Syntax problem on Parallel Analysis

[Bill Wu]

I try to run the syntax provided by Hayton, JC, etal (2004).
Organization Research Methods. 191-205 on the topic of "Factor retention
decisions in exploratory factor analysis: A tutorial on parallel
analysis". The following syntax does not work. Any thought how to make
it work? What does 5/6 mean in the command for generating random
variables? Thanks in advance.

Title "Parallel Analysis"
INPUT PROGRAM.
COMMENT
LOOP LOOP#1 = 1 To 8319
COMMENT
DO REPEAT V = V1 TO V8.
COMMENT
COMPUTE V = RND (NORMAL (5/6) + 3).
COMMENT
COMMENT 5
COMMENT
COMMENT 3
COMMENT
COMMENT
IF (V LT 1)V = 1.
IF (V GT 5)V = 5.
COMMENT
COMMENT
END REPEAT.
END CASE.
END LOOP#1.
END FILE.
END INPUT PROGRAM.
COMMENT
COMMENT
FACTOR
/VARIABLES v1 v2 v3 v4 v5 v6 v7 v8/MISSING LISTWISE/ANALYSIS v1 v2 v3 v4
v5 v6 v7 v8
/PRINT INITIAL EXTRACTION
/CRITERIA MINEIGEN(0) ITERATE(25)
/EXTRACTION PC
/ROTATION NOROTATE
/METHOD=CORRELATION.
COMMENT


Bill

=====================
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

[Jerabek Jindrich]

Hello Bill,

This syntax worked (tested with SPSS ver 15).

Title "Parallel Analysis".
INPUT PROGRAM.
COMMENT.
LOOP LOOP#1 = 1 To 8319.
COMMENT.
DO REPEAT V = V1 TO V8.
COMMENT.
COMPUTE V = RND (NORMAL (5/6) + 3).
COMMENT.
COMMENT 5.
COMMENT.
COMMENT 3.
COMMENT.
COMMENT.
IF (V LT 1)V = 1.
IF (V GT 5)V = 5.
COMMENT.
COMMENT.
END REPEAT.
END CASE.
END LOOP#1.
END FILE.
END INPUT PROGRAM.
COMMENT.
COMMENT.
FACTOR
/VARIABLES v1 v2 v3 v4 v5 v6 v7 v8/MISSING LISTWISE/ANALYSIS v1 v2 v3 v4
v5 v6 v7 v8
/PRINT INITIAL EXTRACTION
/CRITERIA MINEIGEN(0) ITERATE(25)
/EXTRACTION PC
/ROTATION NOROTATE
/METHOD=CORRELATION.
COMMENT.


<What does 5/6 mean in the command for generating random
> variables?

It generates random variable with normal distribution, mean = 0 and stddev = 5/6.
The random variables are then increased by 3, rounded and range trimed to 1 - 5. It results into randomly generated scales 1-5 .

regards
Jindra

> ------------ Původní zpráva ------------
> Od: Bill Wu <[hidden email]>
> Předmět: Syntax problem on Parallel Analysis
> Datum: 28.7.2008 21:56:22
> ----------------------------------------
> I try to run the syntax provided by Hayton, JC, etal (2004).
> Organization Research Methods. 191-205 on the topic of "Factor retention
> decisions in exploratory factor analysis: A tutorial on parallel
> analysis". The following syntax does not work. Any thought how to make
> it work? What does 5/6 mean in the command for generating random
> variables? Thanks in advance.
>
> Title "Parallel Analysis"
> INPUT PROGRAM.
> COMMENT
> LOOP LOOP#1 = 1 To 8319
> COMMENT
> DO REPEAT V = V1 TO V8.
> COMMENT
> COMPUTE V = RND (NORMAL (5/6) + 3).
> COMMENT
> COMMENT 5
> COMMENT
> COMMENT 3
> COMMENT
> COMMENT
> IF (V LT 1)V = 1.
> IF (V GT 5)V = 5.
> COMMENT
> COMMENT
> END REPEAT.
> END CASE.
> END LOOP#1.
> END FILE.
> END INPUT PROGRAM.
> COMMENT
> COMMENT
> FACTOR
> /VARIABLES v1 v2 v3 v4 v5 v6 v7 v8/MISSING LISTWISE/ANALYSIS v1 v2 v3 v4
> v5 v6 v7 v8
> /PRINT INITIAL EXTRACTION
> /CRITERIA MINEIGEN(0) ITERATE(25)
> /EXTRACTION PC
> /ROTATION NOROTATE
> /METHOD=CORRELATION.
> COMMENT
>
>
> Bill
>
> =====================
> 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

