New tutorials on MULT RESPONSE (in syntax) {and a new page on Mark Abrams]
Locked
New tutorials on MULT RESPONSE (in syntax) {and a new page on Mark Abrams]
 
|
I’ve just posted two extensive new tutorials for MULT RESPONSE using SPSS 19, but the exercises will work with all releases back to SPSS 11. They use syntax rather than the GUI for basic operations such as data capture, dictionary specification and tabulation. They proceed gently, step-by-step, with full colour SPSS 19 screenshots at each step, and use real data from a wave of the British Social Attitudes (BSA) series. They relate to earlier tutorials explaining the use of multiple response questions in surveys, and specifically to different ways they are used and coded in the BSA. Two more are in preparation, but the available material is all on page: http://surveyresearch.weebly.com/33-multiple-response-mult-response.html Slightly off-topic, but I’ve also just posted a new page on my old boss Dr Mark Abrams (1906-1994) who is widely accepted as the father of survey and market research (and much else) in the UK. The page contains a brief profile, details of publications, but most important, the transcript of a fascinating 1984 interview in which describes in fascinating detail his early childhood growing up in an émigré Jewish family in North London, his student days at Latymer School and LSE, his early political education and contacts, his research career from the 1930s onwards (including time in the USA) and his work for the Labour Party. Many of the people he refers to, he both knew and worked with: they are household names in the pantheon of the social sciences. The page is: http://surveyresearch.weebly.com/mark-abrams.html . . and should be of interest to all social scientists, not just researchers. If any of you are on the SRMSNET or other lists, please forward this to them. John Hall Email: [hidden email] Website: www.surveyresearch.weebly.com Skype: surveyresearcher1 Phone: (+33) (0) 2.33.45.91.47 |
Re: New tutorials on MULT RESPONSE (in syntax) {and a new page on Mark Abrams]
|
|
I’m working from my 1992 course using spss-x 8 on a Vax for which I already have working examples: I’m simply converting and upgrading them for 19. The tutorials are aimed at absolute beginners and use data in ASCII format from 1986 (same format as many surveys deposited in archives). They are more about process and logic than the mechanics of SPSS, but I’ll check out CTABLES for future reference. If it’s not too complicated I might do something on that, but I still have far too much to do to finish other SPSS stuff. I can’t even replicate the exercises in 15 since IBM/SPSS don’t support it any more and I can’t get a renewal code for my CD. Hopefully they’ll leave 19 up long enough to maintain screenshot comparability. Email: [hidden email] Website: www.surveyresearch.weebly.com Skype: surveyresearcher1 Phone: (+33) (0) 2.33.45.91.47 From: Jon K Peck [mailto:[hidden email]] If this works for V11, then you are not taking advantage of the much superior multiple response capabilities that go with CTABLES. That seems to be doing a disservice to users with mult response problems. I hope you have at least made it clear that in versions since 11.5 there is better technology available, albeit in an option. And, furthermore, that MULT RESPONSE set definitions have nothing to do with the types of mult response sets used in CTABLES and GGRAPH and vice versa. And GGRAPH mult response features are available in the Base product.
|
Re: New tutorials on MULT RESPONSE (in syntax)
|
|
OK Jon (with an h) Here’s the original syntax from the first part of tutorial 3.3.3.1 (if you can be bothered to read it): [the pathway for the raw data file will be different for each user who downloads bsa86.txt from: http://surveyresearch.weebly.com/uploads/2/9/9/8/2998485/bsa86.txt ] I’ve added the comments later. title '3.3.3.1 BSA 1986: Q.117 Educational qualifications' . data list file 'C:\Users\John\Desktop\mybsa\mybsa86\bsa86.txt' records 23 /15 v1565 65 v1566 v1568 v1570 v1572 v1574 v1576 v1578 66-79 /16 v1608 v1610 v1612 v1614 v1616 v1618 v1620 v1622 v1624 8-25 . formats v1565 to v1624 (f2.0) . COMMENT Not available via GUI . list v1565 to v1624 /cases 10 . COMMENT Check data first . frequencies v1566 to v1624 . mult response groups = Q.117b 'Qualifications' (v1566 to v1622 (1,15)) /frequencies q.117b . To get
Variable and value labels can be added either by syntax (my preference) or GUI (which most kids probably prefer today, but is incredibly tedious) variable labels v1565 'Q.117a: Any qualifications?' / v1566 'Q.117b: CSE 2-5' / v1568 'Q.117b: CSE 1, O level,etc' / v1570 'Q.117b: A level, etc' / v1572 'Q.117b: Overseas leavng exam' / v1574 'Q.117b: Trade apprenticeship' / v1576 'Q.117b: RSA,similar clerical' / v1578 'Q.117b: C & Guilds Craft,etc' / v1608 'Q.117b: C & Guilds Advanced,etc' / v1610 'Q.117b: C & Guilds full tech' / v1612 'Q.117b: BEC Ordnary,ONC,OND' / v1614 'Q.117b: BEC Higher,HNC,HND' / v1616 'Q.117b: Teacher training' / v1618 'Q.117b: Nursing qual' / v1620 'Q.117b: Other tech,business' / v1622 'Q.117b: Uni v,CNAA degree,diploma' / v1624 'Q.117b: Other qual or DK,NA' . value labels v1565 1 'Yes' 2 'No' / v1566 to v1624 1 'CSE 2-5' 2 'CSE 1, O level,etc' 3 'A level, etc' 4 'Overseas leavng exam' 5 'Trade apprenticeship' 6 'RSA, similar clerical' 7 'City & Guilds Craft, etc' 8 'City & Guilds Advanced, etc' 9 'City & Guilds full tech' 10 'BEC Ordnary, ONC, OND' 11 'BEC Higher, HNC, HND' 12 'Teacher training' 13 'Nursing qual' 14 'Other tech, business' 15 'Univ, CNAA degree, diploma' 97 'Other qual or DK, NA' . and then run mult response again. mult response groups = Q.117b 'Qualifications' (v1566 to v1622 (1,15)) /frequencies q.117b .
A shortcut is to use a pre-existing saved file (but it has no value labels): http://surveyresearch.weebly.com/uploads/2/9/9/8/2998485/bsa86.sav . . and start from mult response above Method 1 requires: 1: Analyze > Tables > Multiple response sets 2: Scroll down to find variables (luckily adjacent positional not alphabetic mnemonic) 3: Click blue button to transfer to Variables in set box [defaults to dichotomies and says Sets defined here are not available in the Multiple Response Frequencies and Crosstabs procedure. Students have now reached a dead end and go for coffee etc.] Method 2 (yours) rewuires 1: Analyze > Tables > Custom Tables [box with unexpected message advising nominal vars] 2: Continue anyway 3: Scroll down to find variables (luckily adjacent positional not alphabetic mnemonic) 4: Highlight and drag to Rows 5: Bit frightening here as no explanation, but fart about looking for summary statistics (students now yawning or fidgeting) and give up. But settle for COUNT 6: Click PASTE just in case CTABLES /VLABELS VARIABLES=v1565 v1566 v1568 v1570 v1572 v1574 v1576 v1578 v1608 v1610 v1612 v1614 v1616 v1618 v1620 v1622 v1624 DISPLAY=DEFAULT /TABLE v1565 [S][MEAN, COUNT F40.0] + v1566 [S][MEAN, COUNT F40.0] + v1568 [S][MEAN, COUNT F40.0] + v1570 [S][MEAN, COUNT F40.0] + v1572 [S][MEAN, COUNT F40.0] + v1574 [S][MEAN, COUNT F40.0] + v1576 [S][MEAN, COUNT F40.0] + v1578 [S][MEAN, COUNT F40.0] + v1608 [S][MEAN, COUNT F40.0] + v1610 [S][MEAN, COUNT F40.0] + v1612 [S][MEAN, COUNT F40.0] + v1614 [S][MEAN, COUNT F40.0] + v1616 [S][MEAN, COUNT F40.0] + v1618 [S][MEAN, COUNT F40.0] + v1620 [S][MEAN, COUNT F40.0] + v1622 [S][MEAN, COUNT F40.0] + v1624 [S][MEAN, COUNT F40.0]. 7 [If you can find where it went, click on Run > All] or at step 6 above, press OK and get a table. (need to edit to drag column across)
