SPSSX Discussion

Tail wagging dog

Classic

List

Threaded

7 messages Options

John F Hall

Tail wagging dog

Viann, Bruce

Car drives driver, cart drives horse, etc., etc?

http://www.phrases.org.uk/meanings/341850.html helps to explain the meaning of the phrase "tail wagging the dog", but a better example is from Angus Campbell's article in ISSJ 1974 about how to develop and maintain research centres.

Quoting someone else's maxim, he states that it is important that administrators in research institutes understand that there are three groups of people more important than administrators, "the research scientists, the research scientists, and the research scientists". I feel the same way about SPSS, statisticians and computer programmers.

I had a similar problem once with an administrator trying to tell academics how to run the department, in particular my course. She got very upset whan I told her it was her job to make my life easier, not the other way round. That's why I always did my own administration in the Survey Research Unit. The only time I allowed an admin person anywhere near a project led to financial disaster.

Probably extreme, but that's what I mean by "tail wagging dog". SPSS is there to do what I tell it, not the other way round, but nowadays predictive analysis and business requirements seem to have taken over from survey research.

Does that explain or not?

John

ajayohri

Re: Tail wagging dog

Or as they say in here- The Customer is Always Right.

Website- http://decisionstats.com

Graduate Student
University of Tennessee, Knoxville.
Go Vols!

On Sat, Oct 24, 2009 at 12:24 PM, John F Hall <[hidden email]> wrote:

Viann, Bruce

Car drives driver, cart drives horse, etc., etc?

http://www.phrases.org.uk/meanings/341850.html helps to explain the meaning of the phrase "tail wagging the dog", but a better example is from Angus Campbell's article in ISSJ 1974 about how to develop and maintain research centres.

Quoting someone else's maxim, he states that it is important that administrators in research institutes understand that there are three groups of people more important than administrators, "the research scientists, the research scientists, and the research scientists". I feel the same way about SPSS, statisticians and computer programmers.

I had a similar problem once with an administrator trying to tell academics how to run the department, in particular my course. She got very upset whan I told her it was her job to make my life easier, not the other way round. That's why I always did my own administration in the Survey Research Unit. The only time I allowed an admin person anywhere near a project led to financial disaster.

Probably extreme, but that's what I mean by "tail wagging dog". SPSS is there to do what I tell it, not the other way round, but nowadays predictive analysis and business requirements seem to have taken over from survey research.

Does that explain or not?

John

ViAnn Beadle

Re: Tail wagging dog

In reply to this post by John F Hall

SPSS the product does what you tell it to do so I’m not sure that your analogy works.

SPSS the company has never tried to be an evangelist for new and experimental approaches to data analysis but has continued to expand its statistical offerings to be competitive.

For example, as survey research segued into new types of analysis for categorical data. Lazersfeldian elaboration analysis (ca. 1948) beget loglinear analysis (ca. 1980) beget GEE, generalized linear models, the Leiden contributions, etc. So SPSS the company added those kinds of stats. In addition, in support of survey research, SPSS has also added missing value analysis, codebooks, data validation, multiple imputation, raking, etc. So I don’t think that predictive analysis and business requirements have “taken over”.

From: John F Hall [mailto:[hidden email]]
Sent: Saturday, October 24, 2009 10:24 AM
To: ViAnn Beadle; Bruce Weaver; [hidden email]
Cc: Martin Holt
Subject: Tail wagging dog

Viann, Bruce

Car drives driver, cart drives horse, etc., etc?

Does that explain or not?

John

John F Hall

Re: Tail wagging dog

I'm a big fan of Rosenberg's approach (if you don't mind the same tortuous explanation three times over, in translation from the German original): so is Graham Kalton co-author of Moser and Kalton's classic 1971 survey research textbook Survey Methods in Social Investigation. I worked with them both when I was at SSRC in the 1970s.

