How to exclude some results in SPSS

[Promises]

Hi everyone,
I am currently analysing data based on a five point likert scale – from Not at all (1) to extremely (5) on 1. 40 goal items; 2. 40 satisfaction with goal achievement items, and; 3. 40 ‘gap items’ - created by subtracting goal items from satisfaction items, which results in a score of -4 to 4.

I want to exclude any score below 0 (that is -4 to -1) in my analysis using the gap items, and I am wondering if anyone can guide me on how to do this in SPSS (if anyone is also able to provide a syntax code, I will be more than happy to explore this as well).

Thanks in advance.

[Rich Ulrich]

First, points on terminology --
A likert scale is made up of a number of likert items, and the
items are not ordinarily manipulated separately (as you are doing).
Likert items generally express agreement on a symmetrical scale of
agreement/disagreement ... which does not seem to fit "satisfaction".
Some readers and reviewers will not like this looser usage.

You can declare as Missing the values of (Lo through 0)  if you want
to perform some particular comparison that will omit the negative
scores.   -  I am curious why you are doing this, because it feels like
an ad-hoc choice where subsequent results will be more misleading than
revelatory.  What is it that degree of satisfaction is being compared to?
--
Rich Ulrich

> Date: Sat, 15 Nov 2014 16:41:27 -0700

> From: [hidden email]
> Subject: How to exclude some results in SPSS
> To: [hidden email]
>
> Hi everyone,
> I am currently analysing data based on a five point likert scale – from Not
> at all (1) to extremely (5) on 1. 40 goal items; 2. 40 satisfaction with
> goal achievement items, and; 3. 40 ‘gap items’ - created by subtracting goal
> items from satisfaction items, which results in a score of -4 to 4.
>
> I want to exclude any score below 0 (that is -4 to -1) in my analysis using
> the gap items, and I am wondering if anyone can guide me on how to do this
> in SPSS (if anyone is also able to provide a syntax code, I will be more
> than happy to explore this as well).
>
> Thanks in advance.

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

[Art Kendall]

I concur with Rich about the terminology.
Also you end anchors my cause restricted responding, e.g.,
almost not at all vs not at all.

I also wonder why you are doing this restriction on the range of the difference variables.

open a new instance of SPSS. Past the syntax below into a syntax window. Run it.
Is this what you want?

* simulate 2 Extent scales.
* used 1 for none only because OP did.
* it is usually cognitively better to anchor none, not at all,
   never  or almost never  with Zero.

new file.
input program.
   vector ExtentGoal(10,f1).
   vector ExtentSatisfied(10,f1).
   vector Diff(10,f1).
   loop id = 1 to 25.
      loop #index = 1 to 10.
         compute ExtentGoal(#index)      =RND(RV.UNIFORM(.5,5.5)).
         compute ExtentSatisfied(#index) =RND(RV.UNIFORM(.5,5.5)).
         compute Diff(#index) = ExtentSatisfied(#index) -ExtentGoal(#index).
      end loop.
      end case.
   end loop.
   end file.
end input program.
value labels ExtentGoal1 to Extentgoal10, ExtentSatisfied1 to ExtentSatisfied10
   1 'Not at all'
   5 'Extremely'.
frequencies variables = ExtentGoal1 to Extentgoal10, ExtentSatisfied1 to ExtentSatisfied10
   diff1 to diff10.
missing values diff1 to diff10 (low thru -1).
frequencies variables= diff1 to diff10.

[Promises]

Hello,
Thanks for your responses.
Below is the syntax command I’ve got, but it won’t work, I think I’m missing or mixing-up something.

COMPUTE filter_$=(FamsupportGi TO Masters >= 0).
VARIABLE LABEL filter_$ 'FamsupportGi TO Masters >= 0 (FILTER)'.
VALUE LABELS filter_$ 0 'Not Selected' 1 'Selected'.
FORMAT filter_$ (f1.0).
FILTER BY filter_$.
EXECUTE.

I welcome any help you are kindly able offer.

Promise

[Jignesh Sutar]

I suspect you are receiving an error on the statement:

COMPUTE filter_$=(FamsupportGi TO Masters >= 0).


As that is not a valid compute expression i.e. that highlighted in bold font. It's difficult to diagnose without knowing what data you have exactly. But I suspect the greater than zero was to be applied to a derived variable of some transformations of original variables.

[David Marso]

What exactly are you attempting to do here?
--

Promises wrote

Hello,
Thanks for your responses.
Below is the syntax command I’ve got, but it won’t work, I think I’m missing or mixing-up something.

COMPUTE filter_$=(FamsupportGi TO Masters >= 0).
VARIABLE LABEL filter_$ 'FamsupportGi TO Masters >= 0 (FILTER)'.
VALUE LABELS filter_$ 0 'Not Selected' 1 'Selected'.
FORMAT filter_$ (f1.0).
FILTER BY filter_$.
EXECUTE.

I welcome any help you are kindly able offer.

