Tips on Posting questions to the SPSS list.

[David Marso]

Modified from:
http://word.mvps.org/findhelp/posting.htm

<Edited 09/12/2012>

Original in regular font my mods in italic and bold.
-------
1. The first thing to remember is that nobody is paid to answer questions in the forums. Despite the FUBAR economy most have a day job doing something else. Those who answer do so because they enjoy it or are incorrigible masochists. If you respect that, and try to make it easy and enjoyable or painful for people to answer you, then all the other tips here will seem obvious.
<SNIP>  Please make sure you mention which version of SPSS you are using when you put a question (it's often impossible to answer your question if you don't). It also helps if you mention which operating system you are using. OTOH:  My solutions work for all versions.
Please do not send people direct emails unless they ask for them: it's very rude. And you probably won't get an answer: With ??-million users of SPSS out there, no-one would live long enough to answer all the emails that would result if each SPSS user sent them just one email in their whole lifetime.  In fact I would reply with a 5M file of my cat in drag to every one of them.

2.    Please do not send attachments unless we request them. We normally do not need an attachment to answer your question: a reasonably concise thorough well thought out description is sufficient. <SNIP>

3.    Check to see which forum is most relevant to your problem. Read the questions and answers, and see if they are on a topic similar to your question. It may even be that your question has already been asked and answered recently! If you have made an honest effort to find an appropriate forum, people will do their best to answer you, even if the question is at the edge of the main subject area of the forum.
Well, the only other forum other than the IBM/SPSS forums is the SPAM raped google forum (good luck with that.
4 omitted...(just stuff about not posting to GT 1 forum).

5.    Say which version of SPSS god forbid PASW snicker... you are using. .. <SNIP> see 1.
EXCEPT "we can't see your screen from here..." and eSPSS and InterNeTelepathy are in short supply

 6.   Describe your problem as fully (and concisely -THAT IS NOT PARADOXICAL-) as possible, including where appropriate what you have tried so far when attempting to fix it. This is NOT the place to try to be brief: that's for Twitter, and this is not Twitter!  Where possible, use the terms used within SPSS itself to describe what is happening. Describe what you are trying to achieve overall. Sometimes people concentrate so hard on a particular way of solving the problem, that they don't notice that they are using a hammer to drive screws into the wall!
In other words:  Please take at least as much time formulating and posting as you expect others to take to resolve your issue.  My personal limit on time commitment is about 3-5 minutes!!  If it is going to take me longer then it will take you hours and you should pay me my 70 per hour rate for my consulting time. .

7.    Make the message subject heading descriptive of the question. Saying “SPSS problem” or “Help!!!” isn't going to catch the eye of somebody who might know the answer. “Can't install RCI device in ether orifice” or “Error message when running GLM with 500 variables” is much more informative and useful.
In fact I purposely ignore these types of questions even if I know the answer.


8.    Use your real name, or at least something that looks like a real name. It makes us feel a little stupid to start a reply by saying "Hi MadDog". The forums will automatically anonymise your email: we cannot respond to you by email because we do not have your email address (Sure we do.. Are you paranoid yet?).  Unless your name is Black Rock City Hottie then see 8.

 9.   If nobody answers your question, it is probably because nobody knows the answer. Of course we do.  we are just sadists and holding out on you because the question is hazier than the brown acid at Woodstock .
9b. There is no guarantee that anyone will know the answer, though most questions do get answered. There is also no guarantee regarding response times – an answer will appear when somebody who knows the answer chances to read the question. Putting “URGENT” into a message header will make no difference as to when that will happen.
Putting URGENT in the header is a great way to be completely ignored!


 10.    If your English is not very good, don't worry – nobody is going to laugh at you (but you will not get an answer in yout mother tongue so you had best be able to read English or use Babblefish) . Do your best to explain your problem in English, but also include your question in your native language: chances are one of us can read it. If we have trouble understanding something, we will ask you to explain again (or maybe not).