[Meyer, Gregory J]

Bill, it looks like the primary problem is that many commands were not
being read because the previous line was missing a command terminator.
Try the alternative below.

The "5/6" designation was explained in the original Comments, though
they were missing from what you posted; for the example program, 6
indicated the number of response options, 5 was the max value, and 3 was
the midpoint. You would have to change these values to reflect the kind
of rating scale used with your data.

Given that you have to run the syntax 50 to 100 times and save the
factor results after each run so you can compute the mean and 95th
percentile of the eigenvalues, it is a tedious program to use. Simpler
syntax is available from Brian O'Connor's web page
(http://people.ok.ubc.ca/brioconn/), though rely on the principal
components analysis option rather than the principal axis/common factor
analysis option (see Steger, 2006, "An Illustration of Issues in Factor
Extraction
and Identification of Dimensionality in Psychological Assessment Data,"
Journal of Personality Assessment).

Greg

************ .

Title "Parallel Analysis" .

INPUT PROGRAM.
LOOP LOOP#1 = 1 To 8319 .
DO REPEAT V = V1 TO V8.
COMPUTE V = RND (NORMAL (5/6) + 3).
IF (V LT 1) V = 1.
IF (V GT 5) V = 5.
END REPEAT.
END CASE.
END LOOP#1.
END FILE.
END INPUT PROGRAM.
EXECUTE .

FACTOR
 /VARIABLES v1 v2 v3 v4 v5 v6 v7 v8
 /MISSING LISTWISE
 /ANALYSIS v1 v2 v3 v4 v5 v6 v7 v8
 /PRINT INITIAL EXTRACTION
 /CRITERIA MINEIGEN(0) ITERATE(25)
 /EXTRACTION PC
 /ROTATION NOROTATE
 /METHOD=CORRELATION.


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Bill Wu
Sent: Monday, July 28, 2008 3:51 PM
To: [hidden email]
Subject: Syntax problem on Parallel Analysis

I try to run the syntax provided by Hayton, JC, etal (2004).
Organization Research Methods. 191-205 on the topic of "Factor retention
decisions in exploratory factor analysis: A tutorial on parallel
analysis". The following syntax does not work. Any thought how to make
it work? What does 5/6 mean in the command for generating random
variables? Thanks in advance.

Title "Parallel Analysis"
INPUT PROGRAM.
COMMENT
LOOP LOOP#1 = 1 To 8319
COMMENT
DO REPEAT V = V1 TO V8.
COMMENT
COMPUTE V = RND (NORMAL (5/6) + 3).
COMMENT
COMMENT 5
COMMENT
COMMENT 3
COMMENT
COMMENT
IF (V LT 1)V = 1.
IF (V GT 5)V = 5.
COMMENT
COMMENT
END REPEAT.
END CASE.
END LOOP#1.
END FILE.
END INPUT PROGRAM.
COMMENT
COMMENT
FACTOR
/VARIABLES v1 v2 v3 v4 v5 v6 v7 v8/MISSING LISTWISE/ANALYSIS v1 v2 v3 v4
v5 v6 v7 v8
/PRINT INITIAL EXTRACTION
/CRITERIA MINEIGEN(0) ITERATE(25)
/EXTRACTION PC
/ROTATION NOROTATE
/METHOD=CORRELATION.
COMMENT


Bill

=====================
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