so please feel free to take my opinion with a grain of salt if you want
( ... )
Just like any fortune cookie, most of that stuff is so vague and widely applicable that anyone themselves into it.
The standard test for fortune cookies or horoscopes or whatever is the double-blind test. You give someone two fortunes - the one that's supposedly theirs and another one from a profile or sign selected at random - and the subject has to pick which one is theirs. With unreliable methods, subjects won't be able to differentiate the "true" one from the "false" one.
Do you know of any studies using this methodology to demonstrate your claim? I've seen studies where this has been done with astrology and tea leaves, but not the MBTI.
I was referring to the Forer Effect with my comment about testing fortunes or horoscopes. Obviously we agree that the effect exists, there's a potential for the effect here, and that's a reason to be skeptical here. The only question in my mind is to what extent the Forer Effect is actually happening rather than just potentially happening. There's a big difference between saying that acupuncture is potentially all placebo and demonstrating that it is actually all placebo.
(The comment has been removed)
who was one of the most influential folks in psychology in the 20th century!
Reply
(The comment has been removed)
A lot of psychology can't be empirically validated,
that is just the reality of it.
That is why insurance companies have endorsed
shorter-term, behavioral-based therapies more than
longer-term, other kinds of therapies (psychoanalysis,
narrative, self psychology, object relations, etc etc).
Not to mention short-term therapies are more cost-effective.
That doesn't mean other therapies don't work.
You can still get significant data from qualitative studies.
While quantitative studies seem to show that cog-B or say, solution-focused therapy, are more effective on paper than other kinds of therapies;
this should be looked at with caution, as behavioral
therapies are just easier to analyze with scientific methodology,
due to scales and other measures of behavioral change.
That doesn't say much about attitude changes or greater insight.
Anyway, this is obviously a very complicated issue...
I guess the message I want to get across here is
not to throw the baby out with the bathwater.
Reply
I have a hard time
Not reading it like
Free
Verse
Poetry.
Reply
I'll try and remember not to do it when commenting on posts!
Reply
(The comment has been removed)
Reply
The standard test for fortune cookies or horoscopes or whatever is the double-blind test. You give someone two fortunes - the one that's supposedly theirs and another one from a profile or sign selected at random - and the subject has to pick which one is theirs. With unreliable methods, subjects won't be able to differentiate the "true" one from the "false" one.
Do you know of any studies using this methodology to demonstrate your claim? I've seen studies where this has been done with astrology and tea leaves, but not the MBTI.
Reply
(The comment has been removed)
Reply
Leave a comment