10a.    On the other hand, if English is your mother tongue, please try to remember to run the spellchecker before posting – and especially, try to ensure there are no errors in your subject line. In a long thread, especially, these can become irritating for other people. ABOVE ALL DON'T SPEAK IN FREAKING IM!!! and don't use all CAPS it is considered *SHOUTING* and is really freaking annoying!!!!!!!

11.    If you have a follow-up question, or a contribution to make to an answer post it back to the forum! Many people may be interested in the additional information. Also, if the person answering your question is not entirely sure about the answer, post back saying whether it worked or not. That helps for when the question comes up again. Forums work well because everybody can see all the information.
 <SNIP MVP /M$ pay to play spiel>
------------------------------------------
AND!!!!
DMM1.  Don't hijack previous threads with an OT/different topic.  There is a NEW THREAD option and replying to a "how do I mutilate my data using the SPSS ginzu option" with a "My table isn't coming out right" or "Python sucks/Macros rule" is bound to bury your question in the bowels of something completely irrelevant to your cause and the right people won't see it.  Personally I use Nabble to view this group and Changed subject lines end up threaded under the original topic.

DMM2.  Don't expect people to do your thinking and your job for you.  If you have been struggling to solve a problem for days or weeks maybe you should hire a competent consultant or find training.  Ultimately it may save you money and a lot of hair.  SPSS is a complex program.  Don't expect t learn it in 2 days (or even 2 years).  I have used it for 30 years and still look at the manual occasionally.

DMM3. RTFM (read the fine manual).  Not literally "read", but BROWSE it for X's sake to get the lay of the land.

DMM4.  Browse the archives to see if anything remotely similar has been asked in say the past 20 years?
OOPS the archives have been truncated to 1996 so 16 years.
------
ADDED 9/12: After the fact.

DMM5.  If you are attempting to perform some sort of unnatural act with strange data please provide some sort of reasonable representation of the existing structure of the data and an example of what you wish to achieve as a result of the data transformation.  Rather than attempting to describe the data in human language provide an actual snippet or mock up of the relevant fields and their role in the problem.

DMM6.  If in the rare event you should receive a cryptic error message or warning from the Oracle of Cthulhu http://en.wikipedia.org/wiki/Cthulhu_Mythos disguised as coming from your SPSS session PLEASE provide the exact error message or warning so you at least will have some hope of receiving a reply.

DMM7.  If you are attempting to have people here do your homework you would be advised to disguise that fact and not quote the homework question verbatim.  Opacity in such matters is critical.

eg Obvious homework question(s):
1.  If a variable is distributed normally with mean 0 and SD 1.  What is the probability that the mean of 10 numbers drawn from this distribution exceeds 1.9645 .

2. What is the probability the mean of sample size 20 lies between -2 and 1.98.

PLONK!!!

(opacity added)
I am new to this list and was wondering how one computed probabilities for a normal distribution in SPSS.
--

[Albert-Jan Roskam]

<snip>

These are the R guidelines: http://www.r-project.org/posting-guide.html. But having mentioned these, R is good example of a list where "bashing" is quite common *). Guidelines are a means, not a goal in themselves. It's useful to have a conscise problem description, with a self-contained example. But esp. for novice users this is easier said than done. I agree with the technical guidelines to avoid html messages and also, to avoid top posting. Bottom posting makes the archive a lot easier to read.

Albert-Jan

*) This is not to say that many, if not the majority of the people on that list are helpful.

=====================
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

[Robert Jones]

A couple more relevant links

http://www.catb.org/~esr/faqs/smart-questions.html - How To Ask Questions The Smart Way

and

http://www.chiark.greenend.org.uk/~sgtatham/bugs.html - How to Report Bugs Effectively

Robert

[Bruce Weaver]

Re top- vs bottom-posting, I (almost) always bottom-post in usenet newsgroups (e.g., the sci.stat.* groups), where that has long been the convention in usenet.  I top-post on this list, because that *seems* to be the convention here.  (I would guess that's because most folks read via e-mail programs that top-post by default.)  