This has taken several confusing minutes and most students will have lost interest, not just in SPSS but also in the question they were trying to analyse. My table is more useful than yours. My way is quicker, easier to create, follow and understand, and always will be! My MULT RESPONSE command can be saved and used again. Thank God for (obsolete SPSS) syntax and for researchers who are more interested in research questions than the innards of software. Stick to engineering and leave the research to us. John (with an h) Hall Email: [hidden email] Website: www.surveyresearch.weebly.com Skype: surveyresearcher1 Phone: (+33) (0) 2.33.45.91.47 I’m busy with other things for now, but From: Jon K Peck [mailto:[hidden email]] For absolute beginners, the newer technology would be easier.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Re: New tutorials on MULT RESPONSE (in syntax)
|
John I use the new non-antiquated approach to multiple response sets myself, and I don’t find it to be at all as you have described. I find it much easier
than your own approach. MRSETS /MDGROUP NAME=$VarA LABEL=' ' CATEGORYLABELS=VARLABELS VARIABLES=q14a q14b q14c q14d q14e q14f q14g q14h VALUE=1. CTABLES /VLABELS VARIABLES=$VarA DISPLAY=DEFAULT /TABLE $q14SATA [C][COUNT 'N' F40.0, COLPCT.COUNT '%' PCT40.0] /CATEGORIES VARIABLES=$q14SATA EMPTY=INCLUDE /TITLES TITLE='VarA. INSERT TITLE HERE'. Matthew J Poes Research Data Specialist Center for Prevention Research and Development University of Illinois 510 Devonshire Dr. Champaign, IL 61820 Phone: 217-265-4576 email:
[hidden email] From: SPSSX(r) Discussion [mailto:[hidden email]]
On Behalf Of John F Hall OK Jon (with an h) Here’s the original syntax from the first part of tutorial 3.3.3.1 (if you can be bothered to read it): [the pathway for the raw data file will be different for each user who downloads bsa86.txt from: http://surveyresearch.weebly.com/uploads/2/9/9/8/2998485/bsa86.txt
] I’ve added the comments later. title '3.3.3.1 BSA 1986: Q.117 Educational qualifications' . data list file 'C:\Users\John\Desktop\mybsa\mybsa86\bsa86.txt'
records 23 /15 v1565 65 v1566 v1568 v1570 v1572 v1574 v1576 v1578 66-79 /16 v1608 v1610 v1612 v1614 v1616 v1618 v1620 v1622 v1624 8-25 . formats v1565 to v1624 (f2.0) . COMMENT Not available via GUI . list v1565 to v1624 /cases 10 . COMMENT Check data first . frequencies v1566 to v1624 . mult response groups = Q.117b 'Qualifications' (v1566 to v1622 (1,15)) /frequencies q.117b . To get
Variable and value labels can be added either by syntax (my preference) or GUI (which most kids probably prefer today, but is incredibly tedious)
variable labels v1565 'Q.117a: Any qualifications?' / v1566 'Q.117b: CSE 2-5' / v1568 'Q.117b: CSE 1, O level,etc' / v1570 'Q.117b: A level, etc' / v1572 'Q.117b: Overseas leavng exam' / v1574 'Q.117b: Trade apprenticeship' / v1576 'Q.117b: RSA,similar clerical' / v1578 'Q.117b: C & Guilds Craft,etc' / v1608 'Q.117b: C & Guilds Advanced,etc' / v1610 'Q.117b: C & Guilds full tech' / v1612 'Q.117b: BEC Ordnary,ONC,OND' / v1614 'Q.117b: BEC Higher,HNC,HND' / v1616 'Q.117b: Teacher training' / v1618 'Q.117b: Nursing qual' / v1620 'Q.117b: Other tech,business' / v1622 'Q.117b: Uni v,CNAA degree,diploma' / v1624 'Q.117b: Other qual or DK,NA' . value labels v1565 1 'Yes' 2 'No' / v1566 to v1624 1 'CSE 2-5' 2 'CSE 1, O level,etc' 3 'A level, etc' 4 'Overseas leavng exam' 5 'Trade apprenticeship' 6 'RSA, similar clerical' 7 'City & Guilds Craft, etc' 8 'City & Guilds Advanced, etc' 9 'City & Guilds full tech' 10 'BEC Ordnary, ONC, OND' 11 'BEC Higher, HNC, HND' 12 'Teacher training' 13 'Nursing qual' 14 'Other tech, business' 15 'Univ, CNAA degree, diploma' 97 'Other qual or DK, NA' . and then run mult response again. mult response groups = Q.117b 'Qualifications' (v1566 to v1622 (1,15)) /frequencies q.117b .
A shortcut is to use a pre-existing saved file (but it has no value labels): http://surveyresearch.weebly.com/uploads/2/9/9/8/2998485/bsa86.sav
. . and start from mult response above Method 1 requires: 1: Analyze > Tables > Multiple response sets 2: Scroll down to find variables (luckily adjacent positional not alphabetic mnemonic) 3: Click blue button to transfer to Variables in set box
[defaults to dichotomies and says Sets defined here are not available in the Multiple Response Frequencies and Crosstabs
procedure. Students have now reached a dead end and go for coffee etc.] Method 2 (yours) rewuires 1: Analyze > Tables > Custom Tables [box with unexpected message advising nominal vars] 2: Continue anyway 3: Scroll down to find variables (luckily adjacent positional not alphabetic mnemonic) 4: Highlight and drag to Rows 5: Bit frightening here as no explanation, but fart about looking for summary statistics (students now yawning or fidgeting) and give up. But
settle for COUNT 6: Click PASTE just in case
CTABLES /VLABELS VARIABLES=v1565 v1566 v1568 v1570 v1572 v1574 v1576 v1578 v1608 v1610 v1612 v1614 v1616 v1618 v1620 v1622 v1624 DISPLAY=DEFAULT /TABLE v1565 [S][MEAN, COUNT F40.0] + v1566 [S][MEAN, COUNT F40.0] + v1568 [S][MEAN, COUNT F40.0] + v1570 [S][MEAN, COUNT F40.0]
+ v1572 [S][MEAN, COUNT F40.0] + v1574 [S][MEAN, COUNT F40.0] + v1576 [S][MEAN, COUNT F40.0] + v1578 [S][MEAN, COUNT F40.0] + v1608 [S][MEAN, COUNT F40.0] + v1610 [S][MEAN, COUNT F40.0] + v1612 [S][MEAN, COUNT F40.0] + v1614 [S][MEAN, COUNT F40.0] + v1616
[S][MEAN, COUNT F40.0] + v1618 [S][MEAN, COUNT F40.0] + v1620 [S][MEAN, COUNT F40.0] + v1622 [S][MEAN, COUNT F40.0] + v1624 [S][MEAN, COUNT F40.0].
7 [If you can find where it went, click on Run > All] or at step 6 above, press OK and get a table. (need to edit to drag column across)
This has taken several confusing minutes and most students will have lost interest, not just in SPSS but also in the question they were trying
to analyse. My table is more useful than yours. My way is quicker, easier to create, follow and understand, and always will be! My MULT RESPONSE command can be saved and used again. Thank God for (obsolete SPSS) syntax and for researchers who are more
interested in research questions than the innards of software. Stick to engineering and leave the research to us. John (with an h) Hall Email: [hidden email]
Website:
www.surveyresearch.weebly.com Skype: surveyresearcher1 Phone: (+33) (0) 2.33.45.91.47 I’m busy with other things for now, but
From: Jon K Peck
[hidden email] For absolute beginners, the newer technology would be easier.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
Maybe the difference is that you either need the module CTABLES or not ?
