Teaching survey analysis with SPSS

[John F Hall]

Lots of stats out there and one or two interesting logic and programming problems, but is there anyone who teaches how to use SPSS, not for statistics but for processing and analysing data from questionnaire surveys.  After all, that's how SPSS started.  If so, I'd be happy to swap experiences and materials, in my case from 1972 (blue manual) to 15 for Windows today.

[John F Hall]

Debra, Tim, [Not sure why font just got smaller]

 

I'm in France, in Lower Normandy, in the countryside near a small village called Notre Dame de Cenilly (just south of Cerisy-la-Salle which is about halfway between Coutances and St-Lô - scene of Operation Cobra in July 1944) 

 

Quantum: was that the one started in London in the 1970s?  It might have had a different name and been taken over, but the guy who set it up (can't recall his name, but Andrew Westlake or Beverley Rowe might) was on the Study Group on Computers in Survey Analysis (all six of us)

 

If you like syntax you'll love my 2006 Old Dog, Old Tricks paper Old Dog, Old Tricks: Using SPSS Syntax to Avoid the Mouse Trap and the accompanying slide-shows.  It took lierally hundreds of hours to prepare, but I think the final result is worth it.  You should also look at my critical reviews of SPSS Survival Manual (Julie Pallant, Open University Press, 2001, 2nd edition 2005).  The Social Research Association only wanted 350 words, but I went completely over the top when I saw it.  The full length original (3,000 words) is on Review of Julie Pallant "SPSS Survival Manual" but it was chopped down by half for publication in SRA News.  They also published a shorter, and different, review of the 2nd edition, but it's not on my page yet:  I'll forward it separately. 

 

I have a whole load of new stuff ready to upload, but it's quite complicated to get the hyperlinks in because if I upload anything I then have to amend the links in the navigation guides and other documents.  If I edit existing files I have to re-upload them, they get a different reference number and then the hyperlinks in the guides and other documents have to be changed.  The academia team boast that the site is easy to use and edit: not for me it isn't, but it's better than having to set up my own website from scratch.  I've already learned enough software applications in my time and don't particularly fancy tackling yet another.  Any offers of professional help in this area will be gratefully accepted: I can't pay but will happily share the credit.  Robert Louis SAtevenson once wrote, "I do not not write for lone: I write for money, a far nobler deity."  With me it's the other way round.

 

The five new tutorials on COUNT and COMPUTE contain cross contain cross-references to each other: they'll be a nightmare to insert hyperlinks and upload unless I can use bookmarks and hyperlinks within the site.  They use data from a survey of fifth formers (see Fifth form questionna­ire) to generate indices of teenagers' "attachment to status quo" and "sexism" (negative attitudes to women).  The survey was done (under my guidance and supervision) by three of my 2nd year undergraduate students in 1981: it may be old, but (time, resources and existing technology permitting) it was done to professional standards, has a full user manual Playground to Politics: Users' Manual (Hall & Walker 1982) and a carefully designed SPSS saved file Fifth Form Survey and can still serve as model for similar projects.

----- Original Message -----

From: [hidden email]

To: [hidden email]

Sent: Sunday, September 20, 2009 12:17 AM

Subject: Re: Teaching survey analysis with SPSS

John, thank you so much.  I'm courting deadlines right now & won't be able to look at your material yet, but rest assured I am completely capable of giving feedback once I get into seomthing.

 

What part of the world are you in where it's mid-night now?  I'm assuming you're in Europe.

 

And in case I didn't clarify, yes, I am a syntax person.  In my belief system, the more syntax, the safer the code and the less likely for things to happen unintentionally, or that are intentional but cannot be retraced.

 

Debbie

On Sat, Sep 19, 2009 at 4:51 PM, John F Hall <[hidden email]> wrote:

Debra

 

I've had a long exchange with ViAnn Beadle on the "SPSS and aliens" stream, but most of it has not gone to SPSSx listserv.  She was with SPSS for 34 years and is responsible for most of the coding for the Windows version.  She's clearly proud of her drop-down menus, but I'm a died-in-the-wool syntax man myself. 

 

Have a look at my webpage http://independent.academia.edu/JohnFHall to see where I'm coming from and the guide to SPSS materials under teaching http://independent.academia.edu:80/JohnFHall/attachment/345163/full/Guide-to-learning-materials-and-SPSS.  I've  attached a shorter version which is easier to follow.  The SPSS stuff is all based on the postgrad course Survey Analysis Workshop which I designed and taught at PNL until 1992 when I took early retirement.  You might like one of the slide shows (the first item under Talks) which accompanied my mammoth paper to ASSESS (SPSS users) at York in 2006, even the whole paper: it's quite long, but it's fun. 

 

Everything is available for free download.

 

Being retired I've theoretically got all the time in the world, but I do also have a large garden to tend and a backlog of 40-odd films to catch up on.  I've just spent the last week writing five more tutorials to add to my page, but they a need a bit of final editing.  What I'm really short of is feedback on ease of use and understanding.  When I was teaching I had a classroom full of attentive students, but with distance learning I'm working blind, so to speak.  Three of my articles on a different site have had amost 60,000 hits, but I've only ever had one feedback from it, and no idea how many downloads..