Bruce

Albert-Jan Roskam wrote

<snip>

These are the R guidelines: http://www.r-project.org/posting-guide.html. But having mentioned these, R is good example of a list where "bashing" is quite common *). Guidelines are a means, not a goal in themselves. It's useful to have a conscise problem description, with a self-contained example. But esp. for novice users this is easier said than done. I agree with the technical guidelines to avoid html messages and also, to avoid top posting. Bottom posting makes the archive a lot easier to read.

Albert-Jan

*) This is not to say that many, if not the majority of the people on that list are helpful.

=====================
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

[David Marso]

I invariably top post but include the previous post and re-quote the relevant points I am addressing.
Reading bottom posts requires reading through the everything previous and sometimes impossible to find the current post (especially when it is at the bottom of a deep thread where everyone has re-quoted everything without snipping anything  ).  Try reading SEMNET digest these days ;-)))
---

Bruce Weaver wrote

Re top- vs bottom-posting, I (almost) always bottom-post in usenet newsgroups (e.g., the sci.stat.* groups), where that has long been the convention in usenet.  I top-post on this list, because that *seems* to be the convention here.  (I would guess that's because most folks read via e-mail programs that top-post by default.)  

Bruce

Albert-Jan Roskam wrote

<snip>

These are the R guidelines: http://www.r-project.org/posting-guide.html. But having mentioned these, R is good example of a list where "bashing" is quite common *). Guidelines are a means, not a goal in themselves. It's useful to have a conscise problem description, with a self-contained example. But esp. for novice users this is easier said than done. I agree with the technical guidelines to avoid html messages and also, to avoid top posting. Bottom posting makes the archive a lot easier to read.

Albert-Jan

*) This is not to say that many, if not the majority of the people on that list are helpful.

<SNIP>

[Angie Carrico]

My (hardly ever posting) 2 cents.  I have been a reader of this listserv for well over a year.  I personally would think long and hard before posting a question here for fear of doing it "wrong." I have, in fact, searched for more forgiving avenues for those of us who are less skilled.  But the fact remains that there are some great experts here and I haven't found an alternative.

 

I am with Jon (I think that is who said it): If you don't like how the person phrases the question - don't answer. 

 

I will now return to my silent observation...

 

 

On Wed, Sep 12, 2012 at 3:34 AM, Albert-Jan Roskam <[hidden email]> wrote:

<snip>

These are the R guidelines: http://www.r-project.org/posting-guide.html. But having mentioned these, R is good example of a list where "bashing" is quite common *). Guidelines are a means, not a goal in themselves. It's useful to have a conscise problem description, with a self-contained example. But esp. for novice users this is easier said than done. I agree with the technical guidelines to avoid html messages and also, to avoid top posting. Bottom posting makes the archive a lot easier to read.

Albert-Jan

*) This is not to say that many, if not the majority of the people on that list are helpful.

=====================
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

--
Angela Carrico
Director of Institutional Research
Southwestern Michigan College
58900 Cherry Grove Rd
Dowagiac MI 49047
1-800-456-8675, ext. 1323

[Andy W]

Nice list David and links everyone,

I would just add that frequently the hardest part of figuring out the problem is knowing exactly what the OP's data look like. I recently made a blog post to depict several ways to either make fake data or use the example data sets that come with the software. If you don't like my self-promotion, you can just look at old posts here on the list-serve and see a variety of similar examples.

Although Albert-Jan makes a good point that people just don't have the where-with-all to do this frequently (let alone make a reproducible example of the problem), so it should be viewed as a goal (so they need to be nudged to a varying extent). I agree we should attempt to be cordial, though it is difficult sometimes. See for instance some related discussion about being cordial on the R-help forum. IMO I just try to ignore helpless questions, but perhaps we should keep this thread bookmarked and when such a question occurs reply with a link to this discussion.