Promise

[John F Hall]

Looks to me as if you've not run anything. I've not been following this thread, but what happens if you do:

1: File >> New >> syntax:

Count x = FamsupportGi TO Masters ( 1 thru hi).
Recode x (1 thru hi = 1).
List x /cases 10.

Filter by x.
<any analysis you want, even if you use the GUI>

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Promises
Sent: 17 November 2014 12:58
To: [hidden email]
Subject: Re: How to exclude some results in SPSS

Hello,
Thanks for your responses.
Below is the syntax command I’ve got, but it won’t work, I think I’m missing or mixing-up something.

COMPUTE filter_$=(FamsupportGi TO Masters >= 0).
VARIABLE LABEL filter_$ 'FamsupportGi TO Masters >= 0 (FILTER)'.
VALUE LABELS filter_$ 0 'Not Selected' 1 'Selected'.
FORMAT filter_$ (f1.0).
FILTER BY filter_$.
EXECUTE.

I welcome any help you are kindly able offer.

Promise



--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/How-to-exclude-some-results-in-SPSS-tp5727950p5727963.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

[Bruce Weaver]

My ESPss suggests they are attempting something like this!


DATA LIST list / FamsupportGi SomeVar Masters (3F2.0).
BEGIN DATA
0 0 0
1 1 1
1 1 -1
-1 -1 -1
-1 0 0
-1 -1 1
END DATA.

COMPUTE NoNeg = 1.
FORMATS NoNeg(F1).
VECTOR v = FamsupportGi TO Masters.
LOOP # = 1 to 3.
- IF v(#) LT 0 NoNeg = 0.
END LOOP if NoNeg EQ 0.
LIST.
FILTER by NoNeg.
LIST.

Output from the FIRST LIST:

FamsupportGi SomeVar Masters NoNeg
 
      0          0       0     1
      1          1       1     1
      1          1      -1     0
     -1         -1      -1     0
     -1          0       0     0
     -1         -1       1     0
 
Number of cases read:  6    Number of cases listed:  6

Output from the SECOND LIST:

FamsupportGi SomeVar Masters NoNeg
 
      0          0       0     1
      1          1       1     1
 
Number of cases read:  2    Number of cases listed:  2

David Marso wrote

What exactly are you attempting to do here?
--

Promises wrote

Hello,
Thanks for your responses.
Below is the syntax command I’ve got, but it won’t work, I think I’m missing or mixing-up something.

COMPUTE filter_$=(FamsupportGi TO Masters >= 0).
VARIABLE LABEL filter_$ 'FamsupportGi TO Masters >= 0 (FILTER)'.
VALUE LABELS filter_$ 0 'Not Selected' 1 'Selected'.
FORMAT filter_$ (f1.0).
FILTER BY filter_$.
EXECUTE.

I welcome any help you are kindly able offer.

Promise

[David Marso]

OP wrote:
"I want to exclude any score below 0 (that is -4 to -1) in my analysis using the gap items...."
Without indicating what sort of analysis is intended.
I would hardly consider deleting ALL cases where respondent has a negative value on even a single variable a prudent/reasonable/sane course!
So, OP, WHAT IS YOUR ACTUAL ISSUE?
Omission of crucial details will lead only to further confusion!

Bruce Weaver wrote

My ESPss suggests they are attempting something like this!


DATA LIST list / FamsupportGi SomeVar Masters (3F2.0).
BEGIN DATA
0 0 0
1 1 1
1 1 -1
-1 -1 -1
-1 0 0
-1 -1 1
END DATA.

COMPUTE NoNeg = 1.
FORMATS NoNeg(F1).
VECTOR v = FamsupportGi TO Masters.
LOOP # = 1 to 3.
- IF v(#) LT 0 NoNeg = 0.
END LOOP if NoNeg EQ 0.
LIST.
FILTER by NoNeg.
LIST.

Output from the FIRST LIST:

FamsupportGi SomeVar Masters NoNeg
 
      0          0       0     1
      1          1       1     1
      1          1      -1     0
     -1         -1      -1     0
     -1          0       0     0
     -1         -1       1     0
 
Number of cases read:  6    Number of cases listed:  6

Output from the SECOND LIST:

FamsupportGi SomeVar Masters NoNeg
 
      0          0       0     1
      1          1       1     1
 
Number of cases read:  2    Number of cases listed:  2

David Marso wrote

What exactly are you attempting to do here?
--

Promises wrote

Hello,
Thanks for your responses.
Below is the syntax command I’ve got, but it won’t work, I think I’m missing or mixing-up something.

COMPUTE filter_$=(FamsupportGi TO Masters >= 0).
VARIABLE LABEL filter_$ 'FamsupportGi TO Masters >= 0 (FILTER)'.
VALUE LABELS filter_$ 0 'Not Selected' 1 'Selected'.
FORMAT filter_$ (f1.0).
FILTER BY filter_$.
EXECUTE.

I welcome any help you are kindly able offer.

Promise