Reverse coding continuous variables

[Martha]

Hi,

I have to factor analyze many continuous variables which are not all coded in the same direction (some are assumed to increase risk of smoking, others to decrease risk). 1) Do I need to reverse code them and 2) If so, how? Examples of my variables are NDVI index or % employed people.

Thanks for your help!

[Maguin, Eugene]

Hi Martha,

1) No, you don't. Some variables will have negative factor loadings; others
will be positive. If you've chosen your variables well, the factor loadings
will all be large, which is what counts.

2) To reverse score, subtract each value from the theoretical maximum for
the variable. Example: %unemployed= 100-%employed.

Gene Maguin



-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Martha
Sent: Monday, August 08, 2011 10:43 AM
To: [hidden email]
Subject: Reverse coding continuous variables

Hi,

I have to factor analyze many continuous variables which are not all coded
in the same direction (some are assumed to increase risk of smoking, others
to decrease risk). 1) Do I need to reverse code them and 2) If so, how?
Examples of my variables are NDVI index or % employed people.

Thanks for your help!

--
View this message in context:
http://spssx-discussion.1045642.n5.nabble.com/Reverse-coding-continuous-vari
ables-tp4677934p4677934.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