I'm ambivalent about html and top/bottom posting. I follow posts via an RSS reader and reply here on the site (long email threads are IMO always impossible to follow). HTML has the benefit that my code snippets I provide can be copy-pasted appropriately without mysterious line breaks or whatever (besides just generally looking nicer when interpreted as html). Similarly top-bottom posting makes no difference when viewing on NABBLE (I believe it shows up on the bottom on NABBLE regardless). SEMNET digest is just impossible to read.

The big deal is though making the question clear and cogent (the other stuff is nice but just fluff compared to that IMO). This is certainly hard though, and many posters just don't take the time (or lack competence) to be cogent and understandable.

Andy

[Martha Hewett]

I agree with this 100%.




From:        Angie Carrico <[hidden email]>
To:        [hidden email]
Date:        09/12/2012 08:11 AM
Subject:        Re: Tips on Posting questions to the SPSS list.
Sent by:        "SPSSX(r) Discussion" <[hidden email]>

---

My (hardly ever posting) 2 cents.  I have been a reader of this listserv for well over a year.  I personally would think long and hard before posting a question here for fear of doing it "wrong." I have, in fact, searched for more forgiving avenues for those of us who are less skilled.  But the fact remains that there are some great experts here and I haven't found an alternative.
 
I am with Jon (I think that is who said it): If you don't like how the person phrases the question - don't answer. 
 
I will now return to my silent observation...
 
 
On Wed, Sep 12, 2012 at 3:34 AM, Albert-Jan Roskam <fomcl@...> wrote:
<snip>

These are the R guidelines: http://www.r-project.org/posting-guide.html. But having mentioned these, R is good example of a list where "bashing" is quite common *). Guidelines are a means, not a goal in themselves. It's useful to have a conscise problem description, with a self-contained example. But esp. for novice users this is easier said than done. I agree with the technical guidelines to avoid html messages and also, to avoid top posting. Bottom posting makes the archive a lot easier to read.

Albert-Jan

*) This is not to say that many, if not the majority of the people on that list are helpful.

=====================
To manage your subscription to SPSSX-L, send a message to
LISTSERV@... (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



--
Angela Carrico
Director of Institutional Research
Southwestern Michigan College
58900 Cherry Grove Rd
Dowagiac MI 49047
1-800-456-8675, ext. 1323

[statisticsdoc]

I agree completely with Angie and Jon.   It is important to create an atmosphere where it is safe for beginners to ask questions.   It would be a great loss to the user community if less experienced users felt intimidated and even bullied when they asked for help.  I would rather have an open list where an occasional user made a questionable request (that can be ignored) than one in which many sensible requests for help went unanswered.

 

Best,

 

Steve Brand

 

www.StatisticsDoc.com

 

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Angie Carrico
Sent: Wednesday, September 12, 2012 9:04 AM
To: [hidden email]
Subject: Re: Tips on Posting questions to the SPSS list.

 

My (hardly ever posting) 2 cents.  I have been a reader of this listserv for well over a year.  I personally would think long and hard before posting a question here for fear of doing it "wrong." I have, in fact, searched for more forgiving avenues for those of us who are less skilled.  But the fact remains that there are some great experts here and I haven't found an alternative.

 

I am with Jon (I think that is who said it): If you don't like how the person phrases the question - don't answer. 

 

I will now return to my silent observation...

 

 

On Wed, Sep 12, 2012 at 3:34 AM, Albert-Jan Roskam <[hidden email]> wrote:

<snip>

These are the R guidelines: http://www.r-project.org/posting-guide.html. But having mentioned these, R is good example of a list where "bashing" is quite common *). Guidelines are a means, not a goal in themselves. It's useful to have a conscise problem description, with a self-contained example. But esp. for novice users this is easier said than done. I agree with the technical guidelines to avoid html messages and also, to avoid top posting. Bottom posting makes the archive a lot easier to read.

