Ranking categories in mult response
Locked
Ranking categories in mult response
 
|
The 2008 European Values Survey has: Q6 On this list are various groups of people. Could you please sort out any that you would not like to have as neighbours mult resp groups noneighb2 'Not want as neighbours' (v46 to v60 (1)) /freq noneighb2. [UK respondents only: N = 1561]
Is there any way (analogous to /format dfreq) to display the table with categories listed by frequency mentioned)? John F Hall (Mr) [Retired academic survey researcher] SPSS 23: Windows 7 Pro Email: [hidden email] Website: www.surveyresearch.weebly.com SPSS start page: www.surveyresearch.weebly.com/1-survey-analysis-workshop |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Ranking categories in mult response
|
|
The 2008 European Values Survey has:
Q6 On this list are various groups of people. Could you please sort out any that you would not like to have as neighbours
mult resp groups noneighb2 'Not want as neighbours' (v46 to v60 (1)) /freq noneighb2.
[UK respondents only: N = 1561]
Is there any way (analogous to /format dfreq) to display the table with categories listed by frequency mentioned)?
John F Hall (Mr) [Retired academic survey researcher] SPSS 23: Windows 7 Pro
Email: [hidden email] Website: www.surveyresearch.weebly.com SPSS start page: www.surveyresearch.weebly.com/1-survey-analysis-workshop
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Re: Ranking categories in mult response
Administrator
|
In reply to this post by John F Hall
John asked: "Is there any way (analogous to /format dfreq) to display the table with categories listed by frequency mentioned)?"
John, you could always write the table to a new dataset via OMS, then use SORT CASES in that new dataset, and SUMMARIZE to display the table. This is more cumbersome than having a /FORMAT option, but it should work. HTH.
--
Bruce Weaver bweaver@lakeheadu.ca http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." PLEASE NOTE THE FOLLOWING: 1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. 2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (https://listserv.uga.edu/). |
Re: Ranking categories in mult response
Administrator
|
In reply to this post by John F Hall
John, your list excluded politicians and pedophiles!
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me. --- "Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis." Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?" |
|
In reply to this post by John F Hall
Why not use CTABLES for this? It allows you to sort mult response categories (as well as others) by set order, count, label or value. On Mon, Jan 18, 2016 at 5:08 AM, John F Hall <[hidden email]> wrote:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Re: Ranking categories in mult response
|
|
Jon, Bruce Default CTABLES syntax CTABLES /VLABELS VARIABLES=v46 v47 v48 v49 v50 v51 v52 v53 v54 v55 v56 v57 v58 v59 v60 DISPLAY=DEFAULT /TABLE v46 [COUNT F40.0] + v47 [COUNT F40.0] + v48 [COUNT F40.0] + v49 [COUNT F40.0] + v50 [COUNT F40.0] + v51 [COUNT F40.0] + v52 [COUNT F40.0] + v53 [COUNT F40.0] + v54 [COUNT F40.0] + v55 [COUNT F40.0] + v56 [COUNT F40.0] + v57 [COUNT F40.0] + v58 [COUNT F40.0] + v59 [COUNT F40.0] + v60 [COUNT F40.0] /CATEGORIES VARIABLES=v46 v47 v48 v49 v50 v51 v52 v53 v54 v55 v56 v57 v58 v59 v60 ORDER=d KEY=VALUE EMPTY=INCLUDE MISSING=EXCLUDE. Produces
All I want is value ‘1’ tabulated. Presumably I can substitute PCT for COUNT, but can I get both in the same table? John F Hall (Mr) [Retired academic survey researcher] Email: [hidden email] Website: www.surveyresearch.weebly.com SPSS start page: www.surveyresearch.weebly.com/1-survey-analysis-workshop PS (for David) It’s not my questionnaire. It was done by Gallup (35% UK response rate [sic !!] on what by 2012 had become the European Quality of Life Survey). Should they also add “Donald Trump” to their list of undesirable neighbours? From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Jon Peck Why not use CTABLES for this? It allows you to sort mult response categories (as well as others) by set order, count, label or value. On Mon, Jan 18, 2016 at 5:08 AM, John F Hall <[hidden email]> wrote:
-- Jon K Peck ===================== 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 |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