 

My webpage has had a few hits, but feedback is there none so far, except for two uninformative "liked it" and one from an intern at academia who said it was just like being in the classroom with me.  How sweet!

 

Been on this machine since 4:30 am and it's now almost midnight.  Let me know what you think of my stuff.  I can also send you the latest draft tutorials (on COUNT and COMPUTE)

 

John

 

 

----- Original Message -----

From: [hidden email]

To: [hidden email]

Sent: Saturday, September 19, 2009 10:29 PM

Subject: Re: Teaching survey analysis with SPSS

John, I'm interested in processing & analysing data.  Have many years under my belt with Quantum and a couple years with SPSS.  Prefer SAS & would like to learn Stata, but I go with what those around me are using, which is often SPSS these days.  I don't teach SPSS analysis because I'm not sufficiently adept at it.  However, I'm a teacher by nature, and if I become more adept, could see myself teaching.

 

I'm using version 17, and may go to 18.  There's a huge difference between 15, 17 & 18 ... one reason why I'm not happy with SPSS, especially because you can't readily use files from a lower version in a higher version.

 

My time at the moment is also quite limited, but want you to know I'm interested if you put materials together.

 

Debbie Miller
Mixed Methods Consulting

 

----- Original Message -----

From: [hidden email]

To: [hidden email]

Sent: Sunday, September 20, 2009 1:36 AM

Subject: RE: Teaching survey analysis with SPSS

I dont teach - but thats pretty much all I do

On Sat, Sep 19, 2009 at 3:14 PM, John F Hall <[hidden email]> wrote:

Lots of stats out there and one or two interesting logic and programming problems, but is there anyone who teaches how to use SPSS, not for statistics but for processing and analysing data from questionnaire surveys.  After all, that's how SPSS started.  If so, I'd be happy to swap experiences and materials, in my case from 1972 (blue manual) to 15 for Windows today.

[Bruce Weaver]

John F Hall-2 wrote:
>
> Lots of stats out there and one or two interesting logic and programming
> problems, but is there anyone who teaches how to use SPSS, not for
> statistics but for processing and analysing data from questionnaire
> surveys.  After all, that's how SPSS started.  If so, I'd be happy to swap
> experiences and materials, in my case from 1972 (blue manual) to 15 for
> Windows today.
>

Hi John.  I teach a 4th year (so-called) advanced stats course for
psychology students.  For the first 5 weeks, the lab time is used to do a
series of tutorials.  The first tutorial is focused on teaching the students
to exit the dialogs via PASTE rather than OK.  The remaining tutorials
address different data management issues (e.g., ADD FILES, MATCH FILES,
AGGREGATE, and restructuring the data from long-to-wide or vice versa).  The
data management tutorials are modifications of ones found on the UCLA
Computing Services website (under intermediate data management at the link
given below).  My modifications mainly consist of inserting some
screen-shots and fleshing out the instructions a bit.

  http://www.ats.ucla.edu/stat/Spss/topics/data_management.htm

In the remaining part of the course, students then work on 3 assignments
that require some kind of data management step or steps before they can do
the analysis.  The reason I do this is that in my own experience, data
analysts often spend much more time on data management issues than on the
actual analysis.  And in my own experience, most stats courses that use SPSS
(or some other stats package) completely ignore data management.

Not sure if this is what you were after, but I hope it helps.

Cheers,
Bruce


-----
--
Bruce Weaver
[hidden email]
http://sites.google.com/a/lakeheadu.ca/bweaver/
"When all else fails, RTFM."

NOTE:  My Hotmail account is for posting only, and is not monitored
regularly.
If you wish to send me an e-mail, please use the address shown in my sig
file.
--
View this message in context: http://www.nabble.com/Teaching-survey-analysis-with-SPSS-tp25525045p25530322.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

[John F Hall]

I've just added some new tutorials to my webpage.  There's an annotated guide on COUNT and COMPUTE - Annotated guide to tutorials with hyperlinks to each tutorial.  They use data from a real survey done in a North London comprehensive school (11-18 mixed) by three of my undergraduate students for their 2nd year group dissertation.

 

The tutorials give background on the survey, stress the need for careful data checks, then go on to demonstrate (and discuss methodological and technical issues surrounding) the use of SPSS commands COUNT and COMPUTE to create scores on a simple set of attitude scales to measure:

 

1:         Teenagers' attachment to the status quo

2:         Negative attitudes to women.

 

All SPSS exercises are performed first in direct syntax, then repeated using drop-down menus (but without output).  Each tutorial contains appropriate facsimile extracts from the questionnaire or user manual and proceeds at a gentle pace with full SPSS screen-dumps at each step. 

 

There is a separate guide to all other SPSS teaching materials on Guide to learning materials and SPSS but it would have been impossibly complex to add the new tutorials to it.