Some of my tutorials address this approach, but I've never found a way to display the tables in the format required. Students can't be expected to handle macros or R: it's hard enough them coping with social theory and social stats classes (and today in UK no teachers): they need a new SPSS command

elaborate

depvar list = x [ y z ]

/indvarlist = indvar1 by indvar2

/tables = x [ y z ] by indvar1

/x [y,z] by indvar1 by indvar2

/sta epsilon . (or whatever)

After that it's up to them to interpret what's going on in the data.

Do I detect mild paranoia from the SPSS team? One of my ex-SRU researchers has just written a book about psychological attachment to work.

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

From: [hidden email]

To: [hidden email]

Sent: Saturday, October 24, 2009 7:05 PM

Subject: Re: Tail wagging dog

SPSS the product does what you tell it to do so Im not sure that your analogy works.

SPSS the company has never tried to be an evangelist for new and experimental approaches to data analysis but has continued to expand its statistical offerings to be competitive.

For example, as survey research segued into new types of analysis for categorical data. Lazersfeldian elaboration analysis (ca. 1948) beget loglinear analysis (ca. 1980) beget GEE, generalized linear models, the Leiden contributions, etc. So SPSS the company added those kinds of stats. In addition, in support of survey research, SPSS has also added missing value analysis, codebooks, data validation, multiple imputation, raking, etc. So I dont think that predictive analysis and business requirements have taken over.

From: John F Hall [mailto:[hidden email]]
Sent: Saturday, October 24, 2009 10:24 AM
To: ViAnn Beadle; Bruce Weaver; [hidden email]
Cc: Martin Holt
Subject: Tail wagging dog

Viann, Bruce

Car drives driver, cart drives horse, etc., etc?

http://www.phrases.org.uk/meanings/341850.html helps to explain the meaning of the phrase "tail wagging the dog", but a better example is from Angus Campbell's article in ISSJ 1974 about how to develop and maintain research centres.

Quoting someone else's maxim, he states that it is important that administrators in research institutes understand that there are three groups of people more important than administrators, "the research scientists, the research scientists, and the research scientists". I feel the same way about SPSS, statisticians and computer programmers.

I had a similar problem once with an administrator trying to tell academics how to run the department, in particular my course. She got very upset whan I told her it was her job to make my life easier, not the other way round. That's why I always did my own administration in the Survey Research Unit. The only time I allowed an admin person anywhere near a project led to financial disaster.

Probably extreme, but that's what I mean by "tail wagging dog". SPSS is there to do what I tell it, not the other way round, but nowadays predictive analysis and business requirements seem to have taken over from survey research.

Does that explain or not?

John

ajayohri

Re: Tail wagging dog

Regarding Psychological attachment to Which software should we adopt

If X be degree of appropriateness of software to a particular use-
where 0 is totally bad and 1 is perfect for use.

Then the probability p of the software be selected = P/ Q where P is total number of users who Know how to Use software (like R) and Q is total number of users who dont know how to use the Software (like Macros or R)

As the number of users begins to increase
P/Q converges to Integral of X dx

Thanks,

Ajay Ohri

Website- http://decisionstats.com

Graduate Student
University of Tennessee, Knoxville.
Go Vols!

On Sat, Oct 24, 2009 at 1:59 PM, John F Hall <[hidden email]> wrote:

I'm a big fan of Rosenberg's approach (if you don't mind the same tortuous explanation three times over, in translation from the German original): so is Graham Kalton co-author of Moser and Kalton's classic 1971 survey research textbook Survey Methods in Social Investigation. I worked with them both when I was at SSRC in the 1970s.

Some of my tutorials address this approach, but I've never found a way to display the tables in the format required. Students can't be expected to handle macros or R: it's hard enough them coping with social theory and social stats classes (and today in UK no teachers): they need a new SPSS command

elaborate

    depvar list = x [ y z ]

    /indvarlist =   indvar1 by indvar2

    /tables = x [ y z ] by indvar1

    /x [y,z] by indvar1 by indvar2

    /sta epsilon .            (or whatever)

After that it's up to them to interpret what's going on in the data.

Do I detect mild paranoia from the SPSS team? One of my ex-SRU researchers has just written a book about psychological attachment to work.

