Analysis of MRSets
Locked
 
|
Hi All when using MRSet with SPSS Statistics 17 we found out that SPSS does include only those cases which have at least one value to count in the MRSet variables. In our analysis we want to have all cases - either with an answer (coded as 1) or with no ansswer (coded as 0) - in the results. Is there a switch that I can use to tell SPSS that I would like to base the percents on all cases? Best regards Georg |
|
|
Have a new Yahoo! Mail account? Kick start your journey by importing all your contacts! |
Re: Recoding Variables
|
|
Hi Eins,
This will do it (but you'll have to care about
missing values - what happens if there is missing values for X, Y or
Z?):
compute #max = max(x,y,z).
compute #min = min(x,y,z). do if x = #max and x = #min. compute w = 0. else if x = #max. compute w = 3. else if x =#min. compute w = 1. else. compute w = 2. end if. EXECUTE .
|
|
|
In reply to this post by Georg Maubach-2
I always create one more field to add to my MR Set. This
one is labeled “None Selected” and used just like the other fields. For example, if I have a set of dichotomous fields called
q03_01 q03_02 q03_03 q03_04 q03_05, I create q03none to add to it. NUMERIC q03none (F1.0) . RECODE q03_01 TO q03_05 q03none (MISSING=0). IF(NOT ANY(1, q03_01 TO q03_05))q03none=1. Note: recoding the missing values to zero works well for
the analysis I’m doing, it may not work for your purpose. Then my MR set looks like this: ** Dichotomies (0/1). MRSETS /MDGROUP NAME=$MRq03 LABEL ='What aspects do you
like best? (Mark all that apply)' VARIABLES= q03_01 TO q03_05 q03none VALUE=1 /DISPLAY NAME=[$MRq03]. This forces everyone into the table I run: CTABLES /FORMAT EMPTY=ZERO MISSING='NA' /SMISSING
VARIABLE /VLABELS VARIABLES=$MRq03 DISPLAY=DEFAULT /TABLE $MRq03 [COUNT COMMA40.0,
COLPCT.RESPONSES.COUNT PCT40.1, RESPONSES COMMA40.0, COLPCT.RESPONSES PCT40.1] /CATEGORIES VARIABLES=$MRq03 ORDER=D KEY=COUNT
EMPTY=INCLUDE TOTAL=YES POSITION=AFTER /TITLES TITLE= ')TABLE' CAPTION= '* The Count and Column Response %
(Base: Count) columns may sum to more than their ' + 'totals due to multiple responses.' . From: SPSSX(r)
Discussion [mailto:[hidden email]] On
Behalf Of Georg Maubach Hi
All when
using MRSet with SPSS Statistics 17 we found out that SPSS does include only
those cases which have at least one value to count in the MRSet variables. In
our analysis we want to have all cases - either with an answer (coded as 1) or
with no ansswer (coded as 0) - in the results. Is
there a switch that I can use to tell SPSS that I would like to base the
percents on all cases? Best
regards Georg
____________ DefenderMX2. |
|
|
Hi Heidi,
many thanks for your reply. I will do as you suggested
to have a work around.
Many thanks again.
Best regards
Georg
Von: Heidi Green [mailto:[hidden email]] Gesendet: Donnerstag, 19. November 2009 17:37 An: Maubach, Georg MRSC; [hidden email] Betreff: RE: Analysis of MRSets I always create one more field to
add to my MR Set. This one is labeled “None Selected” and used just like the
other fields. For example, if I have a set of
dichotomous fields called q03_01 q03_02 q03_03 q03_04 q03_05, I create q03none
to add to it. NUMERIC q03none (F1.0)
. RECODE q03_01 TO q03_05 q03none
(MISSING=0). IF(NOT ANY(1, q03_01 TO
q03_05))q03none=1. Note: recoding the missing values
to zero works well for the analysis I’m doing, it may not work for your
purpose. Then my MR set looks like
this: ** Dichotomies
(0/1). MRSETS /MDGROUP NAME=$MRq03 LABEL
='What aspects do you like best? (Mark all that
apply)' VARIABLES= q03_01 TO
q03_05 q03none
VALUE=1 /DISPLAY
NAME=[$MRq03]. This forces everyone into the
table I run: CTABLES /FORMAT EMPTY=ZERO
MISSING='NA' /SMISSING VARIABLE /VLABELS VARIABLES=$MRq03
DISPLAY=DEFAULT /TABLE $MRq03 [COUNT
COMMA40.0, COLPCT.RESPONSES.COUNT PCT40.1, RESPONSES COMMA40.0, COLPCT.RESPONSES
PCT40.1] /CATEGORIES
VARIABLES=$MRq03 ORDER=D KEY=COUNT EMPTY=INCLUDE TOTAL=YES
POSITION=AFTER
/TITLES TITLE=
')TABLE' CAPTION=
'* The Count
and Column Response % (Base: Count) columns may sum to more than their '
+ 'totals due to
multiple responses.' . From: SPSSX(r)
Discussion [mailto:[hidden email]] On Behalf Of Georg Maubach Hi
All when using MRSet with SPSS
Statistics 17 we found out that SPSS does include only those cases which have at
least one value to count in the MRSet variables. In our analysis we want to have
all cases - either with an answer (coded as 1) or with no ansswer (coded as 0) -
in the results. Is
there a switch that I can use to tell SPSS that I would like to base the
percents on all cases? Best regards
Georg ____________ DefenderMX2. |
|
|
In reply to this post by Georg Maubach-2
Hi Georg,
the easiest way to get all cases as the percetage base is to use such summary statistics as Column Total N %, Row Total N % etc. For example: data list free / sex v1 to v3. begin data. 1 1 1 1 1 0 0 1 1 0 1 1 1 0 0 0 1 0 0 0 2 1 1 0 2 1 0 0 2 1 1 1 2 0 0 0 2 0 0 0 end data. variable level sex (nominal). * Define Multiple Response Sets. MRSETS /MDGROUP NAME=$V CATEGORYLABELS=VARLABELS VARIABLES=v1 v2 v3 VALUE=1 /DISPLAY NAME=[$V]. * Custom Tables. CTABLES /VLABELS VARIABLES=sex $V DISPLAY=DEFAULT /TABLE $V [COUNT F40.0, COLPCT.COUNT PCT40.1, COLPCT.TOTALN PCT40.1] BY sex /CATEGORIES VARIABLES=sex ORDER=A KEY=VALUE EMPTY=EXCLUDE /CATEGORIES VARIABLES=$V EMPTY=INCLUDE TOTAL=YES POSITION=AFTER. HTH, Dr. Alexander Vinogradov, Associate Professor Sociology and Psychology Faculty National Taras Shevchenko University Ukraine > Hi All > when using MRSet with SPSS Statistics 17 we found out that SPSS does include only those cases which have at least one value to count in the MRSet variables. In our analysis we want to have all cases - either with an answer (coded as 1) or with no ansswer (coded as 0) - in the results. > Is there a switch that I can use to tell SPSS that I would like to base the percents on all cases? > Best regards > Georg ===================== 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 |
| Free forum by Nabble | Edit this page |