 

Everything is available for free download.

 

 

[John F Hall]

ViAnn

 

I have been re-writing  tutorials introducing beginners to the use of COUNT and COMPUTE to generate (admittedly crude) indices of negative attitudes to women i.e. "sexism" on data from the saved file Fifth Form Survey  (questionnaire extract below)

 

<IMG height=696 src="file://C:\DOCUME~1\Owner\LOCALS~1\Temp\msohtml1\01\clip_image002.gif" width=649 v:shapes="_x0000_i1025">

 

In the tutorial I state:

 

It is possible to construct a crude index of "Sexism" from nine of these items, five of which are negative (a,e,f,j,o) and four positive (d,h,m,n ).  To be a sexist a pupil must agree with the negative items (a,e,f,j and o) and disagree with the positive items (d,h,m and n).  Thus we need to construct an index which counts the number of agreements (codes 3 and 4) with the first set together with the number of disagreements (codes 1 and 2) with  the second.  This will yield a score in the range 0 to 9 where 9 indicates high sexism.

 

The tutorial demonstrates how to do this in both direct syntax and using drop-down menus.  Data are from record 2 and have varnames indicating record and field. 

 

The syntax command:

 

count     SEXISM=V248 V252 V253 V256 V261 (3,4)  V251 V255 V259 V260 (1,2).

 

works perfectly and (if printed up) uses hardly any ink or paper. 

 

However, when I try to do this using the drop down menus, I have not yet found a way to generate this index in a single pass using two different source lists, each with different values.  As far as I can make out, it is necessary to create two intermediate variables, one to count agreement with items a, e, f, j and o and another to count disagreement with items d, j, m, and n and then add the two together.  This process is incredibly cumbersome, and clearly demonstrates the superiority of syntax over menus for this kind of analysis.

 

The addition of menus has added several unnecessary pages (ink, paper?) to the tutorials, far too much to reproduce here.

 

The syntax generated by SPSS from the menus is:

 

 

 

COUNT

  sexneg = v248 v252 v253 v256 v261  (3)  v248 v252 v253 v256 v261  (4)  .

VARIABLE LABELS sexneg 'q33 Agree a,e,f,j,o' .

EXECUTE .

COUNT

  sexpos = v251 v255 v259 v260  (1)  v251 v255 v259 v260  (2)  .

VARIABLE LABELS sexpos 'q33 Disagree d,h,m,n' .

EXECUTE .

COMPUTE sexism = sexneg + sexpos .

EXECUTE .

 

 

Have I missed something?

 

My students would have given up and gone home long before they got through the menus.

 

John

 

PS  It's nearly as bad using RECODE and COMPUTE to generate a superior measure which takes into account disagreement as well as agreement, and also the level of each.

[J P-6]

Hello,

 

As the title implies, this is really a stat quesiton. But I have no one else to turn to.....

 

Can someone advise me on a statistical test comparing correlated proportions when there is significant attrition (75%) between time 1 and time 2. The initial t1 n is 96 and the second t2 n is 23. Proportion of 'pass' at time 1 is 31% (n= 30) and time 2 it is 83% (n=20).

 

The best I can come up with is a z-test of whether proportion at t1 is equal to porporion at t2.

 

Any advice is appreciated.

 

Best,

John

[John F Hall]

This is a course for beginners, so reliabilty test way ahead of anything they need to know.  This is about the survey process, not about statistics, nor for that matter about SPSS.  Any software will do, but SPSS is what we used.

 

Items were initially selected by factor and reliability long before any preparation of tutorials, but students don't need to know that.  In class I used anyway to tell them I'd cheated a bit.

 

----- Original Message -----

From: [hidden email]

To: [hidden email]

Sent: Tuesday, September 29, 2009 10:09 PM

Subject: RE: Teaching survey analysis with SPSS

No you have not missed anything here—there is no dialog box UI for the COUNT command. Note, however, if you are going to show your users how to use the dialog boxes generate syntax be sure and talk about the Execute command (what it is and why it’s used) and how to turn it off when pasting transformations. BTW, do you actually run a reliability analysis on these items before glomming them into a one scale? That might be a useful exercise.

 

From: John F Hall [mailto:[hidden email]]
Sent: Tuesday, September 29, 2009 12:59 PM
To: ViAnn Beadle; [hidden email]
Subject: Re: Teaching survey analysis with SPSS

 

ViAnn

 

I have been re-writing  tutorials introducing beginners to the use of COUNT and COMPUTE to generate (admittedly crude) indices of negative attitudes to women i.e. "sexism" on data from the saved file Fifth Form Survey  (questionnaire extract below)

 

<IMG id=_x0000_i1025 height=696 src="file:///C:\DOCUME~1\Owner\LOCALS~1\Temp\msohtml1\01\clip_image002.gif" width=649 border=0>

 

In the tutorial I state:

 

It is possible to construct a crude index of "Sexism" from nine of these items, five of which are negative (a,e,f,j,o) and four positive (d,h,m,n ).  To be a sexist a pupil must agree with the negative items (a,e,f,j and o) and disagree with the positive items (d,h,m and n).  Thus we need to construct an index which counts the number of agreements (codes 3 and 4) with the first set together with the number of disagreements (codes 1 and 2) with  the second.  This will yield a score in the range 0 to 9 where 9 indicates high sexism.