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

From: [hidden email]

To: [hidden email]

Sent: Saturday, October 24, 2009 7:05 PM

Subject: Re: Tail wagging dog

SPSS the product does what you tell it to do so I’m not sure that your analogy works.

SPSS the company has never tried to be an evangelist for new and experimental approaches to data analysis but has continued to expand its statistical offerings to be competitive.

For example, as survey research segued into new types of analysis for categorical data. Lazersfeldian elaboration analysis (ca. 1948) beget loglinear analysis (ca. 1980) beget GEE, generalized linear models, the Leiden contributions, etc. So SPSS the company added those kinds of stats. In addition, in support of survey research, SPSS has also added missing value analysis, codebooks, data validation, multiple imputation, raking, etc. So I don’t think that predictive analysis and business requirements have “taken over”.

From: John F Hall [mailto:[hidden email]]
Sent: Saturday, October 24, 2009 10:24 AM
To: ViAnn Beadle; Bruce Weaver; [hidden email]
Cc: Martin Holt
Subject: Tail wagging dog

Viann, Bruce

Car drives driver, cart drives horse, etc., etc?

http://www.phrases.org.uk/meanings/341850.html helps to explain the meaning of the phrase "tail wagging the dog", but a better example is from Angus Campbell's article in ISSJ 1974 about how to develop and maintain research centres.

Quoting someone else's maxim, he states that it is important that administrators in research institutes understand that there are three groups of people more important than administrators, "the research scientists, the research scientists, and the research scientists". I feel the same way about SPSS, statisticians and computer programmers.

I had a similar problem once with an administrator trying to tell academics how to run the department, in particular my course. She got very upset whan I told her it was her job to make my life easier, not the other way round. That's why I always did my own administration in the Survey Research Unit. The only time I allowed an admin person anywhere near a project led to financial disaster.

Probably extreme, but that's what I mean by "tail wagging dog". SPSS is there to do what I tell it, not the other way round, but nowadays predictive analysis and business requirements seem to have taken over from survey research.

Does that explain or not?

John

John F Hall

Re: Tail wagging dog

In reply to this post by John F Hall

Oops! Got sent too soon. Here's the link for Christeen George's new book on the psychological contract.

http://www.google.fr/search?hl=en&source=hp&q=+christeen+george+psychological+contract&aq=f&oq=

Another useful English proverb is, "Don't teach Grandma to suck eggs."

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

From: [hidden email]

They need a new SPSS command, something like this:

elaborate

depvar = x

/indvarlist = indvar1 by indvar2

/tables = x by indvar1

/x by indvar1 by indvar2

/sta epsilon .

After that it's up to them to interpret what's going on in the data.

Do I detect mild paranoia from the SPSS team? One of my ex-SRU researchers Dr Christeen George has just published a book about psychological attachment to work.

PS Does anyone know where Graham Kalton is now? He used to be at ISR Ann Arbor or Maryland.

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

From: [hidden email]

To: [hidden email]

Sent: Saturday, October 24, 2009 7:05 PM

Subject: Re: Tail wagging dog

SPSS the product does what you tell it to do so Im not sure that your analogy works.

SPSS the company has never tried to be an evangelist for new and experimental approaches to data analysis but has continued to expand its statistical offerings to be competitive.

For example, as survey research segued into new types of analysis for categorical data. Lazersfeldian elaboration analysis (ca. 1948) beget loglinear analysis (ca. 1980) beget GEE, generalized linear models, the Leiden contributions, etc. So SPSS the company added those kinds of stats. In addition, in support of survey research, SPSS has also added missing value analysis, codebooks, data validation, multiple imputation, raking, etc. So I dont think that predictive analysis and business requirements have taken over.

From: John F Hall [mailto:[hidden email]]
Sent: Saturday, October 24, 2009 10:24 AM
To: ViAnn Beadle; Bruce Weaver; [hidden email]
Cc: Martin Holt
Subject: Tail wagging dog

Viann, Bruce

Car drives driver, cart drives horse, etc., etc?

http://www.phrases.org.uk/meanings/341850.html helps to explain the meaning of the phrase "tail wagging the dog", but a better example is from Angus Campbell's article in ISSJ 1974 about how to develop and maintain research centres.

