creating nets quickly in custom tables

Hello,
I have several nominal, numeric type variables (satisfaction ratings measured on a 1-10 scale). In order to create nets for items 1-3 and 8-10 in Custom Tables, I've just been creating subtotals within the custom tables dialogue. Does anyone know of a quicker/easier way to do this? Either through the interface or through syntax? It gets cumbersome when trying to do this to several variables, so just looking for a streamlined process.  

Thank you.

=====================
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 am not sure of the term "nets".

If I am guessing correctly about what you mean, try something like this.

Recode myvar1, myvar5, myvar11 (1 thru 3 =3)(8 thru 10 =8) (else = copy)
into
myNewVar1, myNewVar5 . . .
crosstabs tables = myvar1 by myNewVar1 . . .

If are you sure your variable in only nominal?  It should be at least
ordinal.  Such scales are rarely severely discrepant for interval.



-----
Art Kendall
Social Research Consultants
--
Sent from: http://spssx-discussion.1045642.n5.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

Thanks Art. By nets, I mean I want to compare those who answer a 1,2, or 3 on
the scale vs. those who answer an 8, 9, or 10. I do this in custom tables so
I can create nested cross tabs and apply significance testing to the netted
categories. Currently, the only way I've been able to create the
nets/subtotals in custom tables is to change the variable format to nominal
(otherwise it's normally a scale/interval variable). Would the syntax you
suggest work with the ctables function?

Here is the pasted syntax if I do it via custom tables:
CTABLES
  /VLABELS VARIABLES=Q17_OverallSat SVC_LEVELS DISPLAY=LABEL
  /TABLE Q17_OverallSat [C] BY SVC_LEVELS [C][COUNT F40.0, COLPCT.COUNT
PCT40.1]
  /CATEGORIES VARIABLES=Q17_OverallSat [1, 2, 3, SUBTOTAL='B3B', 4, 5, 6, 7,
SUBTOTAL='Middle', 8,
    9, 10, SUBTOTAL='T3B', OTHERNM] EMPTY=INCLUDE POSITION=AFTER
  /CATEGORIES VARIABLES=SVC_LEVELS ORDER=A KEY=VALUE EMPTY=INCLUDE
  /CRITERIA CILEVEL=95
  /COMPARETEST TYPE=PROP ALPHA=0.05 ADJUST=BONFERRONI ORIGIN=COLUMN
INCLUDEMRSETS=YES
    CATEGORIES=SUBTOTALS MERGE=YES STYLE=SIMPLE SHOWSIG=NO.

Hope that helps clarify. Thanks.



--
Sent from: http://spssx-discussion.1045642.n5.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

[John F Hall]

Satisfaction ratings on a 1-10 scale are not Nominal, they are Ordinal, but are often treated as Scale.

 

Why don't you try instead using direct syntax to recode your ratings variables into three groups each?

 

From the Data Editor:

 

File >> New >> Syntax

 

Modify to suit your data: supply your own variable labels

 

*As a check on the original distributions.

FREQUENCIES rating1 rating2 rating3 /format notable /histogram.

*Group variables into fewer categories.

RECODE rating1 rating2 rating3 (1 thru 3 =1)(4 thru 7 =2)(8 thru 10 =3)

               into r_ rating1 r_rating2 r-rating3.

FORMATS r_ rating1 r_rating2 r-rating3 (f2.0).

VARIABLE LEVEL

               rating1 rating2 rating3  r_ rating1 r_rating2 r-rating3

               (ordinal).

VALUE LABELS

r_ rating1 r_rating2 r-rating3 1 'Low 1-3' 2 'Medium 4-7' 3 ' High 8-10'.

*Check new variables.

FREQUENCIES r_ rating1 r_rating2 r-rating3 /barchart.

 

John F Hall  MA (Cantab) Dip Ed (Dunelm)

[Retired academic survey researcher]

 

Email:          [hidden email]

Website:     Journeys in Survey Research

Course:       Survey Analysis Workshop (SPSS)

Research:   Subjective Social Indicators (Quality of Life)

 

-----Original Message-----
From: SPSSX(r) Discussion <[hidden email]> On Behalf Of Darcylynn
Sent: 02 August 2018 21:05
To: [hidden email]
Subject: creating nets quickly in custom tables

 

Hello,

I have several nominal, numeric type variables (satisfaction ratings measured on a 1-10 scale). In order to create nets for items 1-3 and 8-10 in Custom Tables, I've just been creating subtotals within the custom tables dialogue. Does anyone know of a quicker/easier way to do this? Either through the interface or through syntax? It gets cumbersome when trying to do this to several variables, so just looking for a streamlined process. 