 

The tutorial demonstrates how to do this in both direct syntax and using drop-down menus.  Data are from record 2 and have varnames indicating record and field. 

 

The syntax command:

 

count     SEXISM=V248 V252 V253 V256 V261 (3,4)  V251 V255 V259 V260 (1,2).

 

works perfectly and (if printed up) uses hardly any ink or paper. 

 

However, when I try to do this using the drop down menus, I have not yet found a way to generate this index in a single pass using two different source lists, each with different values.  As far as I can make out, it is necessary to create two intermediate variables, one to count agreement with items a, e, f, j and o and another to count disagreement with items d, j, m, and n and then add the two together.  This process is incredibly cumbersome, and clearly demonstrates the superiority of syntax over menus for this kind of analysis.

 

The addition of menus has added several unnecessary pages (ink, paper?) to the tutorials, far too much to reproduce here.

 

The syntax generated by SPSS from the menus is:

 

 

 

 

COUNT

  sexneg = v248 v252 v253 v256 v261  (3)  v248 v252 v253 v256 v261  (4)  .

VARIABLE LABELS sexneg 'q33 Agree a,e,f,j,o' .

EXECUTE .

COUNT

  sexpos = v251 v255 v259 v260  (1)  v251 v255 v259 v260  (2)  .

VARIABLE LABELS sexpos 'q33 Disagree d,h,m,n' .

EXECUTE .

COMPUTE sexism = sexneg + sexpos .

EXECUTE .

 

 

Have I missed something?

 

My students would have given up and gone home long before they got through the menus.

 

John

 

PS It's nearly as bad using RECODE and COMPUTE to generate a superior measure which takes into account disagreement as well as agreement, and also the level of each.

[John F Hall]

I think we're talking at cross-purposes here.  This particular sample is 100% of the population anyway, so statistical inference doesn't really come into it, unless comparisons are sought with other studies from which some questions were replicated.

 

At this stage of the course, my students (many of whom had little or no previous experience of statistics or computing) were still learning the language of survey research and grappling with computer technology and  the mechanics of data handling.  They were much happier starting from initial research questions:

 

Is there such a thing as "sexism"? 

How do we measure it?

Has anyone else measured it before us? 

 

...and only later moved to questions of methodology and/or statistical inference:

 

Is our measure valid? 

Is our measure reliable? 

Are there differences between groups in levels of "sexism" ? 

Are the differences real or could they arise by chance? 

How do we account for variations in levels of "sexism" ? 

What variables (singly or in combination) are likely to affect  levels of "sexism", and to what extent?

 

Tutorials and exercises using this small survey (230 variables, N=142) as a teaching aid, together with others using data from the fun questionnaire completed by students at the beginning of the course (21 variables, N = 169 in waves of 25 or so per semester) helped to keep students motivated and get them up to speed with computer technology, survey research materials and routine SPSS usage.  In subsequent class and homework exercises, and for the independent analysis and reporting for their chosen assessment topic, they were able to tackle, with confidence and minimal technical assistance, data  from a major national survey in the public domain http://www.data-archive.ac.uk/findingData/snDescription.asp?sn=2723 (British Social Attitudes 1989, 1000+ variables, N= 3,000+)

 

By the end of the course we had actually covered a great deal of inferential statistics specifically geared to, and in sequence with, the analyses peformed and survey data sets used.  Notes written by Jim Ring Survey Analysis Workshop - Statistical Notes (63 pages, 395kb) derive from teaching the statistical component of professional practice oriented courses in survey data collection, data management, computer processing and statistical analysis to social science students at the Polytechnic of North London from 1976 to 1992.  They represented an attempt to fill a gap in the textbook provision for students who found computers and statistics daunting, and were mostly written before the appearance of the original and still much sought-after SPSS Guide to Data Analysis (Norusis,1990: later editions relating to SPSS for Windows are less helpful for beginners). They were not intended as a replacement, and should be used in conjuction with the recommended texts.

 

 

----- Original Message -----

From: [hidden email]

To: [hidden email]

Sent: Tuesday, September 29, 2009 11:40 PM

Subject: RE: Teaching survey analysis with SPSS

I guess I think that the survey process is about using statistics to make inferences for what is essentially categorical data. How can you make any inferences from the sample to the  unmeasured population unless you’re willing to look at the statistics.

 

From: John F Hall [mailto:[hidden email]]
Sent: Tuesday, September 29, 2009 3:13 PM
To: ViAnn Beadle
Cc: [hidden email]
Subject: Re: Teaching survey analysis with SPSS

 

This is a course for beginners, so reliabilty test way ahead of anything they need to know.  This is about the survey process, not about statistics, nor for that matter about SPSS.  Any software will do, but SPSS is what we used.

