correlation between binomial and continue variables

[ro]

Hi all,

I d like know if exist some method for run a correlation between 2 variables: one continue (and normal  variate) and other binomial variate (presence and absence).

Thanks in advance
Rodrigo

[Maguin, Eugene]

Sure, it's a point-biserial and computable using the correlation command.
Gene Maguin

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of ro
Sent: Wednesday, September 28, 2011 1:13 PM
To: [hidden email]
Subject: correlation between binomial and continue variables

Hi all,

I d like know if exist some method for run a correlation between 2
variables: one continue (and normal  variate) and other binomial variate
(presence and absence).

Thanks in advance
Rodrigo


--
View this message in context:
http://spssx-discussion.1045642.n5.nabble.com/correlation-between-binomial-a
nd-continue-variables-tp4850031p4850031.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Swank, Paul R]

If the binary variable has an underlying normal distribution, that is, the construct is continuous and normal but artificially dichotomized, then a biserial correlation would be appropriate. Otherwise, it is the point biserial.

Dr. Paul R. Swank,
Children's Learning Institute
Professor, Department of Pediatrics, Medical School
Adjunct Professor, School of Public Health
University of Texas Health Science Center-Houston


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of ro
Sent: Wednesday, September 28, 2011 12:13 PM
To: [hidden email]
Subject: correlation between binomial and continue variables

Hi all,

I d like know if exist some method for run a correlation between 2
variables: one continue (and normal  variate) and other binomial variate
(presence and absence).

Thanks in advance
Rodrigo


--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/correlation-between-binomial-and-continue-variables-tp4850031p4850031.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[ro]

Thanks for you reply, its nice heard that i can do it... just one simple and fool question: the comand name is "point-biserial"??

Regards
Rodrigo

[Maguin, Eugene]

Rodrigo,

If the dichotomous variable is a true dichotomy, like gender, then use the
correlation command.

Gene Maguin

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of ro
Sent: Wednesday, September 28, 2011 2:21 PM
To: [hidden email]
Subject: Re: correlation between binomial and continue variables

Thanks for you reply, its nice heard that i can do it... just one simple and
fool question: the comand name is "point-biserial"??

Regards
Rodrigo

--
View this message in context:
http://spssx-discussion.1045642.n5.nabble.com/correlation-between-binomial-a
nd-continue-variables-tp4850031p4850248.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Mike]

Use the ordinary Pearson r.  The point-biserial is the Pearson r
calculated on one "true" dichotomous variable and one interval/ratio
scale variable.

-Mike Palij
New York University
[hidden email]


----- Original Message -----
From: "ro" <[hidden email]>
To: <[hidden email]>
Sent: Wednesday, September 28, 2011 2:21 PM
Subject: Re: correlation between binomial and continue variables

> Thanks for you reply, its nice heard that i can do it... just one simple and
> fool question: the comand name is "point-biserial"??
>
> Regards
> Rodrigo
>
> --
> View this message in context: http://spssx-discussion.1045642.n5.nabble.com/correlation-between-binomial-and-continue-variables-tp4850031p4850248.html
> Sent from the SPSSX Discussion mailing list archive at Nabble.com.
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> [hidden email] (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Swank, Paul R]

The point biserial is just a Pearson Product Moment correlation. It's distinction is  just a hold-over from the old days before computers made it easy to compute correlations. It's formula is a littler simpler to compute on a had calculator but just do a Pearson on the data (assuming the binary variable is numeric (i.e. 0,1) to get it.

Dr. Paul R. Swank,
Children's Learning Institute
Professor, Department of Pediatrics, Medical School
Adjunct Professor, School of Public Health
University of Texas Health Science Center-Houston


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of ro
Sent: Wednesday, September 28, 2011 1:21 PM
To: [hidden email]
Subject: Re: correlation between binomial and continue variables

Thanks for you reply, its nice heard that i can do it... just one simple and
fool question: the comand name is "point-biserial"??

Regards
Rodrigo

--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/correlation-between-binomial-and-continue-variables-tp4850031p4850248.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[David Marso]

Rodrigo,
Reread Gene's post (perhaps also google "point biserial correlation" -There is *NO* "point biserial" command.  Why?  Because it is *EQUIVALENT* to the Pearson PM correlation ie. CORRELATION command ;-)
HTH, David
--

ro wrote

Thanks for you reply, its nice heard that i can do it... just one simple and fool question: the comand name is "point-biserial"??

Regards
Rodrigo

[ro]

Hi,

ok, is just more simple that i thought...

Thanks for your reply
Rodrigo