Which costs some 900+ € here in France, if I remember well. F. Thomas On 07/03/2012 18:21, Poes, Matthew Joseph wrote: > John I use the new non-antiquated approach to multiple response sets > myself, and I don’t find it to be at all as you have described. I find > it much easier than your own approach. > > MRSETS > > /MDGROUP NAME=$VarA LABEL=' ' CATEGORYLABELS=VARLABELS VARIABLES=q14a > q14b q14c q14d q14e q14f q14g q14h VALUE=1. > > CTABLES > > /VLABELS VARIABLES=$VarA DISPLAY=DEFAULT > > /TABLE $q14SATA [C][COUNT 'N' F40.0, COLPCT.COUNT '%' PCT40.0] > > /CATEGORIES VARIABLES=$q14SATA EMPTY=INCLUDE > > /TITLES > > TITLE='VarA. INSERT TITLE HERE'. > > *Matthew J Poes* > > Research Data Specialist > > Center for Prevention Research and Development > > University of Illinois > > 510 Devonshire Dr. > > Champaign, IL 61820 > > Phone: 217-265-4576 > > email: [hidden email] <mailto:[hidden email]> > > *From:*SPSSX(r) Discussion [mailto:[hidden email]] *On Behalf > Of *John F Hall > *Sent:* Wednesday, March 07, 2012 11:13 AM > *To:* [hidden email] > *Subject:* Re: New tutorials on MULT RESPONSE (in syntax) > > OK Jon (with an h) > > Here’s the original syntax from the first part of tutorial 3.3.3.1 (if > you can be bothered to read it): > > [the pathway for the raw data file will be different for each user who > downloads bsa86.txt from: > > http://surveyresearch.weebly.com/uploads/2/9/9/8/2998485/bsa86.txt ] > I’ve added the comments later. > > ------------------------------------------------------------------------ > > title '3.3.3.1 BSA 1986: Q.117 Educational qualifications' . > > data list > > file 'C:\Users\John\Desktop\mybsa\mybsa86\bsa86.txt' > > records 23 > > /15 v1565 65 v1566 v1568 v1570 v1572 v1574 v1576 v1578 66-79 > > /16 v1608 v1610 v1612 v1614 v1616 v1618 v1620 v1622 v1624 8-25 . > > formats v1565 to v1624 (f2.0) . > > COMMENT Not available via GUI . > > list v1565 to v1624 /cases 10 . > > COMMENT Check data first . > > frequencies v1566 to v1624 . > > mult response groups = > > Q.117b 'Qualifications' > > (v1566 to v1622 (1,15)) > > /frequencies q.117b . > > To get > > *Q.117b Frequencies* > > > > Responses > > > > Percent of Cases > > N > > > > Percent > > Qualifications^a > > > > 1 > > > > 374 > > > > 10.9% > > > > 22.2% > > 2 > > > > 1073 > > > > 31.3% > > > > 63.6% > > 3 > > > > 426 > > > > 12.4% > > > > 25.2% > > 4 > > > > 26 > > > > .8% > > > > 1.5% > > 5 > > > > 192 > > > > 5.6% > > > > 11.4% > > 6 > > > > 257 > > > > 7.5% > > > > 15.2% > > 7 > > > > 204 > > > > 5.9% > > > > 12.1% > > 8 > > > > 101 > > > > 2.9% > > > > 6.0% > > 9 > > > > 47 > > > > 1.4% > > > > 2.8% > > 10 > > > > 76 > > > > 2.2% > > > > 4.5% > > 11 > > > > 69 > > > > 2.0% > > > > 4.1% > > 12 > > > > 95 > > > > 2.8% > > > > 5.6% > > 13 > > > > 98 > > > > 2.9% > > > > 5.8% > > 14 > > > > 185 > > > > 5.4% > > > > 11.0% > > 15 > > > > 207 > > > > 6.0% > > > > 12.3% > > Total > > > > 3430 > > > > 100.0% > > > > 203.2% > > a. Group > > ------------------------------------------------------------------------ > > Variable and value labels can be added either by syntax (my preference) > or GUI (which most kids probably prefer today, but is incredibly tedious) > > variable labels > > v1565 'Q.117a: Any qualifications?' > > / v1566 'Q.117b: CSE 2-5' > > / v1568 'Q.117b: CSE 1, O level,etc' > > / v1570 'Q.117b: A level, etc' > > / v1572 'Q.117b: Overseas leavng exam' > > / v1574 'Q.117b: Trade apprenticeship' > > / v1576 'Q.117b: RSA,similar clerical' > > / v1578 'Q.117b: C & Guilds Craft,etc' > > / v1608 'Q.117b: C & Guilds Advanced,etc' > > / v1610 'Q.117b: C & Guilds full tech' > > / v1612 'Q.117b: BEC Ordnary,ONC,OND' > > / v1614 'Q.117b: BEC Higher,HNC,HND' > > / v1616 'Q.117b: Teacher training' > > / v1618 'Q.117b: Nursing qual' > > / v1620 'Q.117b: Other tech,business' > > / v1622 'Q.117b: Uni v,CNAA degree,diploma' > > / v1624 'Q.117b: Other qual or DK,NA' . > > value labels > > v1565 > > 1 'Yes' 2 'No' > > / v1566 to v1624 > > 1 'CSE 2-5' > > 2 'CSE 1, O level,etc' > > 3 'A level, etc' > > 4 'Overseas leavng exam' > > 5 'Trade apprenticeship' > > 6 'RSA, similar clerical' > > 7 'City & Guilds Craft, etc' > > 8 'City & Guilds Advanced, etc' > > 9 'City & Guilds full tech' > > 10 'BEC Ordnary, ONC, OND' > > 11 'BEC Higher, HNC, HND' > > 12 'Teacher training' > > 13 'Nursing qual' > > 14 'Other tech, business' > > 15 'Univ, CNAA degree, diploma' > > 97 'Other qual or DK, NA' . > > and then run mult response again. > > mult response groups = > > Q.117b 'Qualifications' > > (v1566 to v1622 (1,15)) > > /frequencies q.117b . > > *Q.117b Frequencies* > > > > Responses > > > > Percent of Cases > > N > > > > Percent > > Qualifications^a > > > > CSE 2-5 > > > > 374 > > > > 10.9% > > > > 22.2% > > CSE 1, O level,etc > > > > 1073 > > > > 31.3% > > > > 63.6% > > A level, etc > > > > 426 > > > > 12.4% > > > > 25.2% > > Overseas leavng exam > > > > 26 > > > > .8% > > > > 1.5% > > Trade apprenticeship > > > > 192 > > > > 5.6% > > > > 11.4% > > RSA, similar clerical > > > > 257 > > > > 7.5% > > > > 15.2% > > City & Guilds Craft, etc > > > > 204 > > > > 5.9% > > > > 12.1% > > City & Guilds Advanced, etc > > > > 101 > > > > 2.9% > > > > 6.0% > > City & Guilds full tech > > > > 47 > > > > 1.4% > > > > 2.8% > > BEC Ordnary, ONC, OND > > > > 76 > > > > 2.2% > > > > 4.5% > > BEC Higher, HNC, HND > > > > 69 > > > > 2.0% > > > > 4.1% > > Teacher training > > > > 95 > > > > 2.8% > > > > 5.6% > > Nursing qual > > > > 98 > > > > 2.9% > > > > 5.8% > > Other tech, business > > > > 185 > > > > 5.4% > > > > 11.0% > > Univ, CNAA degree, diploma > > > > 207 > > > > 6.0% > > > > 12.3% > > Total > > > > 3430 > > > > 100.0% > > > > 203.2% > > a. Group > > ------------------------------------------------------------------------ > > A shortcut is to use a pre-existing saved file (but it has no value labels): > > http://surveyresearch.weebly.com/uploads/2/9/9/8/2998485/bsa86.sav > > . . and start from mult response above > > Method 1 requires: > > 1: Analyze > Tables > Multiple response sets > > 2: Scroll down to find variables (luckily adjacent positional not > alphabetic mnemonic) > > 3: Click blue button to transfer to Variables in set box > > [defaults to dichotomies and says Sets defined here are not available in > the Multiple Response Frequencies and Crosstabs procedure. Students have > now reached a dead end and go for coffee etc.] > > Method 2 (yours) rewuires > > 1: Analyze > Tables > Custom Tables > > [box with unexpected message advising nominal vars] > > 2: Continue anyway > > 3: Scroll down to find variables (luckily adjacent positional not > alphabetic mnemonic) > > 4: Highlight and drag to Rows > > 5: Bit frightening here as no explanation, but fart about looking for > summary statistics (students now yawning or fidgeting) and give up. But > settle for COUNT > > 6: Click PASTE just in case > > CTABLES > > /VLABELS VARIABLES=v1565 v1566 v1568 v1570 v1572 v1574 v1576 v1578 v1608 > v1610 v1612 v1614 v1616 v1618 v1620 v1622 v1624 > > DISPLAY=DEFAULT > > /TABLE v1565 [S][MEAN, COUNT F40.0] + v1566 [S][MEAN, COUNT F40.0] + > v1568 [S][MEAN, COUNT F40.0] + v1570 [S][MEAN, COUNT F40.0] + v1572 > [S][MEAN, COUNT F40.0] + v1574 [S][MEAN, COUNT F40.0] + v1576 [S][MEAN, > COUNT F40.0] + v1578 [S][MEAN, COUNT F40.0] > > + v1608 [S][MEAN, COUNT F40.0] + v1610 [S][MEAN, COUNT F40.0] + v1612 > [S][MEAN, COUNT F40.0] + v1614 [S][MEAN, COUNT F40.0] + v1616 [S][MEAN, > COUNT F40.0] + v1618 [S][MEAN, COUNT F40.0] + v1620 [S][MEAN, COUNT > F40.0] + v1622 [S][MEAN, COUNT F40.0] + v1624 > > [S][MEAN, COUNT F40.0]. > > 7 [If you can find where it went, click on Run > All] or at step 6 > above, press OK and get a