 

Items were initially selected by factor and reliability long before any preparation of tutorials, but students don't need to know that.  In class I used anyway to tell them I'd cheated a bit.

 

----- Original Message -----

From: [hidden email]

To: [hidden email]

Sent: Tuesday, September 29, 2009 10:09 PM

Subject: RE: Teaching survey analysis with SPSS

 

No you have not missed anything here—there is no dialog box UI for the COUNT command. Note, however, if you are going to show your users how to use the dialog boxes generate syntax be sure and talk about the Execute command (what it is and why it’s used) and how to turn it off when pasting transformations. BTW, do you actually run a reliability analysis on these items before glomming them into a one scale? That might be a useful exercise.

 

From: John F Hall [mailto:[hidden email]]
Sent: Tuesday, September 29, 2009 12:59 PM
To: ViAnn Beadle; [hidden email]
Subject: Re: Teaching survey analysis with SPSS

 

ViAnn

 

I have been re-writing  tutorials introducing beginners to the use of COUNT and COMPUTE to generate (admittedly crude) indices of negative attitudes to women i.e. "sexism" on data from the saved file Fifth Form Survey  (questionnaire extract below)

 

<IMG id=_x0000_i1025 height=696 src="file:///C:\DOCUME~1\Owner\LOCALS~1\Temp\msohtml1\01\clip_image002.gif" width=649 border=0>

 

In the tutorial I state:

 

It is possible to construct a crude index of "Sexism" from nine of these items, five of which are negative (a,e,f,j,o) and four positive (d,h,m,n ).  To be a sexist a pupil must agree with the negative items (a,e,f,j and o) and disagree with the positive items (d,h,m and n).  Thus we need to construct an index which counts the number of agreements (codes 3 and 4) with the first set together with the number of disagreements (codes 1 and 2) with  the second.  This will yield a score in the range 0 to 9 where 9 indicates high sexism.

 

The tutorial demonstrates how to do this in both direct syntax and using drop-down menus.  Data are from record 2 and have varnames indicating record and field. 

 

The syntax command:

 

count     SEXISM=V248 V252 V253 V256 V261 (3,4)  V251 V255 V259 V260 (1,2).

 

works perfectly and (if printed up) uses hardly any ink or paper. 

 

However, when I try to do this using the drop down menus, I have not yet found a way to generate this index in a single pass using two different source lists, each with different values.  As far as I can make out, it is necessary to create two intermediate variables, one to count agreement with items a, e, f, j and o and another to count disagreement with items d, j, m, and n and then add the two together.  This process is incredibly cumbersome, and clearly demonstrates the superiority of syntax over menus for this kind of analysis.

 

The addition of menus has added several unnecessary pages (ink, paper?) to the tutorials, far too much to reproduce here.

 

The syntax generated by SPSS from the menus is:

 

 

 

 

COUNT

  sexneg = v248 v252 v253 v256 v261  (3)  v248 v252 v253 v256 v261  (4)  .

VARIABLE LABELS sexneg 'q33 Agree a,e,f,j,o' .

EXECUTE .

COUNT

  sexpos = v251 v255 v259 v260  (1)  v251 v255 v259 v260  (2)  .

VARIABLE LABELS sexpos 'q33 Disagree d,h,m,n' .

EXECUTE .

COMPUTE sexism = sexneg + sexpos .

EXECUTE .

 

 

Have I missed something?

 

My students would have given up and gone home long before they got through the menus.

 

John

 

PS It's nearly as bad using RECODE and COMPUTE to generate a superior measure which takes into account disagreement as well as agreement, and also the level of each.

[Martin Holt]

A few days ago I sent a query asking whether a 100% sample taken at one time
constituted "the population" if the individuals that made up that 100%
changed over time. I'm thinking here that a 95% confidence interval is
so-called because the population mean falls within that interval 95%
_of_the_time_. I've not had a reply, and reviewing my 'sent' tray I cannot
see the original posting, so I'm just sending this in to see if I'm on
track. I guess if the 100% is made up of certain defined individuals whom
you are going to study irrespective of whether other individuals join the
sample (you're going to ignore these), then the 100% sample is your
population. Otherwise, I guess the original 100% sample is a sample.

BW,
Martin Holt

=====================
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]

I've just spent the last 3 days on major revisons to the new set of tutorials introducing the use of  COUNT and COMPUTE to create scores on simple attitude scales.  This has confirmed my allegiance to direct syntax over drop-down menus, but with one exception I have used both methods in all the exercises.  The whole exercise has been horrendous, but there are now seven tutorials uploaded and also an annotated guide, with hyperlinks to all seven.  Abstract and URL below

  

---

 

Guide to COUNT and COMPUTE tutorials

This set of tutorials uses data from a survey of fifth formers to demonstrate the construction of scores on a simple set of attitude scales using SPSS commands COUNT and COMPUTE.

