Thank you so much for helping me with this – lots of useful feedback on tips 
and back channel (I quickly ran out of my 3 daily posts so I was talking 
directly to folks). TIPSters rock!

Now just one last question. I’m writing an article where this is a problem for 
just two correlations (in a big table with lots of correlations). Do I just add 
a footnote to those two cells and say “disregard – see Simpson’s paradox (ref)”?

Marie Helweg-Larsen, Ph.D.
Associate Professor l Department of Psychology
Kaufman 168 l Dickinson College
Phone 717.245.1562 l Fax 717.245.1971
Office Hours: Mondays and Tuesdays 2:00-3:30
http://users.dickinson.edu/~helwegm/index.html


From: John Kulig [mailto:[email protected]]
Sent: Monday, February 27, 2012 2:33 PM
To: Teaching in the Psychological Sciences (TIPS)
Subject: Re: [tips] Statistical question-correlations












OK, I think the mystery of Marie's data is solved. I was on the right track 
inititially. If you look at non-Smokers, notice the Y axis where ALL the data 
is above 2, going up to 5, and there is a slight positive correlation. Now take 
the smoker data and visually superimpose it or sketch it in. Notice two things: 
the smoker data is mostly lower, with data between 1.5 and 4.5 .. Now you'd 
think these effects would wash out, but also notice this: Non-smokers, chance 
of lung cancer = 4 OR 5, there are only a few points, all high on Y axis. When 
you look at smokers, chance of lung cancer = 4 OR 5, there are much more data 
points, between 2.5 and 4.5 on Y, enough to tilt the best fit line downward as 
X increases

Now, when you look at the combined data, on the one hand it looks positive, 
because when X = 4 or 5 the Y values overall seem higher because the _range_ of 
data seems higher, but notice that the _concentration_ of high Y points when X 
= 1 and 2 .. they are bunched up at the upper end of Y and the circles are 
squashed on top of each other.. that will pull the regression line up when X is 
low, and my first paragraph shows how it will pull down when X is high ...
==========================
John W. Kulig, Ph.D.
Professor of Psychology
Coordinator, University Honors
Plymouth State University
Plymouth NH 03264
==========================

________________________________
From: "John Kulig" <[email protected]<mailto:[email protected]>>
To: "Teaching in the Psychological Sciences (TIPS)" 
<[email protected]<mailto:[email protected]>>
Sent: Monday, February 27, 2012 2:02:25 PM
Subject: Re: [tips] Statistical question-correlations









OOPS .. didn't see the scatterplot , just notice it, let me look
==========================
John W. Kulig, Ph.D.
Professor of Psychology
Coordinator, University Honors
Plymouth State University
Plymouth NH 03264
==========================

________________________________
From: "Arlie Belliveau" 
<[email protected]<mailto:[email protected]>>
To: "Teaching in the Psychological Sciences (TIPS)" 
<[email protected]<mailto:[email protected]>>
Sent: Monday, February 27, 2012 1:37:05 PM
Subject: Re: [tips] Statistical question-correlations






Hi Marie,

Have you calculated your effect sizes? It could be that the positive
correlations of .23 and .16 are so small that, when the groups are
combined, the error (or noise) turns them into negative correlations. At a
glance, the scatter plots don't looks as though the relationships between
variables are very significant.

Just a thought.
Cheers,

Arlie
--
Arlie R. Belliveau, BA (Hons), MA
History & Theory of Psychology, PhD2
York University Department of Psychology
059 Behavioural Science Building
4700 Keele St. Toronto, ON
[email protected]<mailto:[email protected]>


On Mon, Feb 27, 2012 at 11:30 AM, Helweg-Larsen, Marie 
<[email protected]<mailto:[email protected]>> wrote:









I have a simple statistical question.

I have a sample of 307 people. 111 are in the red group and 196 are the blue 
group.
The correlation between variables x and y in the red group is r= .226 (n=111), 
p <.05 and in the blue group r=.164 (n=196), p<.05. However, when I run the 
correlation between x and y in the entire sample (red and blue combined, no 
missing data) I get a negative correlation, r=-.142 (n=307), p < .05.
Now what doesn’t make sense to me that two groups individually have positive 
and significant correlations but the two groups combined can have a negative 
and significant correlation.
So you stats tipsters. Is that statistically possible?

I have checked everything I possibly can in terms of errors in the data or the 
analyses and have found none. Some suggestions about what I ought to look at?

Marie

Marie Helweg-Larsen, Ph.D.
Associate Professor l Department of Psychology
Kaufman 168 l Dickinson College
Phone 717.245.1562<tel:717.245.1562> l Fax 717.245.1971<tel:717.245.1971>
Office Hours: Mondays and Tuesdays 2:00-3:30
http://users.dickinson.edu/~helwegm/index.html<http://users.dickinson.edu/%7Ehelwegm/index.html>



---

You are currently subscribed to tips as: 
[email protected]<mailto:[email protected]>.

To unsubscribe click here: 
http://fsulist.frostburg.edu/u?id=13091.5fbcba6b94d471416a45a34246e4403b&n=T&l=tips&o=16294

(It may be necessary to cut and paste the above URL if the line is broken)

or send a blank email to 
leave-16294-13091.5fbcba6b94d471416a45a34246e44...@fsulist.frostburg.edu<mailto:leave-16294-13091.5fbcba6b94d471416a45a34246e44...@fsulist.frostburg.edu>








---

You are currently subscribed to tips as: 
[email protected]<mailto:[email protected]>.

To unsubscribe click here: 
http://fsulist.frostburg.edu/u?id=13338.f659d005276678c0696b7f6beda66454&n=T&l=tips&o=16303

(It may be necessary to cut and paste the above URL if the line is broken)

or send a blank email to 
leave-16303-13338.f659d005276678c0696b7f6beda66...@fsulist.frostburg.edu<mailto:leave-16303-13338.f659d005276678c0696b7f6beda66...@fsulist.frostburg.edu>







---

You are currently subscribed to tips as: 
[email protected]<mailto:[email protected]>.

To unsubscribe click here: 
http://fsulist.frostburg.edu/u?id=13338.f659d005276678c0696b7f6beda66454&n=T&l=tips&o=16305

(It may be necessary to cut and paste the above URL if the line is broken)

or send a blank email to 
leave-16305-13338.f659d005276678c0696b7f6beda66...@fsulist.frostburg.edu<mailto:leave-16305-13338.f659d005276678c0696b7f6beda66...@fsulist.frostburg.edu>









---

You are currently subscribed to tips as: 
[email protected]<mailto:[email protected]>.

To unsubscribe click here: 
http://fsulist.frostburg.edu/u?id=13234.b0e864a6eccfc779c8119f5a4468797f&n=T&l=tips&o=16307

(It may be necessary to cut and paste the above URL if the line is broken)

or send a blank email to 
leave-16307-13234.b0e864a6eccfc779c8119f5a44687...@fsulist.frostburg.edu<mailto:leave-16307-13234.b0e864a6eccfc779c8119f5a44687...@fsulist.frostburg.edu>







---
You are currently subscribed to tips as: [email protected].
To unsubscribe click here: 
http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=16327
or send a blank email to 
leave-16327-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu

Reply via email to