You can list as many statistics as you like along with COUNT in the syntax above, and you can factor that out like this (v46+v47+ ...)[count ...]. However, I thought this was supposed to be a mult response set, but it looks like you didn't define that. Use Analyze > Custom Tables > Multiple Response Sets to create it and then you can just select the set in the CTABLES canvas and select appropriate statistics. Note that there are two kind of these sets - on for a collection of dichotomies and one for lists of items. I think the multiple dichotomies is probably what you want. That will then tabulate just the counts or responses as needed. On Tue, Jan 19, 2016 at 11:20 AM, John F Hall <[hidden email]> wrote:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Re: Ranking categories in mult response
|
|
Jon Yes you are right: the variables effectively constitute a multiple dichotomy set. For the last forty years or so I have always used MULT RESP and stored the group commands in separate files for future use. Would you believe that I have never actually used MRSETS? * Define Multiple Response Sets. MRSETS /MDGROUP NAME=$Noneighb LABEL="Don't want as neighbours" CATEGORYLABELS=VARLABELS VARIABLES=v46 v47 v48 v49 v50 v51 v52 v53 v54 v55 v56 v57 v58 v59 v60 VALUE=1 /DISPLAY NAME=[$Noneighb]. After much perusal of the FM, I eventually found the answer in a table (ranking provenance of cars) and associated syntax on p472. My modified version worked first time: CTABLES /VLABELS VARIABLES=$Noneighb DISPLAY=DEFAULT /TABLE $Noneighb [COLPCT.COUNT COLPCT.RESPONSES] /CATEGORIES VARIABLES=$Noneighb ORDER=d KEY=COUNT EMPTY=INCLUDE MISSING=EXCLUDE. [NB: On the last line COUNT remains black: it does not display in yellow/orange in the syntax editor]
. . which is exactly what I wanted in category order. The appalling variable labels supplied by EVS/GESIS are a different matter , and clearly need the Peck Python treatment to capitalise the first letters. One alternative would perhaps be to recode v46 to v60 value 1 to 1 – 15 into a new variable set r46 to r60 with a new set of value labels. The table itself needs the Jignesh treatment to get rid of the clutter of % signs, but both the above can be partly resolved by manually editing the table with Ctrl+H.
John F Hall (Mr) [Retired academic survey researcher] Email: [hidden email] Website: www.surveyresearch.weebly.com SPSS start page: www.surveyresearch.weebly.com/1-survey-analysis-workshop PS To David’s politicians and paedophiles, I might consider adding Palin and Trump! From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Jon Peck You can list as many statistics as you like along with COUNT in the syntax above, and you can factor that out like this (v46+v47+ ...)[count ...]. However, I thought this was supposed to be a mult response set, but it looks like you didn't define that. Use Analyze > Custom Tables > Multiple Response Sets to create it and then you can just select the set in the CTABLES canvas and select appropriate statistics. Note that there are two kind of these sets - on for a collection of dichotomies and one for lists of items. I think the multiple dichotomies is probably what you want. That will then tabulate just the counts or responses as needed. On Tue, Jan 19, 2016 at 11:20 AM, John F Hall <[hidden email]> wrote:
-- Jon K Peck ===================== 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 |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
Note that once you have defined the MR sets, they will be saved with the data, so they will be automatically available when you use the data file later. These definitions can also be used with GGRAPH, but they cannot be used with the old MULT RESPONSE command. If you don't want the percent signs in the table body, just specify a different format for the statistics in the TABLE subcommand. On Wed, Jan 20, 2016 at 3:04 AM, John F Hall <[hidden email]> wrote:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Re: Ranking categories in mult response
|
Administrator
|
In reply to this post by John F Hall
John, thanks for posting that example. This is far more succinct and elegant than the OMS approach I suggested earlier.
--
Bruce Weaver bweaver@lakeheadu.ca http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." PLEASE NOTE THE FOLLOWING: 1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. 2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (https://listserv.uga.edu/). |
| Free forum by Nabble | Edit this page |