Introducti­on to COUNT and COMPUTE

Guide, with hyperlinks, to a set of tutorials demonstrating the use of COUNT and COMPUTE to create simple attitude scores. Two sets of items are used, one to measure attachment to status quo, the other negative attitudes to women. Two measures are generated for each, one with COUNT and the other with COMPUTE. All examples are fully worked out using both SPSS syntax and drop-down menus together with embedded facsimile questionnaire extracts and SPSS screen dumps. All SPSS exercises are performed first in direct syntax, then repeated using drop-down menus. Each tutorial contains appropriate facsimile extracts from the questionnaire or user manual and proceeds at a gentle pace with full SPSS screen-dumps at each step.

 

 

---- Original Message -----

From: [hidden email]

To: [hidden email]

Cc: [hidden email] ; [hidden email] ; [hidden email] ; [hidden email] ; [hidden email] ; [hidden email] ; [hidden email] ; [hidden email] ; [hidden email] ; [hidden email] ; [hidden email] ; [hidden email] ; [hidden email] ; [hidden email] ; [hidden email] ; [hidden email]

Sent: Saturday, September 26, 2009 8:58 AM

Subject: Fw: Teaching survey analysis with SPSS

One of the tutorials The COMPUTE command 1 – Attachment to status quo listed on COUNT and COMPUTE - Annotated guide to tutorials is somehow missing a whole section.  No idea how it happened, but I'll have to upload a new version.  This means having to change the guide and reupload that as well.  Must be more careful in future.

 

----- Original Message -----

From: [hidden email]

Sent: Friday, September 25, 2009 5:35 PM

Subject: Re: Teaching survey analysis with SPSS

I've just added some new tutorials to my webpage.  There's an annotated guide on COUNT and COMPUTE - Annotated guide to tutorials with hyperlinks to each tutorial.  They use data from a real survey done in a North London comprehensive school (11-18 mixed) by three of my undergraduate students for their 2nd year group dissertation.

 

The tutorials give background on the survey, stress the need for careful data checks, then go on to demonstrate (and discuss methodological and technical issues surrounding) the use of SPSS commands COUNT and COMPUTE to create scores on a simple set of attitude scales to measure:

 

1:         Teenagers' attachment to the status quo

2:         Negative attitudes to women.

 

All SPSS exercises are performed first in direct syntax, then repeated using drop-down menus (but without output).  Each tutorial contains appropriate facsimile extracts from the questionnaire or user manual and proceeds at a gentle pace with full SPSS screen-dumps at each step. 

 

There is a separate guide to all other SPSS teaching materials on Guide to learning materials and SPSS but it would have been impossibly complex to add the new tutorials to it.

 

Everything is available for free download.

 

 

[John F Hall]

Apologies for duplicate posting, but more people might read this when the subject is changed to "Teaching survey analysis with SPSS."

 

This could well explain why the best-selling SPSS Survival Manuals for 10/11 and 12  (Julie Pallant, Open University Press, 1st edition, SPSS 10/11, 2001, 2nd edition, SPSS 12 2005) use only drop-down menus.

 

I reviewed both editions, critically and at length, for the Social Research Association (UK).  The first was heavily abridged for publication, but the full length original (3,100 words) is on Review of Julie Pallant "SPSS Survival Manual" 1st edition 2001  and the second (1,800 words) on Review of Julie Pallant, SPSS Survival Manual 2nd edition 2005

 

These excellent and deservedly best-selling books are more helpful to students and researchers in psychology than those in sociology or social research.   In my reviews I drew attention to important SPSS facilities and procedures, and to analysis involving tabulation, which get no mention in either book, but was particularly appalled by the exclusive use of drop-down menus and complete lack of any direct syntax.  One of the slide shows Old Dog, Old Tricks 5: Exercises from SPSS Survival Manual for my Old Dog, Old Tricks paper Old Dog, Old Tricks: Using SPSS Syntax to Avoid the Mouse Trap even repeats a full set of exercises  from the second edition.  Both reviews and the slide show contain detailed examples demonstrating the clear superiority, in my opinion, of syntax over drop-down menus.

 

Were syntax and PASTE available in Student and GradPack versions 10, 11 and 12?

 

If not, I owe Julie an unreserved and craven apology.

 

On the bright side, if I hadn't requested the SRA's review copy of the 1st edition I may never have requested a licence for SPSS for Windows, never have recreated important data sets from 40 years in survey research and never have started updating, converting and writing SPSS learning materials (See Guide to learning materials and SPSS and Introducti­on to COUNT and COMPUTE) based on 38 years of using and teaching various releases of SPSS on a range of mainframes and, since 2002, SPSS for Windows on a PC.

 

For all this I owe Julie heartfelt thanks.

 

----- Original Message -----

From: [hidden email]

To: [hidden email]

Sent: Thursday, October 01, 2009 11:41 PM

Subject: Re: Unable to access Syntax Window (SPSS 16.0 for Windows, running on Vista)

Grad Pack is a different product from the Student version, and does have a syntax window and Paste functionality.