 

Thank you.

 

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

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

[Art Kendall]

The RECODE suggestion coarsens your variable into a new one with fewer
categories.

The purpose of the CROSSTAB is to document and clarify what the new variable
is.

Try substituting the new variable name for the old variable name in the
CTABLE syntax or any other syntax you draft.



-----
Art Kendall
Social Research Consultants
--
Sent from: http://spssx-discussion.1045642.n5.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

[Javier Figueroa]

Hello!!! I hope I can help you with this.

First of all I would assign a name to the standard variables, that is, instead of Q17_OverallSat I would name it Q17_1, then Q17_2 and so on until I finished with the variable battery.

the syntax of this form would be easier to create one and then copy and only modify the variable to analyze.

If this were too complicated, I would use the recoding that Art Kendall suggests, which is a good solution.

next I leave an example using the PCOMPUTE that is similar to SUBTOTAL.

/* procedimiento para quitar multiples.

data list list

/var1 to var10.

begin data.

1 1 5 10 8 3 2 5 2 3

1 3 5 3 9 8 5 5 8 10

1 2 5 3 4 2 10 10 9 6

2 5 2 4 10 2 1 8 7 6

2 8 10 8 10 5 3 2 1 6

2 10 3 5 10 2 10 10 9 6

end data.

execute.

numeric id(f8.0).

compute id=$casenum.

execute.

variable level var1 to var10 (NOMINAL).

variable level ID (scale).

VARIABLE LABELS var1 'COMPRAR'.

VARIABLE LABELS var2 'VAR 1'.

VARIABLE LABELS var3 'VAR 2'.

VARIABLE LABELS var4 'VAR 3'.

VARIABLE LABELS var5 'VAR 4'.

VARIABLE LABELS var6 'VAR 5'.

VARIABLE LABELS var7 'VAR 6'.

VARIABLE LABELS var8 'VAR 7'.

VARIABLE LABELS var9 'VAR 8'.

VARIABLE LABELS var10 'VAR 9'.

VALUE LABELS var1

1 1

2 2.

VALUE LABELS var2 to var10

1 1

2 2

3 3

4 4

5 5

6 6

7 7

8 8

9 9

10 10.

* Tablas personalizadas.

CTABLES

  /FORMAT EMPTY='  -' MISSING='.'

  /SMISSING VARIABLE

  /VLABELS VARIABLES=var2 var1 DISPLAY=DEFAULT

  /PCOMPUTE &cat3 = EXPR(([1.00] + [2.00] + [3.00]))

  /PPROPERTIES &cat3 LABEL = "B3B (1+2+3)" FORMAT=COUNT F40.0, COLPCT.COUNT PCT40.0 HIDESOURCECATS=NO

  /PCOMPUTE &cat4 = EXPR(([8.00] + [9.00] + [10.00]))

  /PPROPERTIES &cat4 LABEL = "B3B (8+9+10)" FORMAT=COUNT F40.0, COLPCT.COUNT PCT40.0 HIDESOURCECATS=NO

  /TABLE var2 [C] BY var1 [C][COUNT 'n=' F40.0, COLPCT.COUNT '%' PCT40.0]

  /CATEGORIES VARIABLES=var2 [1.00, 2.00, 3.00, 4.00, 5.00, 6.00, 7.00, 8.00, 9.00, 10.00, &cat3, &cat4, OTHERNM] EMPTY=INCLUDE TOTAL=YES POSITION=AFTER

  /CATEGORIES VARIABLES=var1 ORDER=A KEY=VALUE EMPTY=INCLUDE

  /TITLES

    TITLE='PRUEBA'

  /COMPARETEST TYPE=PROP ALPHA=0.05 ADJUST=BONFERRONI ORIGIN=COLUMN INCLUDEMRSETS=YES CATEGORIES=ALLVISIBLE MERGE=NO.

I hope I help you in something.

Greetings.

atte

2018-08-03 9:52 GMT-06:00 Art Kendall <[hidden email]>:

The RECODE suggestion coarsens your variable into a new one with fewer
categories.

The purpose of the CROSSTAB is to document and clarify what the new variable
is.

Try substituting the new variable name for the old variable name in the
CTABLE syntax or any other syntax you draft.



-----
Art Kendall
Social Research Consultants

--
Sent from: http://spssx-discussion.1045642.n5.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

Thank you all for your suggestions.



--
Sent from: http://spssx-discussion.1045642.n5.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