table. (need to edit to drag column across) > > > > Mean > > > > Count > > A109B117A NO QUALIFICATION OBTAINED BY R > > > > 1 > > > > 3100 > > A109B117B QUAL OBTND: CSE 2-5 > > > > 1 > > > > 3100 > > A109B117B QUAL OBTND: CSE 1, O LEVEL,ETC > > > > 2 > > > > 3100 > > A109B117B QUAL OBTND: A LEVEL, ETC > > > > 3 > > > > 3100 > > A109B117B QUAL OBTND: O-SEAS LEAVNG EXAM > > > > 4 > > > > 3100 > > A109B117B QUAL OBTND:TRADE APPRENTICESHP > > > > 5 > > > > 3100 > > A109B117B QUAL OBTND: RSA,SIMLR CLERICAL > > > > 6 > > > > 3100 > > A109B117B QUAL OBTND: C & GLDS CRAFT,ETC > > > > 7 > > > > 3100 > > A109B117B QUAL OBTND:C & GLDS ADVNCD,ETC > > > > 8 > > > > 3100 > > A109B117B QUAL OBTND: C & GLDS FULL TECH > > > > 9 > > > > 3100 > > A109B117B QUAL OBTND:BEC ORDNARY,ONC,OND > > > > 10 > > > > 3100 > > A109B117B QUAL OBTND: BEC HIGHER,HNC,HND > > > > 11 > > > > 3100 > > A109B117B QUAL OBTND: TEACHER TRAIN QUAL > > > > 12 > > > > 3100 > > A109B117B QUAL OBTND: NURSING QUAL > > > > 13 > > > > 3100 > > A109B117B QUAL OBTND: OTH TECH,BSNS QUAL > > > > 14 > > > > 3100 > > A109B117B QUAL OBTND: UNIV,CNAA DEGREE,DIP > > > > 15 > > > > 3100 > > A109B117B QUAL OBTND: OTHER OR DK,NA > > > > 97 > > > > 3100 > > This has taken several confusing minutes and most students will have > lost interest, not just in SPSS but also in the question they were > trying to analyse. My table is more useful than yours. My way is > quicker, easier to create, follow and understand, and always will be! My > MULT RESPONSE command can be saved and used again. Thank God for > (obsolete SPSS) syntax and for researchers who are more interested in > research questions than the innards of software. Stick to engineering > and leave the research to us. > > John (with an h) Hall > > Email: [hidden email] <mailto:[hidden email]> > > Website: www.surveyresearch.weebly.com <http://surveyresearch.weebly.com/> > > Skype: surveyresearcher1 > > Phone: (+33) (0) 2.33.45.91.47 > > I’m busy with other things for now, but > > *From:*Jon K Peck [mailto:[hidden email]] > <mailto:[mailto:[hidden email]]> > *Sent:* 06 March 2012 23:02 > *To:* John F Hall > *Subject:* RE: [SPSSX-L] New tutorials on MULT RESPONSE (in syntax) {and > a new page on Mark Abrams] > > For absolute beginners, the newer technology would be easier. > > In general, though, I don't see why you would spend time updating > something that is obsolete, even if still functional, instead of > spending the time on the newer features. That may be more work, but it > isn't doing someone a favor to be leading them to an old solution when > there is now a better one. It is your judgment, of course, which way is > better, but I would hope that you would first make that judgment. > > Jon Peck (no "h") aka Kim > Senior Software Engineer, IBM > [hidden email] <mailto:[hidden email]> > new phone: 720-342-5621 > > > > > From: "John F Hall" <[hidden email] <mailto:[hidden email]>> > To: Jon K Peck/Chicago/IBM@IBMUS > Cc: <[hidden email] <mailto:[hidden email]>> > Date: 03/06/2012 02:35 PM > Subject: RE: [SPSSX-L] New tutorials on MULT RESPONSE (in syntax) {and a > new page on Mark Abrams] > > ------------------------------------------------------------------------ > > > > > I’m working from my 1992 course using spss-x 8 on a Vax for which I > already have working examples: I’m simply converting and upgrading them > for 19. The tutorials are aimed at absolute beginners and use data in > ASCII format from 1986 (same format as many surveys deposited in > archives). They are more about process and logic than the mechanics of > SPSS, but I’ll check out CTABLES for future reference. If it’s not too > complicated I might do something on that, but I still have far too much > to do to finish other SPSS stuff. I can’t even replicate the exercises > in 15 since IBM/SPSS don’t support it any more and I can’t get a renewal > code for my CD. Hopefully they’ll leave 19 up long enough to maintain > screenshot comparability. > > > Email: [hidden email] <mailto:[hidden email]> > Website: www.surveyresearch.weebly.com <http://surveyresearch.weebly.com/> > Skype: surveyresearcher1 > Phone: (+33) (0) 2.33.45.91.47 > > > > > *From:*Jon K Peck [mailto:[hidden email]] * > Sent:* 06 March 2012 19:09* > To:* John F Hall* > Subject:* Re: [SPSSX-L] New tutorials on MULT RESPONSE (in syntax) {and > a new page on Mark Abrams] > > If this works for V11, then you are not taking advantage of the much > superior multiple response capabilities that go with CTABLES. That seems > to be doing a disservice to users with mult response problems. I hope > you have at least made it clear that in versions since 11.5 there is > better technology available, albeit in an option. And, furthermore, that > MULT RESPONSE set definitions have nothing to do with the types of mult > response sets used in CTABLES and GGRAPH and vice versa. And GGRAPH mult > response features are available in the Base product. > > Jon Peck (no "h") aka Kim > Senior Software Engineer, IBM_ > [hidden email] <mailto:[hidden email]> > new phone: 720-342-5621 > > > > > From: John F Hall <[hidden email] <mailto:[hidden email]>> > To: [hidden email] <mailto:[hidden email]> > Date: 03/06/2012 11:02 AM > Subject: [SPSSX-L] New tutorials on MULT RESPONSE (in syntax) {and a new > page on Mark Abrams] > Sent by: "SPSSX(r) Discussion" <[hidden email] > <mailto:[hidden email]>> > > ------------------------------------------------------------------------ > > > > > > I’ve just posted two extensive new tutorials for MULT RESPONSE using > SPSS 19, but the exercises will work with all releases back to SPSS 11. > They use syntax rather than the GUI for basic operations such as data > capture, dictionary specification and tabulation. They proceed gently, > step-by-step, with full colour SPSS 19 screenshots at each step, and use > real data from a wave of the British Social Attitudes (BSA) series. They > relate to earlier tutorials explaining the use of multiple response > questions in surveys, and specifically to different ways they are used > and coded in the BSA. Two more are in preparation, but the available > material is all on page:_ > _http://surveyresearch.weebly.com/33-multiple-response-mult-response.html > > Slightly off-topic, but I’ve also just posted a new page on my old boss > Dr Mark Abrams (1906-1994) who is widely accepted as the father of > survey and market research (and much else) in the UK. The page contains > a brief profile, details of publications, but most important, the > transcript of a fascinating 1984 interview in which describes in > fascinating detail his early childhood growing up in an émigré Jewish > family in North London, his student days at Latymer School and LSE, his > early political education and contacts, his research career from the > 1930s onwards (including time in the USA) and his work for the Labour > Party. Many of the people he refers to, he both knew and worked with: > they are household names in the pantheon of the social sciences. > > The page is: > _ > _http://surveyresearch.weebly.com/mark-abrams.html > > . . and should be of interest to all social scientists, not just > researchers. If any of you are on the SRMSNET or other lists, please > forward this to them. > > John Hall > > > Email: [hidden email] <mailto:[hidden email]> > Website: www.surveyresearch.weebly.com <http://surveyresearch.weebly.com/> > Skype: surveyresearcher1 > Phone: (+33) (0) 2.33.45.91.47 > > > > ===================== 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 |
Re: New tutorials on MULT RESPONSE (in syntax)
|
|
Maybe. It's included in the program I get here at the University, so I didn't realize it was an add on.