 

---

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Debra Miller
Sent: Thursday, October 01, 2009 4:14 PM
To: [hidden email]
Subject: Re: Unable to access Syntax Window (SPSS 16.0 for Windows, running on Vista)

 

Jonathon, I have a student grad pack version (17) that I operate on Windows XP.  I'm able to use syntax in a separate window.  Is a grad version different than other student versions, and/or is my grad syntax window different than a syntax window available on full professional versions?

 

Debbie Miller

Mixed Methods Consulting

On Thu, Oct 1, 2009 at 1:58 PM, John F Hall <[hidden email]> wrote:

Not much point in me writing tutorials then, is there?  How on earth do they expect students to learn anything if all they can do is point and click?

 

See also threads on

 

Teaching survey analysis with SPSS

SPSS and aliens

 

Plus all the stuff on PASW, IBM takeovers etc..  Doesn't anyone remember how SPSS started?  I thought it was by and for social scientists doing survey analysis.

----- Original Message -----

From: [hidden email]

To: [hidden email] ; [hidden email]

Sent: Thursday, October 01, 2009 7:10 PM

Subject: RE: Re: Unable to access Syntax Window (SPSS 16.0 for Windows, running on Vista)

 

There is no syntax window in any Student version, and no Paste function.

 

---

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of John F Hall
Sent: Thursday, October 01, 2009 10:25 AM
To: [hidden email]
Subject: Re: Unable to access Syntax Window (SPSS 16.0 for Windows, running on Vista)

 

Just in case it wasn't clear what I meant::

 

Go to your Data Editor and click on :

 

 

File

   New

      Syntax

 

write:

 

corr var1 with var2.

 

in the box and run the job with [CTRL]R or Run...Current line.

 

 

----- Original Message -----

From: [hidden email]

To: [hidden email] ; [hidden email]

Sent: Thursday, October 01, 2009 3:17 PM

Subject: Re: Unable to access Syntax Window (SPSS 16.0 for Windows, running on Vista)

 

Not sure what you mean by bivariate, but have you tried syntax?

 

File

   New

      Syntax

 

corr x with y.

 

 

----- Original Message -----

From: [hidden email]

To: [hidden email]

Sent: Thursday, October 01, 2009 2:40 PM

Subject: Unable to access Syntax Window (SPSS 16.0 for Windows, running on Vista)

 

Does anyone know how to fix this problem? When running a bivariate correlation analysis

, there is no paste button visible in the dialog box, even after I move variables into the variable box. The only buttons visible are OK, Reset, Cancel and Help. [I am running student version 16.0 on Vista].

 

 

[Kornbrot, Diana]

Think it should also go to computer science, comq and the life sciences [don;t know the address, as may be more than one ? Bioq, ?env].
It seems to me that as a community we should distribute widely and let those for whom its irrelevant delete, rather than have people opt in –we’d be bound to miss people, inclduing grad students]

Also desiurrable to go in events on  both HHRI and psy web site, partly for us and partly for next Rae ‘research environment’
Best

Diana


On 02/10/2009 08:16, "John F Hall" <johnfhall@...> wrote:

Apologies for duplicate posting, but more people might read this when the subject is changed to "Teaching survey analysis with SPSS."
 
This could well explain why the best-selling SPSS Survival Manuals for 10/11 and 12  (Julie Pallant, Open University Press, 1st edition, SPSS 10/11, 2001, 2nd edition, SPSS 12 2005) use only drop-down menus.
 
I reviewed both editions, critically and at length, for the Social Research Association (UK).  The first was heavily abridged for publication, but the full length original (3,100 words) is on Review of Julie Pallant "SPSS Survival Manual" 1st edition 2001 <http://independent.academia.edu/JohnFHall/Papers/76935/Review-of-Julie-Pallant--SPSS-Survival-Manual--1st-edition-2001>   and the second (1,800 words) on Review of Julie Pallant, SPSS Survival Manual 2nd edition 2005 <http://independent.academia.edu/JohnFHall/Papers/111393/Review-of-Julie-Pallant--SPSS-Survival-Manual-2nd-edition-2005>
 
These excellent and deservedly best-selling books are more helpful to students and researchers in psychology than those in sociology or social research.   In my reviews I drew attention to important SPSS facilities and procedures, and to analysis involving tabulation, which get no mention in either book, but was particularly appalled by the exclusive use of drop-down menus and complete lack of any direct syntax.  One of the slide shows Old Dog, Old Tricks 5: Exercises from SPSS Survival Manual  <http://independent.academia.edu/JohnFHall/attachment/96022/full/Old-Dog--Old-Tricks-5--Exercises-from-SPSS-Survival-Manual-> for my Old Dog, Old Tricks paper Old Dog, Old Tricks: Using SPSS Syntax to Avoid the Mouse Trap <http://independent.academia.edu/documents/0009/1070/Old_Dog__Old_Tricks.doc> even repeats a full set of exercises  from the second edition.  Both reviews and the slide show contain detailed examples demonstrating the clear superiority, in my opinion, of syntax over drop-down menus.
 