Albert-Jan

*) This is not to say that many, if not the majority of the people on that list are helpful.

=====================
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

--
Angela Carrico
Director of Institutional Research
Southwestern Michigan College
58900 Cherry Grove Rd
Dowagiac MI 49047
1-800-456-8675, ext. 1323

[Zuluaga, Juan]

There's the issue about questions that are not really about SPSS but about statistical approaches ("does method Z make sense to answer question Q given this data?").  Are there better email lists to direct those to?

-j

=====================
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

[Bruce Weaver]

It depends on what method Z is.  If Z is a fairly standard method that has been around for a while, the sci.stat.* usenet groups (where * = consult, edu or math) have been pretty good places to get help over the years.  (Some might tell you they have more spam than they used to, but depending on how you access them, you can do a pretty good job of filtering it out.)  For medical statistics, the Medstats group (in Google Groups) is pretty good.  There are also other mailing lists for specific methods (e.g., multilevel models, classification methods), but I don't have much (if any) first-hand knowledge of those.

HTH.

Zuluaga, Juan wrote

There's the issue about questions that are not really about SPSS but about statistical approaches ("does method Z make sense to answer question Q given this data?").  Are there better email lists to direct those to?

-j

=====================
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

[Jon K Peck]

One site that I like for statistical questions is Cross Validated
stats.stackexchange.com

Statistical questions and answers there are generally pretty high quality.

Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
[hidden email]
new phone: [phone]




From:        Bruce Weaver <[hidden email]>
To:        [hidden email]
Date:        09/13/2012 08:06 AM
Subject:        Re: [SPSSX-L] Tips on Posting questions to the SPSS list.
Sent by:        "SPSSX(r) Discussion" <[hidden email]>

---

It depends on what method Z is.  If Z is a fairly standard method that has
been around for a while, the sci.stat.* usenet groups (where * = consult,
edu or math) have been pretty good places to get help over the years.  (Some
might tell you they have more spam than they used to, but depending on how
you access them, you can do a pretty good job of filtering it out.)  For
medical statistics, the Medstats group (in Google Groups) is pretty good.
There are also other mailing lists for specific methods (e.g., multilevel
models, classification methods), but I don't have much (if any) first-hand
knowledge of those.

HTH.



Zuluaga, Juan wrote
>
> There's the issue about questions that are not really about SPSS but about
> statistical approaches ("does method Z make sense to answer question Q
> given this data?").  Are there better email lists to direct those to?
>
> -j
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> LISTSERV@.UGA (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
>




-----
--
Bruce Weaver
[hidden email]
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.

--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Tips-on-Posting-questions-to-the-SPSS-list-tp5715038p5715064.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

[Art Kendall]

For top vs bottom posting, I try to follow whatever seems to be prevalent on a particular list.

When messages are threaded (on some software this is called putting them in "discussions"), my personal preference is for top posting. I have often just read the previous posting.

Without having done any checking I have the impression that most webmail and email software can display HTML.
I wonder how many people have a problem with HTML other than when dealing with the UGA archives? I think that it is unfortunate that UGA is running a 90's version of LISTSERV.  This messes up retrievals from the archives.
However, Bruce has pointed out that one can use NABBLE to see formatted returns from the archives.



if a post is getting long , it is helpful to snip out irrelevant information, e.g.,

=====================
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

On 9/12/2012 7:58 AM, Bruce Weaver wrote:

Re top- vs bottom-posting, I (almost) always bottom-post in usenet newsgroups
(e.g., the sci.stat.* groups), where that has long been the convention in
usenet.  I top-post on this list, because that *seems* to be the convention
here.  (I would guess that's because most folks read via e-mail programs
that top-post by default.)

Bruce



Albert-Jan Roskam wrote

<snip>

These are the R guidelines: http://www.r-project.org/posting-guide.html.
But having mentioned these, R is good example of a list where "bashing" is
quite common *). Guidelines are a means, not a goal in themselves. It's
useful to have a conscise problem description, with a self-contained
example. But esp. for novice users this is easier said than done. I agree
with the technical guidelines to avoid html messages and also, to avoid
top posting. Bottom posting makes the archive a lot easier to read.