Matthew J Poes Research Data Specialist Center for Prevention Research and Development University of Illinois 510 Devonshire Dr. Champaign, IL 61820 Phone:� 217-265-4576 email: [hidden email] -----Original Message----- From: ftr [mailto:[hidden email]] Sent: Tuesday, March 13, 2012 1:01 PM To: Poes, Matthew Joseph Cc: [hidden email] Subject: Re: [SPSSX-L] New tutorials on MULT RESPONSE (in syntax) Maybe the difference is that you either need the module CTABLES or not ? Which costs some 900+ € here in France, if I remember well. F. Thomas On 07/03/2012 18:21, Poes, Matthew Joseph wrote: > John I use the new non-antiquated approach to multiple response sets > myself, and I don’t find it to be at all as you have described. I find > it much easier than your own approach. > > MRSETS > > /MDGROUP NAME=$VarA LABEL=' ' CATEGORYLABELS=VARLABELS VARIABLES=q14a > q14b q14c q14d q14e q14f q14g q14h VALUE=1. > > CTABLES > > /VLABELS VARIABLES=$VarA DISPLAY=DEFAULT > > /TABLE $q14SATA [C][COUNT 'N' F40.0, COLPCT.COUNT '%' PCT40.0] > > /CATEGORIES VARIABLES=$q14SATA EMPTY=INCLUDE > > /TITLES > > TITLE='VarA. INSERT TITLE HERE'. > > *Matthew J Poes* > > Research Data Specialist > > Center for Prevention Research and Development > > University of Illinois > > 510 Devonshire Dr. > > Champaign, IL 61820 > > Phone: 217-265-4576 > > email: [hidden email] <mailto:[hidden email]> > > *From:*SPSSX(r) Discussion [mailto:[hidden email]] *On > Behalf Of *John F Hall > *Sent:* Wednesday, March 07, 2012 11:13 AM > *To:* [hidden email] > *Subject:* Re: New tutorials on MULT RESPONSE (in syntax) > > OK Jon (with an h) > > Here’s the original syntax from the first part of tutorial 3.3.3.1 (if > you can be bothered to read it): > > [the pathway for the raw data file will be different for each user who > downloads bsa86.txt from: > > http://surveyresearch.weebly.com/uploads/2/9/9/8/2998485/bsa86.txt ] > I’ve added the comments later. > > ---------------------------------------------------------------------- > -- > > title '3.3.3.1 BSA 1986: Q.117 Educational qualifications' . > > data list > > file 'C:\Users\John\Desktop\mybsa\mybsa86\bsa86.txt' > > records 23 > > /15 v1565 65 v1566 v1568 v1570 v1572 v1574 v1576 v1578 66-79 > > /16 v1608 v1610 v1612 v1614 v1616 v1618 v1620 v1622 v1624 8-25 . > > formats v1565 to v1624 (f2.0) . > > COMMENT Not available via GUI . > > list v1565 to v1624 /cases 10 . > > COMMENT Check data first . > > frequencies v1566 to v1624 . > > mult response groups = > > Q.117b 'Qualifications' > > (v1566 to v1622 (1,15)) > > /frequencies q.117b . > > To get > > *Q.117b Frequencies* > > > > Responses > > > > Percent of Cases > > N > > > > Percent > > Qualifications^a > > > > 1 > > > > 374 > > > > 10.9% > > > > 22.2% > > 2 > > > > 1073 > > > > 31.3% > > > > 63.6% > > 3 > > > > 426 > > > > 12.4% > > > > 25.2% > > 4 > > > > 26 > > > > .8% > > > > 1.5% > > 5 > > > > 192 > > > > 5.6% > > > > 11.4% > > 6 > > > > 257 > > > > 7.5% > > > > 15.2% > > 7 > > > > 204 > > > > 5.9% > > > > 12.1% > > 8 > > > > 101 > > > > 2.9% > > > > 6.0% > > 9 > > > > 47 > > > > 1.4% > > > > 2.8% > > 10 > > > > 76 > > > > 2.2% > > > > 4.5% > > 11 > > > > 69 > > > > 2.0% > > > > 4.1% > > 12 > > > > 95 > > > > 2.8% > > > > 5.6% > > 13 > > > > 98 > > > > 2.9% > > > > 5.8% > > 14 > > > > 185 > > > > 5.4% > > > > 11.0% > > 15 > > > > 207 > > > > 6.0% > > > > 12.3% > > Total > > > > 3430 > > > > 100.0% > > > > 203.2% > > a. Group > > ---------------------------------------------------------------------- > -- > > Variable and value labels can be added either by syntax (my > preference) or GUI (which most kids probably prefer today, but is > incredibly tedious) > > variable labels > > v1565 'Q.117a: Any qualifications?' > > / v1566 'Q.117b: CSE 2-5' > > / v1568 'Q.117b: CSE 1, O level,etc' > > / v1570 'Q.117b: A level, etc' > > / v1572 'Q.117b: Overseas leavng exam' > > / v1574 'Q.117b: Trade apprenticeship' > > / v1576 'Q.117b: RSA,similar clerical' > > / v1578 'Q.117b: C & Guilds Craft,etc' > > / v1608 'Q.117b: C & Guilds Advanced,etc' > > / v1610 'Q.117b: C & Guilds full tech' > > / v1612 'Q.117b: BEC Ordnary,ONC,OND' > > / v1614 'Q.117b: BEC Higher,HNC,HND' > > / v1616 'Q.117b: Teacher training' > > / v1618 'Q.117b: Nursing qual' > > / v1620 'Q.117b: Other tech,business' > > / v1622 'Q.117b: Uni v,CNAA degree,diploma' > > / v1624 'Q.117b: Other qual or DK,NA' . > > value labels > > v1565 > > 1 'Yes' 2 'No' > > / v1566 to v1624 > > 1 'CSE 2-5' > > 2 'CSE 1, O level,etc' > > 3 'A level, etc' > > 4 'Overseas leavng exam' > > 5 'Trade apprenticeship' > > 6 'RSA, similar clerical' > > 7 'City & Guilds Craft, etc' > > 8 'City & Guilds Advanced, etc' > > 9 'City & Guilds full tech' > > 10 'BEC Ordnary, ONC, OND' > > 11 'BEC Higher, HNC, HND' > > 12 'Teacher training' > > 13 'Nursing qual' > > 14 'Other tech, business' > > 15 'Univ, CNAA degree, diploma' > > 97 'Other qual or DK, NA' . > > and then run mult response again. > > mult response groups = > > Q.117b 'Qualifications' > > (v1566 to v1622 (1,15)) > > /frequencies q.117b . > > *Q.117b Frequencies* > > > > Responses > > > > Percent of Cases > > N > > > > Percent > > Qualifications^a > > > > CSE 2-5 > > > > 374 > > > > 10.9% > > > > 22.2% > > CSE 1, O level,etc > > > > 1073 > > > > 31.3% > > > > 63.6% > > A level, etc > > > > 426 > > > > 12.4% > > > > 25.2% > > Overseas leavng exam > > > > 26 > > > > .8% > > > > 1.5% > > Trade apprenticeship > > > > 192 > > > > 5.6% > > > > 11.4% > > RSA, similar clerical > > > > 257 > > > > 7.5% > > > > 15.2% > > City & Guilds Craft, etc > > > > 204 > > > > 5.9% > > > > 12.1% > > City & Guilds Advanced, etc > > > > 101 > > > > 2.9% > > > > 6.0% > > City & Guilds full tech > > > > 47 > > > > 1.4% > > > > 2.8% > > BEC Ordnary, ONC, OND > > > > 76 > > > > 2.2% > > > > 4.5% > > BEC Higher, HNC, HND > > > > 69 > > > > 2.0% > > > > 4.1% > > Teacher training > > > > 95 > > > > 2.8% > > > > 5.6% > > Nursing qual > > > > 98 > > > > 2.9% > > > > 5.8% > > Other tech, business > > > > 185 > > > > 5.4% > > > > 11.0% > > Univ, CNAA degree, diploma > > > > 207 > > > > 6.0% > > > > 12.3% > > Total > > > > 3430 > > > > 100.0% > > > > 203.2% > > a. Group > > ---------------------------------------------------------------------- > -- > > A shortcut is to use a pre-existing saved file (but it has no value labels): > > http://surveyresearch.weebly.com/uploads/2/9/9/8/2998485/bsa86.sav > > . . and start from mult response above > > Method 1 requires: > > 1: Analyze > Tables > Multiple response sets > > 2: Scroll down to find variables (luckily adjacent positional not > alphabetic mnemonic) > > 3: Click blue button to transfer to Variables in set box > > [defaults to dichotomies and says Sets defined here are not available > in the Multiple Response Frequencies and Crosstabs procedure. Students > have now reached a dead end and go for coffee etc.] > > Method 2 (yours) rewuires > > 1: Analyze > Tables > Custom Tables > > [box with unexpected message advising nominal vars] > > 2: Continue anyway > > 3: Scroll down to find variables (luckily adjacent positional not > alphabetic mnemonic) > > 4: Highlight and drag to Rows > > 5: Bit frightening here as no explanation, but fart about looking for > summary statistics (students now yawning or fidgeting) and give up. > But settle for COUNT > > 6: Click PASTE just in case > > CTABLES > > /VLABELS VARIABLES=v1565 v1566 v1568 v1570 v1572 v1574 v1576 v1578 > v1608 > v1610 v1612 v1614 v1616 v1618 v1620 v1622 v1624 > > DISPLAY=DEFAULT > > /TABLE v1565 [S][MEAN, COUNT F40.0] + v1566 [S][MEAN, COUNT F40.0] + > v1568 [S][MEAN, COUNT F40.0] + v1570 [S][MEAN, COUNT F40.0] + v1572 > [S][MEAN, COUNT F40.0] + v1574 [S][MEAN, COUNT F40.0] + v1576 > [S][MEAN, COUNT F40.0] + v1578 [S][MEAN, COUNT F40.0] > > + v1608 [S][MEAN, COUNT F40.0] + v1610 [S][MEAN, COUNT F40.0] + v1612 > [S][MEAN, COUNT F40.0] + v1614 [S][MEAN, COUNT F40.0] + v1616 > [S][MEAN, COUNT F40.0] + v1618 [S][MEAN, COUNT F40.0] + v1620 > [S][MEAN, COUNT F40.0] + v1622 [S][MEAN, COUNT F40.0] + v1624 > > [S][MEAN, COUNT F40.0]. > > 7 [If you can find where it went, click on Run > All] or at step 6 > above, press OK and get a table. (need to edit to drag column across) > > > > Mean > > > > Count > > A109B117A NO QUALIFICATION OBTAINED BY R > > > > 1 > > > > 3100 > > A109B117B QUAL OBTND: CSE 2-5 > > > > 1 > > > > 3100 > > A109B117B QUAL OBTND: CSE 1, O LEVEL,ETC > > > > 2 > > > > 3100 > > A109B117B QUAL OBTND: A LEVEL, ETC > > > > 3 > > > > 3100 > > A109B117B QUAL OBTND: O-SEAS LEAVNG EXAM > > > > 4 > > > > 3100 > > A109B117B QUAL OBTND:TRADE APPRENTICESHP > > > > 5 > > > > 3100 > > A109B117B QUAL OBTND: RSA,SIMLR CLERICAL > > > > 6 > > > > 3100 > > A109B117B QUAL OBTND: C & GLDS CRAFT,ETC > > > > 7 > > > > 3100 > > A109B117B QUAL OBTND:C & GLDS ADVNCD,ETC > > > > 8 > > > > 3100 > > A109B117B QUAL OBTND: C & GLDS FULL TECH > > > > 9 > > > > 3100 > > A109B117B QUAL OBTND:BEC ORDNARY,ONC,OND > > > > 10 > > > > 3100 > > A109B117B QUAL OBTND: BEC HIGHER,HNC,HND > > > > 11 > > > > 3100 > > A109B117B QUAL OBTND: TEACHER TRAIN QUAL > > > > 12 > > > > 3100 > > A109B117B QUAL OBTND: NURSING QUAL > > > > 13 > > > > 3100 > > A109B117B QUAL OBTND: OTH TECH,BSNS QUAL > > > > 14 > > > > 3100 > > A109B117B QUAL OBTND: UNIV,CNAA DEGREE,DIP > > > > 15 > > > > 3100 > > A109B117B QUAL OBTND: OTHER OR DK,NA > > > > 97 > > > > 3100 > > This has taken several confusing minutes and most students will have > lost interest, not just in SPSS but also in the question they were > trying to analyse. My table is more useful than yours. My way is > quicker, easier to create, follow and understand, and always will be! > My MULT RESPONSE command can be saved and used again. Thank God for > (obsolete SPSS) syntax and for researchers who are more interested in > research questions than the innards of software. Stick to engineering > and leave the research to us. > > John (with an h) Hall > > Email: [hidden email] <mailto:[hidden email]> > > Website: www.surveyresearch.weebly.com > <http://surveyresearch.weebly.com/> > > Skype: surveyresearcher1 > > Phone: (+33) (0) 2.33.45.91.47 > > I’m busy with other things for now, but > > *From:*Jon K Peck [mailto:[hidden email]] > <mailto:[mailto:[hidden email]]> > *Sent:* 06 March 2012 23:02 > *To:* John F Hall > *Subject:* RE: [SPSSX-L] New tutorials on MULT RESPONSE (in syntax) > {and a new page on Mark Abrams] > > For absolute beginners, the newer technology would be easier. > > In general, though, I don't see why you would spend time updating > something that is obsolete, even if still functional, instead of > spending the time on the newer features. That may be more work, but it > isn't doing someone a favor to be leading them to an old solution when > there is now a better one. It is your judgment, of course, which way > is better, but I would hope that you would first make that judgment. > > Jon Peck (no "h") aka Kim > Senior Software Engineer, IBM > [hidden email] <mailto:[hidden email]> new phone: 720-342-5621 > > > > > From: "John F Hall" <[hidden email] <mailto:[hidden email]>> > To: Jon K Peck/Chicago/IBM@IBMUS > Cc: <[hidden email] <mailto:[hidden email]>> > Date: 03/06/2012 02:35 PM > Subject: RE: [SPSSX-L] New tutorials on MULT RESPONSE (in syntax) {and a > new page on Mark Abrams] > > ------------------------------------------------------------------------ > > > > > I’m working from my 1992 course using spss-x 8 on a Vax for which I > already have working examples: I’m simply converting and upgrading them > for 19. The tutorials are aimed at absolute beginners and use data in > ASCII format from 1986 (same format as many surveys deposited in > archives). They are more about process and logic than the mechanics of > SPSS, but I’ll check out CTABLES for future reference. If it’s not too > complicated I might do something on that, but I still have far too much > to do to finish other SPSS stuff. I can’t even replicate the exercises > in 15 since IBM/SPSS don’t support it any more and I can’t get a renewal > code for my CD. Hopefully they’ll leave 19 up long enough to maintain > screenshot comparability. > > > Email: [hidden email] <mailto:[hidden email]> > Website: www.surveyresearch.weebly.com <http://surveyresearch.weebly.com/> > Skype: surveyresearcher1 > Phone: (+33) (0) 2.33.45.91.47 > > > > > *From:*Jon K Peck [mailto:[hidden email]] * > Sent:* 06 March 2012 19:09* > To:* John F Hall* > Subject:* Re: [SPSSX-L] New tutorials on MULT RESPONSE (in syntax) {and > a new page on Mark Abrams] > > If this works for V11, then you are not taking advantage of the much > superior multiple response capabilities that go with CTABLES. That seems > to be doing a disservice to users with mult response problems. I hope > you have at least made it clear that in versions since 11.5 there is > better technology available, albeit in an option. And, furthermore, that > MULT RESPONSE set definitions have nothing to do with the types of mult > response sets used in CTABLES and GGRAPH and vice versa. And GGRAPH mult > response features are available in the Base product. > > Jon Peck (no "h") aka Kim > Senior Software Engineer, IBM_ > [hidden email] <mailto:[hidden email]> > new phone: 720-342-5621 > > > > > From: John F Hall <[hidden email] <mailto:[hidden email]>> > To: [hidden email] <mailto:[hidden email]> > Date: 03/06/2012 11:02 AM > Subject: [SPSSX-L] New tutorials on MULT RESPONSE (in syntax) {and a new > page on Mark Abrams] > Sent by: "SPSSX(r) Discussion" <[hidden email] > <mailto:[hidden email]>> > > ------------------------------------------------------------------------ > > > > > > I’ve just posted two extensive new tutorials for MULT RESPONSE using > SPSS 19, but the exercises will work with all releases back to SPSS 11. > They use syntax rather than the GUI for basic operations such as data > capture, dictionary specification and tabulation. They proceed gently, > step-by-step, with full colour SPSS 19 screenshots at each step, and use > real data from a wave of the British Social Attitudes (BSA) series. They > relate to earlier tutorials explaining the use of multiple response > questions in surveys, and specifically to different ways they are used > and coded in the BSA. Two more are in preparation, but the available > material is all on page:_ > _http://surveyresearch.weebly.com/33-multiple-response-mult-response.html > > Slightly off-topic, but I’ve also just posted a new page on my old boss > Dr Mark Abrams (1906-1994) who is widely accepted as the father of > survey and market research (and much else) in the UK. The page contains > a brief profile, details of publications, but most important, the > transcript of a fascinating 1984 interview in which describes in > fascinating detail his early childhood growing up in an émigré Jewish > family in North London, his student days at Latymer School and LSE, his > early political education and contacts, his research career from the > 1930s onwards (including time in the USA) and his work for the Labour > Party. Many of the people he refers to, he both knew and worked with: > they are household names in the pantheon of the social sciences. > > The page is: > _ > _http://surveyresearch.weebly.com/mark-abrams.html > > . . and should be of interest to all social scientists, not just > researchers. If any of you are on the SRMSNET or other lists, please > forward this to them. > > John Hall > > > Email: [hidden email] <mailto:[hidden email]> > Website: www.surveyresearch.weebly.com <http://surveyresearch.weebly.com/> > Skype: surveyresearcher1 > Phone: (+33) (0) 2.33.45.91.47 > > > > ===================== 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 |
Re: New tutorials on MULT RESPONSE (in syntax)
|
|
In reply to this post by news
Precisely