Were syntax and PASTE available in Student and GradPack versions 10, 11 and 12?
 
If not, I owe Julie an unreserved and craven apology.
 
On the bright side, if I hadn't requested the SRA's review copy of the 1st edition I may never have requested a licence for SPSS for Windows, never have recreated important data sets from 40 years in survey research and never have started updating, converting and writing SPSS learning materials (See Guide to learning materials and SPSS <http://independent.academia.edu/JohnFHall/attachment/345163/full/Guide-to-learning-materials-and-SPSS>  and Introducti­on to COUNT and COMPUTE <http://independent.academia.edu/JohnFHall/attachment/392119/full/Introduction-to-COUNT-and-COMPUTE> ) based on 38 years of using and teaching various releases of SPSS on a range of mainframes and, since 2002, SPSS for Windows on a PC.
 
For all this I owe Julie heartfelt thanks.
 
----- Original Message -----

From:  Fry, Jonathan B. <[hidden email]>  
 
To: SPSSX-L@...
 
Sent: Thursday, October 01, 2009 11:41  PM
 
Subject: Re: Unable to access Syntax  Window (SPSS 16.0 for Windows, running on Vista)
 

 
 

Grad Pack is a  different product from the Student version, and does have a syntax window and  Paste functionality.



 

---

From:  SPSSX(r) Discussion [[hidden email]] On Behalf Of Debra Miller
Sent: Thursday, October 01, 2009 4:14  PM
To: SPSSX-L@...
Subject: Re: Unable to access Syntax  Window (SPSS 16.0 for Windows, running on Vista)



 

Jonathon, I have a student grad pack version (17) that  I operate on Windows XP.  I'm able to use syntax in a separate  window.  Is a grad version different than other student versions, and/or  is my grad syntax window different than a syntax window available on full  professional versions?

 



 

Debbie Miller

 

Mixed Methods  Consulting

 

On Thu, Oct 1, 2009 at 1:58 PM, John F Hall <johnfhall@...>  wrote:

 
 
 

Not much point in me writing  tutorials then, is there?  How on earth do they expect students to learn  anything if all they can do is point and  click?

 



 

See also threads  on

 



 

Teaching survey analysis with  SPSS

 

SPSS and  aliens

 



 

Plus all the stuff on PASW, IBM  takeovers etc..  Doesn't anyone remember how SPSS started?  I  thought it was by and for social scientists doing survey  analysis.

 
 

 

----- Original Message -----  

 

From: Fry, Jonathan  B. <[hidden email]>  

 

To: John F Hall <[hidden email]>  ; SPSSX-L@...  

 

Sent:  Thursday, October 01, 2009 7:10 PM

 

Subject: RE: Re:  Unable to access Syntax Window (SPSS 16.0 for Windows, running on Vista)

 



 

There is no syntax  window in any Student version, and no Paste  function.




 

---

From:  SPSSX(r) Discussion [[hidden email]] On Behalf Of John F Hall
Sent: Thursday, October 01, 2009 10:25  AM
To: SPSSX-L@...
Subject: Re: Unable to access Syntax  Window (SPSS 16.0 for Windows, running on Vista)




 

Just in case it wasn't clear  what I meant::

 



 

Go to your Data Editor  and click on :

 



 
 



 

File

 

   New

 

      Syntax

 



 

write:

 



 

corr var1 with  var2.

 



 

in the box and run the job with  [CTRL]R or Run...Current line.

 



 

 

----- Original Message -----  

 

From: John F Hall <[hidden email]>  

 

To: Ursula Mills <[hidden email]>  ; SPSSX-L@...  

 

Sent:  Thursday, October 01, 2009 3:17 PM

 

Subject: Re:  Unable to access Syntax Window (SPSS 16.0 for Windows, running on  Vista)

 



 

Not sure what you mean by  bivariate, but have you tried syntax?

 



 

File

 

   New

 

      Syntax

 



 

corr x with  y.

 



 

 

----- Original Message -----  

 

From: Ursula Mills <[hidden email]>  

 

To: SPSSX-L@...  

 

Sent:  Thursday, October 01, 2009 2:40 PM

 

Subject:  Unable to access Syntax Window (SPSS 16.0 for Windows, running on  Vista)

 


  
  Does anyone know how  to fix this problem? When running a bivariate correlation  analysis , there is no paste  button visible in the dialog box, even after I move variables into  the variable box. The only buttons visible are OK, Reset, Cancel  and Help. [I am running student version 16.0 on Vista].

---

Professor Diana Kornbrot
email:? d.e.kornbrot@...    
web:    http://web.me.com/kornbrot/KornbrotHome.html
Work
School of Psychology
 University of Hertfordshire
 College Lane, Hatfield, Hertfordshire AL10 9AB, UK
 voice:   [phone]
   fax:     [phone]
Home
 19 Elmhurst Avenue
 London N2 0LT, UK
    voice:   [phone]
    mobile: [phone]
   fax:      [phone]