Albert-Jan

*) This is not to say that many, if not the majority of the people on that
list are helpful.

=====================
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

-----
--
Bruce Weaver
[hidden email]
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.

--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Tips-on-Posting-questions-to-the-SPSS-list-tp5715038p5715041.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

===================== 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

[bilal khalid]

This is Bilal Khalid. I am doing research work in the field of Construction Engineering and Management. The Statistical analyses techniques which I am using are confirmatory factor analysis (CFA) along with structure equation Modelling (SEM). Initially I started with PCA to shrink my factor List (I had 3 Principle categories in which 26 sub categories are there along with 106 variables/statements in sub categories). Finally i have 9 factors in which 29 variables/statements are there. Now, Project my supervisor suggests me to go with Principle Axis factoring technique because he is not satisfied with the final outputs of PCA. My 1st question is, should i go with PAF again to validate my factor results? if yes then kindly mention here some link which can help me in understanding the basics of PAF and also the threshold values of outputs of PAF.

Regards,
Bilal Khalid
Urban & Infrastructure Engineering department,
NED University of Engineering & Technology.

[David Marso]

Notice you hijacked the thread titled "Tips on Posting questions to the SPSS list."....
Please begin a new topic after deleting this unrelated question from the thread!
--
"AND!!!!
DMM1.  Don't hijack previous threads with an OT/different topic.  There is a NEW THREAD option and replying to a "how do I mutilate my data using the SPSS ginzu option" with a "My table isn't coming out right" or "Python sucks/Macros rule" is bound to bury your question in the bowels of something completely irrelevant to your cause and the right people won't see it.  Personally I use Nabble to view this group and Changed subject lines end up threaded under the original topic. "

I suspect your question is rather general and tangential to SPSS.  PCA vs PAF vs SEM?
I'll bet if you Google these terms you will find more than enough to keep you busy for months.
---

bilal khalid wrote

This is Bilal Khalid. I am doing research work in the field of Construction Engineering and Management. The Statistical analyses techniques which I am using are confirmatory factor analysis (CFA) along with structure equation Modelling (SEM). Initially I started with PCA to shrink my factor List (I had 3 Principle categories in which 26 sub categories are there along with 106 variables/statements in sub categories). Finally i have 9 factors in which 29 variables/statements are there. Now, Project my supervisor suggests me to go with Principle Axis factoring technique because he is not satisfied with the final outputs of PCA. My 1st question is, should i go with PAF again to validate my factor results? if yes then kindly mention here some link which can help me in understanding the basics of PAF and also the threshold values of outputs of PAF.

Regards,
Bilal Khalid
Urban & Infrastructure Engineering department,
NED University of Engineering & Technology.

[Maguin, Eugene]

Bilal,
The practical answer is that you should do what your supervisor requests. I understand that faculty-student relationship expectations at your school may be different in some ways than would be true in the US but he's the teacher so he should be able to coherently (and correctly) explain his reasoning, which means you need to have some background. Why haven't you googled 'factor analysis'? Other, smarter, people have written book on factor analysis.

Gene Maguin


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of bilal khalid
Sent: Wednesday, September 19, 2012 3:19 AM
To: [hidden email]
Subject: Re: Tips on Posting questions to the SPSS list.

This is Bilal Khalid. I am doing research work in the field of Construction Engineering and Management. The Statistical analyses techniques which I am using are confirmatory factor analysis (CFA) along with structure equation Modelling (SEM). Initially I started with PCA to shrink my factor List (I had 3 Principle categories in which 26 sub categories are there along with
106 variables/statements in sub categories). Finally i have 9 factors in which 29 variables/statements are there. Now, Project my supervisor suggests me to go with Principle Axis factoring technique because he is not satisfied with the final outputs of PCA. My 1st question is, should i go with PAF again to validate my factor results? if yes then kindly mention here some link which can help me in understanding the basics of PAF and also the threshold values of outputs of PAF.