Quoting someone else's maxim, he states that it is important that administrators in research institutes understand that there are three groups of people more important than administrators, "the research scientists, the research scientists, and the research scientists". I feel the same way about SPSS, statisticians and computer programmers.

I had a similar problem once with an administrator trying to tell academics how to run the department, in particular my course. She got very upset whan I told her it was her job to make my life easier, not the other way round. That's why I always did my own administration in the Survey Research Unit. The only time I allowed an admin person anywhere near a project led to financial disaster.

Probably extreme, but that's what I mean by "tail wagging dog". SPSS is there to do what I tell it, not the other way round, but nowadays predictive analysis and business requirements seem to have taken over from survey research.

Does that explain or not?

John

Art Kendall

Re: Tail wagging dog

In reply to this post by John F Hall

Can you post the link to some table that do what you want to see?

Art Kendall

John F Hall wrote:

I'm a big fan of Rosenberg's approach (if you don't mind the same tortuous explanation three times over, in translation from the German original): so is Graham Kalton co-author of Moser and Kalton's classic 1971 survey research textbook Survey Methods in Social Investigation. I worked with them both when I was at SSRC in the 1970s.

Some of my tutorials address this approach, but I've never found a way to display the tables in the format required. Students can't be expected to handle macros or R: it's hard enough them coping with social theory and social stats classes (and today in UK no teachers): they need a new SPSS command

elaborate

    depvar list = x [ y z ]

    /indvarlist =   indvar1 by indvar2

    /tables = x [ y z ] by indvar1

    /x [y,z] by indvar1 by indvar2

    /sta epsilon .            (or whatever)

After that it's up to them to interpret what's going on in the data.

Do I detect mild paranoia from the SPSS team? One of my ex-SRU researchers has just written a book about psychological attachment to work.

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

From: [hidden email]

To: [hidden email]

Sent: Saturday, October 24, 2009 7:05 PM

Subject: Re: Tail wagging dog

SPSS the product does what you tell it to do so I’m not sure that your analogy works.

SPSS the company has never tried to be an evangelist for new and experimental approaches to data analysis but has continued to expand its statistical offerings to be competitive.

For example, as survey research segued into new types of analysis for categorical data. Lazersfeldian elaboration analysis (ca. 1948) beget loglinear analysis (ca. 1980) beget GEE, generalized linear models, the Leiden contributions, etc. So SPSS the company added those kinds of stats. In addition, in support of survey research, SPSS has also added missing value analysis, codebooks, data validation, multiple imputation, raking, etc. So I don’t think that predictive analysis and business requirements have “taken over”.

From: John F Hall [[hidden email]]
Sent: Saturday, October 24, 2009 10:24 AM
To: ViAnn Beadle; Bruce Weaver; [hidden email]
Cc: Martin Holt
Subject: Tail wagging dog

Viann, Bruce

Car drives driver, cart drives horse, etc., etc?

http://www.phrases.org.uk/meanings/341850.html helps to explain the meaning of the phrase "tail wagging the dog", but a better example is from Angus Campbell's article in ISSJ 1974 about how to develop and maintain research centres.

Quoting someone else's maxim, he states that it is important that administrators in research institutes understand that there are three groups of people more important than administrators, "the research scientists, the research scientists, and the research scientists". I feel the same way about SPSS, statisticians and computer programmers.

I had a similar problem once with an administrator trying to tell academics how to run the department, in particular my course. She got very upset whan I told her it was her job to make my life easier, not the other way round. That's why I always did my own administration in the Survey Research Unit. The only time I allowed an admin person anywhere near a project led to financial disaster.

Probably extreme, but that's what I mean by "tail wagging dog". SPSS is there to do what I tell it, not the other way round, but nowadays predictive analysis and business requirements seem to have taken over from survey research.

Does that explain or not?

John

===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

Art Kendall
Social Research Consultants