I'm aiming at students and researchers who only have Gradpack or who have never heard of SPSS, but have survey data to process (a growing population now that academic institutions are looking for cheaper alternatives) for many of whom English is a second language. My tutorials should work with every release from 11 onwards and my syntax is much easier to understand and follow than anything created by the GUI. John Hall Email: [hidden email] Website: www.surveyresearch.weebly.com Skype: surveyresearcher1 Phone: (+33) (0) 2.33.45.91.47 -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of ftr Sent: 13 March 2012 19:01 To: [hidden email] Subject: Re: New tutorials on MULT RESPONSE (in syntax) Maybe the difference is that you either need the module CTABLES or not ? Which costs some 900+ € here in France, if I remember well. F. Thomas On 07/03/2012 18:21, Poes, Matthew Joseph wrote: > John I use the new non-antiquated approach to multiple response sets > myself, and I don’t find it to be at all as you have described. I find > it much easier than your own approach. > > MRSETS > > /MDGROUP NAME=$VarA LABEL=' ' CATEGORYLABELS=VARLABELS VARIABLES=q14a > q14b q14c q14d q14e q14f q14g q14h VALUE=1. > > CTABLES > > /VLABELS VARIABLES=$VarA DISPLAY=DEFAULT > > /TABLE $q14SATA [C][COUNT 'N' F40.0, COLPCT.COUNT '%' PCT40.0] > > /CATEGORIES VARIABLES=$q14SATA EMPTY=INCLUDE > > /TITLES > > TITLE='VarA. INSERT TITLE HERE'. > > *Matthew J Poes* > > Research Data Specialist > > Center for Prevention Research and Development > > University of Illinois > > 510 Devonshire Dr. > > Champaign, IL 61820 > > Phone: 217-265-4576 > > email: [hidden email] <mailto:[hidden email]> > > *From:*SPSSX(r) Discussion [mailto:[hidden email]] *On > Behalf Of *John F Hall > *Sent:* Wednesday, March 07, 2012 11:13 AM > *To:* [hidden email] > *Subject:* Re: New tutorials on MULT RESPONSE (in syntax) > > OK Jon (with an h) > > Here’s the original syntax from the first part of tutorial 3.3.3.1 (if > you can be bothered to read it): > > [the pathway for the raw data file will be different for each user who > downloads bsa86.txt from: > > http://surveyresearch.weebly.com/uploads/2/9/9/8/2998485/bsa86.txt ] > I’ve added the comments later. > > ---------------------------------------------------------------------- > -- > > title '3.3.3.1 BSA 1986: Q.117 Educational qualifications' . > > data list > > file 'C:\Users\John\Desktop\mybsa\mybsa86\bsa86.txt' > > records 23 > > /15 v1565 65 v1566 v1568 v1570 v1572 v1574 v1576 v1578 66-79 > > /16 v1608 v1610 v1612 v1614 v1616 v1618 v1620 v1622 v1624 8-25 . > > formats v1565 to v1624 (f2.0) . > > COMMENT Not available via GUI . > > list v1565 to v1624 /cases 10 . > > COMMENT Check data first . > > frequencies v1566 to v1624 . > > mult response groups = > > Q.117b 'Qualifications' > > (v1566 to v1622 (1,15)) > > /frequencies q.117b . > > To get > > *Q.117b Frequencies* > > > > Responses > > > > Percent of Cases > > N > > > > Percent > > Qualifications^a > > > > 1 > > > > 374 > > > > 10.9% > > > > 22.2% > > 2 > > > > 1073 > > > > 31.3% > > > > 63.6% > > 3 > > > > 426 > > > > 12.4% > > > > 25.2% > > 4 > > > > 26 > > > > .8% > > > > 1.5% > > 5 > > > > 192 > > > > 5.6% > > > > 11.4% > > 6 > > > > 257 > > > > 7.5% > > > > 15.2% > > 7 > > > > 204 > > > > 5.9% > > > > 12.1% > > 8 > > > > 101 > > > > 2.9% > > > > 6.0% > > 9 > > > > 47 > > > > 1.4% > > > > 2.8% > > 10 > > > > 76 > > > > 2.2% > > > > 4.5% > > 11 > > > > 69 > > > > 2.0% > > > > 4.1% > > 12 > > > > 95 > > > > 2.8% > > > > 5.6% > > 13 > > > > 98 > > > > 2.9% > > > > 5.8% > > 14 > > > > 185 > > > > 5.4% > > > > 11.0% > > 15 > > > > 207 > > > > 6.0% > > > > 12.3% > > Total > > > > 3430 > > > > 100.0% > > > > 203.2% > > a. Group > > ---------------------------------------------------------------------- > -- > > Variable and value labels can be added either by syntax (my > preference) or GUI (which most kids probably prefer today, but is > incredibly tedious) > > variable labels > > v1565 'Q.117a: Any qualifications?' > > / v1566 'Q.117b: CSE 2-5' > > / v1568 'Q.117b: CSE 1, O level,etc' > > / v1570 'Q.117b: A level, etc' > > / v1572 'Q.117b: Overseas leavng exam' > > / v1574 'Q.117b: Trade apprenticeship' > > / v1576 'Q.117b: RSA,similar clerical' > > / v1578 'Q.117b: C & Guilds Craft,etc' > > / v1608 'Q.117b: C & Guilds Advanced,etc' > > / v1610 'Q.117b: C & Guilds full tech' > > / v1612 'Q.117b: BEC Ordnary,ONC,OND' > > / v1614 'Q.117b: BEC Higher,HNC,HND' > > / v1616 'Q.117b: Teacher training' > > / v1618 'Q.117b: Nursing qual' > > / v1620 'Q.117b: Other tech,business' > > / v1622 'Q.117b: Uni v,CNAA degree,diploma' > > / v1624 'Q.117b: Other qual or DK,NA' . > > value labels > > v1565 > > 1 'Yes' 2 'No' > > / v1566 to v1624 > > 1 'CSE 2-5' > > 2 'CSE 1, O level,etc' > > 3 'A level, etc' > > 4 'Overseas leavng exam' > > 5 'Trade apprenticeship' > > 6 'RSA, similar clerical' > > 7 'City & Guilds Craft, etc' > > 8 'City & Guilds Advanced, etc' > > 9 'City & Guilds full tech' > > 10 'BEC Ordnary, ONC, OND' > > 11 'BEC Higher, HNC, HND' > > 12 'Teacher training' > > 13 'Nursing qual' > > 14 'Other tech, business' > > 15 'Univ, CNAA degree, diploma' > > 97 'Other qual or DK, NA' . > > and then run mult response again. > > mult response groups = > > Q.117b 'Qualifications' > > (v1566 to v1622 (1,15)) > > /frequencies q.117b . > > *Q.117b Frequencies* > > > > Responses > > > > Percent of Cases > > N > > > > Percent > > Qualifications^a > > > > CSE 2-5 > > > > 374 > > > > 10.9% > > > > 22.2% > > CSE 1, O level,etc > > > > 1073 > > > > 31.3% > > > > 63.6% > > A level, etc > > > > 426 > > > > 12.4% > > > > 25.2% > > Overseas leavng exam > > > > 26 > > > > .8% > > > > 1.5% > > Trade apprenticeship > > > > 192 > > > > 5.6% > > > > 11.4% > > RSA, similar clerical > > > > 257 > > > > 7.5% > > > > 15.2% > > City & Guilds Craft, etc > > > > 204 > > > > 5.9% > > > > 12.1% > > City & Guilds Advanced, etc > > > > 101 > > > > 2.9% > > > > 6.0% > > City & Guilds full tech > > > > 47 > > > > 1.4% > > > > 2.8% > > BEC Ordnary, ONC, OND > > > > 76 > > > > 2.2% > > > > 4.5% > > BEC Higher, HNC, HND > > > > 69 > > > > 2.0% > > > > 4.1% > > Teacher training > > > > 95 > > > > 2.8% > > > > 5.6% > > Nursing qual > > > > 98 > > > > 2.9% > > > > 5.8% > > Other tech, business > > > > 185 > > > > 5.4% > > > > 11.0% > > Univ, CNAA degree, diploma > > > > 207 > > > > 6.0% > > > > 12.3% > > Total > > > > 3430 > > > > 100.0% > > > > 203.2% > > a. Group > > ---------------------------------------------------------------------- > -- > > A shortcut is to use a pre-existing saved file (but it has no value labels): > > http://surveyresearch.weebly.com/uploads/2/9/9/8/2998485/bsa86.sav > > . . and start from mult response above > > Method 1 requires: > > 1: Analyze > Tables > Multiple response sets > > 2: Scroll down to find variables (luckily adjacent positional not > alphabetic mnemonic) > > 3: Click blue button to transfer to Variables in set box > > [defaults to dichotomies and says Sets defined here are not available > in the Multiple Response Frequencies and Crosstabs procedure. Students > have now reached a dead end and go for coffee etc.] > > Method 2 (yours) rewuires > > 1: Analyze > Tables > Custom Tables > > [box with unexpected message advising nominal vars] > > 2: Continue anyway > > 3: Scroll down to find variables (luckily adjacent positional not > alphabetic mnemonic) > > 4: Highlight and drag to Rows > > 5: Bit frightening here as no explanation, but fart about looking for > summary statistics (students now yawning or fidgeting) and give up. > But settle for COUNT > > 6: Click PASTE just in case > > CTABLES > > /VLABELS VARIABLES=v1565 v1566 v1568 v1570 v1572 v1574 v1576 v1578 > v1608 > v1610 v1612 v1614 v1616 v1618 v1620 v1622 v1624 > > DISPLAY=DEFAULT > > /TABLE v1565 [S][MEAN, COUNT F40.0] + v1566 [S][MEAN, COUNT F40.0] + > v1568 [S][MEAN, COUNT F40.0] + v1570 [S][MEAN, COUNT F40.0] + v1572 > [S][MEAN, COUNT F40.0] + v1574 [S][MEAN, COUNT F40.0] + v1576 > [S][MEAN, COUNT F40.0] + v1578 [S][MEAN, COUNT F40.0] > > + v1608 [S][MEAN, COUNT F40.0] + v1610 [S][MEAN, COUNT F40.0] + v1612 > [S][MEAN, COUNT F40.0] + v1614 [S][MEAN, COUNT F40.0] + v1616 > [S][MEAN, COUNT F40.0] + v1618 [S][MEAN, COUNT F40.0] + v1620 > [S][MEAN, COUNT F40.0] + v1622 [S][MEAN, COUNT F40.0] + v1624 > > [S][MEAN, COUNT F40.0]. > > 7 [If you can find where it went, click on Run > All] or at step 6 > above, press OK and get a table. (need to edit to drag column across) > > > > Mean > > > > Count > > A109B117A NO QUALIFICATION OBTAINED BY R > > > > 1 > > > > 3100 > > A109B117B QUAL OBTND: CSE 2-5 > > > > 1 > > > > 3100 > > A109B117B QUAL OBTND: CSE 1, O LEVEL,ETC > > > > 2 > > > > 3100 > > A109B117B QUAL OBTND: A LEVEL, ETC > > > > 3 > > > > 3100 > > A109B117B QUAL OBTND: O-SEAS LEAVNG EXAM > > > > 4 > > > > 3100 > > A109B117B QUAL OBTND:TRADE APPRENTICESHP > > > > 5 > > > > 3100 > > A109B117B QUAL OBTND: RSA,SIMLR CLERICAL > > > > 6 > > > > 3100 > > A109B117B QUAL OBTND: C & GLDS CRAFT,ETC > > > > 7 > > > > 3100 > > A109B117B QUAL OBTND:C & GLDS ADVNCD,ETC > > > > 8 > > > > 3100 > > A109B117B QUAL OBTND: C & GLDS FULL TECH > > > > 9 > > > > 3100 > > A109B117B QUAL OBTND:BEC ORDNARY,ONC,OND > > > > 10 > > > > 3100 > > A109B117B QUAL OBTND: BEC HIGHER,HNC,HND > > > > 11 > > > > 3100 > > A109B117B QUAL OBTND: TEACHER TRAIN QUAL > > > > 12 > > > > 3100 > > A109B117B QUAL OBTND: NURSING QUAL > > > > 13 > > > > 3100 > > A109B117B QUAL OBTND: OTH TECH,BSNS QUAL > > > > 14 > > > > 3100 > > A109B117B QUAL OBTND: UNIV,CNAA DEGREE,DIP > > > > 15 > > > > 3100 > > A109B117B QUAL OBTND: OTHER OR DK,NA > > > > 97 > > > > 3100 > > This has taken several confusing minutes and most students will have > lost interest, not just in SPSS but also in the question they were > trying to analyse. My table is more useful than yours. My way is > quicker, easier to create, follow and understand, and always will be! > My MULT RESPONSE command can be saved and used again. Thank God for > (obsolete SPSS) syntax and for researchers who are more interested in > research questions than the innards of software. Stick to engineering > and leave the research to us. > > John (with an h) Hall > > Email: [hidden email] <mailto:[hidden email]> > > Website: www.surveyresearch.weebly.com > <http://surveyresearch.weebly.com/> > > Skype: surveyresearcher1 > > Phone: (+33) (0) 2.33.45.91.47 > > I’m busy with other things for now, but > > *From:*Jon K Peck [mailto:[hidden email]] > <mailto:[mailto:[hidden email]]> > *Sent:* 06 March 2012 23:02 > *To:* John F Hall > *Subject:* RE: [SPSSX-L] New tutorials on MULT RESPONSE (in syntax) > {and a new page on Mark Abrams] > > For absolute beginners, the newer technology would be easier. > > In general, though, I don't see why you would spend time updating > something that is obsolete, even if still functional, instead of > spending the time on the newer features. That may be more work, but it > isn't doing someone a favor to be leading them to an old solution when > there is now a better one. It is your judgment, of course, which way > is better, but I would hope that you would first make that judgment. > > Jon Peck (no "h") aka Kim > Senior Software Engineer, IBM > [hidden email] <mailto:[hidden email]> new phone: 720-342-5621 > > > > > From: "John F Hall" <[hidden email] <mailto:[hidden email]>> > To: Jon K Peck/Chicago/IBM@IBMUS > Cc: <[hidden email] <mailto:[hidden email]>> > Date: 03/06/2012 02:35 PM > Subject: RE: [SPSSX-L] New tutorials on MULT RESPONSE (in syntax) {and a > new page on Mark Abrams] > > ------------------------------------------------------------------------ > > > > > I’m working from my 1992 course using spss-x 8 on a Vax for which I > already have working examples: I’m simply converting and upgrading them > for 19. The tutorials are aimed at absolute beginners and use data in > ASCII format from 1986 (same format as many surveys deposited in > archives). They are more about process and logic than the mechanics of > SPSS, but I’ll check out CTABLES for future reference. If it’s not too > complicated I might do something on that, but I still have far too much > to do to finish other SPSS stuff. I can’t even replicate the exercises > in 15 since IBM/SPSS don’t support it any more and I can’t get a renewal > code for my CD. Hopefully they’ll leave 19 up long enough to maintain > screenshot comparability. > > > Email: [hidden email] <mailto:[hidden email]> > Website: www.surveyresearch.weebly.com <http://surveyresearch.weebly.com/> > Skype: surveyresearcher1 > Phone: (+33) (0) 2.33.45.91.47 > > > > > *From:*Jon K Peck [mailto:[hidden email]] * > Sent:* 06 March 2012 19:09* > To:* John F Hall* > Subject:* Re: [SPSSX-L] New tutorials on MULT RESPONSE (in syntax) {and > a new page on Mark Abrams] > > If this works for V11, then you are not taking advantage of the much > superior multiple response capabilities that go with CTABLES. That seems > to be doing a disservice to users with mult response problems. I hope > you have at least made it clear that in versions since 11.5 there is > better technology available, albeit in an option. And, furthermore, that > MULT RESPONSE set definitions have nothing to do with the types of mult > response sets used in CTABLES and GGRAPH and vice versa. And GGRAPH mult > response features are available in the Base product. > > Jon Peck (no "h") aka Kim > Senior Software Engineer, IBM_ > [hidden email] <mailto:[hidden email]> > new phone: 720-342-5621 > > > > > From: John F Hall <[hidden email] <mailto:[hidden email]>> > To: [hidden email] <mailto:[hidden email]> > Date: 03/06/2012 11:02 AM > Subject: [SPSSX-L] New tutorials on MULT RESPONSE (in syntax) {and a new > page on Mark Abrams] > Sent by: "SPSSX(r) Discussion" <[hidden email] > <mailto:[hidden email]>> > > ------------------------------------------------------------------------ > > > > > > I’ve just posted two extensive new tutorials for MULT RESPONSE using > SPSS 19, but the exercises will work with all releases back to SPSS 11. > They use syntax rather than the GUI for basic operations such as data > capture, dictionary specification and tabulation. They proceed gently, > step-by-step, with full colour SPSS 19 screenshots at each step, and use > real data from a wave of the British Social Attitudes (BSA) series. They > relate to earlier tutorials explaining the use of multiple response > questions in surveys, and specifically to different ways they are used > and coded in the BSA. Two more are in preparation, but the available > material is all on page:_ > _http://surveyresearch.weebly.com/33-multiple-response-mult-response.html > > Slightly off-topic, but I’ve also just posted a new page on my old boss > Dr Mark Abrams (1906-1994) who is widely accepted as the father of > survey and market research (and much else) in the UK. The page contains > a brief profile, details of publications, but most important, the > transcript of a fascinating 1984 interview in which describes in > fascinating detail his early childhood growing up in an émigré Jewish > family in North London, his student days at Latymer School and LSE, his > early political education and contacts, his research career from the > 1930s onwards (including time in the USA) and his work for the Labour > Party. Many of the people he refers to, he both knew and worked with: > they are household names in the pantheon of the social sciences. > > The page is: > _ > _http://surveyresearch.weebly.com/mark-abrams.html > > . . and should be of interest to all social scientists, not just > researchers. If any of you are on the SRMSNET or other lists, please > forward this to them. > > John Hall > > > Email: [hidden email] <mailto:[hidden email]> > Website: www.surveyresearch.weebly.com <http://surveyresearch.weebly.com/> > Skype: surveyresearcher1 > Phone: (+33) (0) 2.33.45.91.47 > > > > ===================== 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 |
Re: New tutorials on MULT RESPONSE (in syntax)
|
|
Jon Peck (no "h") aka Kim Senior Software Engineer, IBM [hidden email] new phone: 720-342-5621 From: John F Hall <[hidden email]> To: [hidden email] Date: 03/13/2012 02:02 PM Subject: Re: [SPSSX-L] New tutorials on MULT RESPONSE (in syntax) Sent by: "SPSSX(r) Discussion" <[hidden email]> Precisely I'm aiming at students and researchers who only have Gradpack or who have never heard of SPSS, but have survey data to process (a growing population now that academic institutions are looking for cheaper alternatives) for many of whom English is a second language. My tutorials should work with every release from 11 onwards and my syntax is much easier to understand and follow than anything created by the GUI. >>>I have nothing against syntax - I designed a lot of it (as well as a lot of the SPSS gui), but I'm genuinely baffled. I taught a lot of students in days of yore, and I cannot believe that this syntax solution is easier to understand than the gui. Not to mention that the gui is translated while syntax is not. And current students would be on versions long past 11, so compatibility that far back doesn't seem to be of much benefit. >>>Jon Peck John Hall [snip] Email: [hidden email] Website: www.surveyresearch.weebly.com Skype: surveyresearcher1 |
| Free forum by Nabble | Edit this page |