Regards,
Bilal Khalid
Urban & Infrastructure Engineering department, NED University of Engineering & Technology.



--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Tips-on-Posting-questions-to-the-SPSS-list-tp5715038p5715155.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

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

It is not possible to give a very complete answer without more information.

The substantive nature of you research makes a lot of difference.
What is the purpose of your analysis?
What are you trying to find out or do with the factor analysis?
Are there preexisting scales (aka, summary measures, composite indices) that came from previous studies and you want to see if they stand up in another sample of population?
What is a case in your data?
How were the cases chosen? How many are there? Do they come from different (sub)populations?
What are your three broad categories of items?
What are your 26 subcategories  of items?
How did you get you items?

What do you mean by 

the
threshold values of outputs of PAF.

-------
A brief distinction between PCA and PAF.
Principal Axis Factoring is the most common kind of factor analysis used in creating summative scales. (Other purposes vary in what kind of factor analysis is used.)
The idea is that an item variable has three parts to its total variance:
reliable/consistent variance due to a common latent or underlying construct
variance due to the uniqueness to item and
random error (noise) variance.
Total variance = common variance + unique variance + error variance.
For example, spelling ability might be measured via the common variance within a set of items. The more words that are spelled correctly the higher spelling ability is considered to be.
Each word (item) has unique characteristics due to the particular word chosen.
The items can be considered as repeated rough measures of a construct.  The construct is considered to be better represented by the sum or mean of the items.

Principal Axis Factoring aims to account for the common variance in a set of items.  It tries to find meaningful artificial dimensions (factors)  that account for as much of the common variance as it can with as few factors as is practical.

On the other hand Principal components factor analysis tries to account  for the total variance. It tries to find meaningful artificial dimensions (factors)  that account for as much of the total variance as it can with as few factors as is practical.

PAF uses estimates of the item's reliability on the diagonal of the matrix that is being analyzed. That is what the procedure tries to account for in a parsimonious manner. These numbers are less than 1.00 because they represent only the common variance.

PCA uses 1.00 on the diagonal of the matrix to represent the total variance.


On 9/19/2012 3:18 AM, bilal khalid wrote:

This is Bilal Khalid. I am doing research work in the field of Construction
Engineering and Management. The Statistical analyses techniques which I am
using are confirmatory factor analysis (CFA) along with structure equation
Modelling (SEM). Initially I started with PCA to shrink my factor List (I
had 3 Principle categories in which 26 sub categories are there along with
106 variables/statements in sub categories). Finally i have 9 factors in
which 29 variables/statements are there. Now, Project my supervisor suggests
me to go with Principle Axis factoring technique because he is not satisfied
with the final outputs of PCA. My 1st question is, should i go with PAF
again to validate my factor results? if yes then kindly mention here some
link which can help me in understanding the basics of PAF and also the
threshold values of outputs of PAF.

Regards,
Bilal Khalid
Urban & Infrastructure Engineering department,
NED University of Engineering & Technology.



--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Tips-on-Posting-questions-to-the-SPSS-list-tp5715038p5715155.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

===================== 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

[David Marso]

After reviewing recent activity!!!
BUMP!!!!
Need I say more?
About 3 posts away from a 3 mo LOA!!!
Maybe this is TLTR and I should do a Cliff Notes version for the ADD peeps?
----

David Marso wrote

Modified from:
http://word.mvps.org/findhelp/posting.htm

<Edited 09/12/2012>

Original in regular font my mods in italic and bold.
-------
1. The first thing to remember is that nobody is paid to answer questions in the forums. Despite the FUBAR economy most have a day job doing something else. Those who answer do so because they enjoy it or are incorrigible masochists. If you respect that, and try to make it easy and enjoyable or painful for people to answer you, then all the other tips here will seem obvious.
<SNIP>  Please make sure you mention which version of SPSS you are using when you put a question (it's often impossible to answer your question if you don't). It also helps if you mention which operating system you are using. OTOH:  My solutions work for all versions.
Please do not send people direct emails unless they ask for them: it's very rude. And you probably won't get an answer: With ??-million users of SPSS out there, no-one would live long enough to answer all the emails that would result if each SPSS user sent them just one email in their whole lifetime.  In fact I would reply with a 5M file of my cat in drag to every one of them.

