Date: Sun, 3 Jul 2011 08:02:59 -0400
Reply-To: Mike Palij <mp26@nyu.edu>
Sender: "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From: Mike Palij <mp26@nyu.edu>
Subject: Re: Factor Analysis - Error - Pl. Help Me
Content-Type: text/plain; charset="iso-8859-1"
Run descriptives for the 34 variables and
(1) See if any of the variables has only one or no valid
response.
(2) See if any of the variables have zero standard
deviation (i.e., all respondent chose the same rating
scale value).
(3) Examine the remaining variables to see if the
Reponses use only part of rating scale values (e.g.,
values 1-3 or 2-4 or 4-7). Variables with such limited
ranges of response can affect correlation (i.e.,
reduce the Pearson r).
Eliminate the variables that meet condition (1) or (2).
Re-run the factor analysis.
Before doing a factor analysis it is a good idea to look
at the distribution of values for each variable -- one
can do this through frequencies and request "/histogram"
as well as examining the frequency tables (one could also
look at the boxplots for the items in the examine procedure).
One would be able to identify "pathological variables" and
exclude them from the factor analysis. However, if there
are many pathological variables, one should be concerned
with the "quality" of the scale -- running the reliability
procedure would provide more information especially
about the relationship among the variables. If the responses
are badly distributed, there would be no point in doing
a factor analysis.
-Mike Palij
New York University
mp26@nyu.edu
----- Original Message -----
From: "miryalaramesh" <miryalaramesh@gmail.com>
To: <SPSSX-L@LISTSERV.UGA.EDU>
Sent: Sunday, July 03, 2011 1:41 AM
Subject: Factor Analysis - Error - Pl. Help Me
> While working with SPSS 17.0 on the data collected and tabulated by myself,
> it is throwing an error as follows
> "*There are fewer than two cases, at least one of the variables has zero
> variance, there is only one variable in the analysis, or correlation
> coefficients could not be computed for all pairs of variables. No further
> statistics will be computed.*".
>
> Basically the collected data is about Doctors job satisfaction on about 34
> aspects considering the scale 1 = extremely satisfied to 7= extremely
> dissatisfied and 4= neutral response. The data has been collected from 200
> Doctors.
>
> pl help me how i can overcome the error specified above.
>
> Dr Ramesh Kumar Miryala
>
> --
> View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Factor-Analysis-Error-Pl-Help-Me-tp4546546p4546546.html
> Sent from the SPSSX Discussion mailing list archive at Nabble.com.
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> LISTSERV@LISTSERV.UGA.EDU (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
LISTSERV@LISTSERV.UGA.EDU (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