2.    Please do not send attachments unless we request them. We normally do not need an attachment to answer your question: a reasonably concise thorough well thought out description is sufficient. <SNIP>

3.    Check to see which forum is most relevant to your problem. Read the questions and answers, and see if they are on a topic similar to your question. It may even be that your question has already been asked and answered recently! If you have made an honest effort to find an appropriate forum, people will do their best to answer you, even if the question is at the edge of the main subject area of the forum.
Well, the only other forum other than the IBM/SPSS forums is the SPAM raped google forum (good luck with that.
4 omitted...(just stuff about not posting to GT 1 forum).

5.    Say which version of SPSS god forbid PASW snicker... you are using. .. <SNIP> see 1.
EXCEPT "we can't see your screen from here..." and eSPSS and InterNeTelepathy are in short supply

 6.   Describe your problem as fully (and concisely -THAT IS NOT PARADOXICAL-) as possible, including where appropriate what you have tried so far when attempting to fix it. This is NOT the place to try to be brief: that's for Twitter, and this is not Twitter!  Where possible, use the terms used within SPSS itself to describe what is happening. Describe what you are trying to achieve overall. Sometimes people concentrate so hard on a particular way of solving the problem, that they don't notice that they are using a hammer to drive screws into the wall!
In other words:  Please take at least as much time formulating and posting as you expect others to take to resolve your issue.  My personal limit on time commitment is about 3-5 minutes!!  If it is going to take me longer then it will take you hours and you should pay me my 70 per hour rate for my consulting time. .

7.    Make the message subject heading descriptive of the question. Saying “SPSS problem” or “Help!!!” isn't going to catch the eye of somebody who might know the answer. “Can't install RCI device in ether orifice” or “Error message when running GLM with 500 variables” is much more informative and useful.
In fact I purposely ignore these types of questions even if I know the answer.


8.    Use your real name, or at least something that looks like a real name. It makes us feel a little stupid to start a reply by saying "Hi MadDog". The forums will automatically anonymise your email: we cannot respond to you by email because we do not have your email address (Sure we do.. Are you paranoid yet?).  Unless your name is Black Rock City Hottie then see 8.

 9.   If nobody answers your question, it is probably because nobody knows the answer. Of course we do.  we are just sadists and holding out on you because the question is hazier than the brown acid at Woodstock .
9b. There is no guarantee that anyone will know the answer, though most questions do get answered. There is also no guarantee regarding response times – an answer will appear when somebody who knows the answer chances to read the question. Putting “URGENT” into a message header will make no difference as to when that will happen.
Putting URGENT in the header is a great way to be completely ignored!


 10.    If your English is not very good, don't worry – nobody is going to laugh at you (but you will not get an answer in yout mother tongue so you had best be able to read English or use Babblefish) . Do your best to explain your problem in English, but also include your question in your native language: chances are one of us can read it. If we have trouble understanding something, we will ask you to explain again (or maybe not).

10a.    On the other hand, if English is your mother tongue, please try to remember to run the spellchecker before posting – and especially, try to ensure there are no errors in your subject line. In a long thread, especially, these can become irritating for other people. ABOVE ALL DON'T SPEAK IN FREAKING IM!!! and don't use all CAPS it is considered *SHOUTING* and is really freaking annoying!!!!!!!

11.    If you have a follow-up question, or a contribution to make to an answer post it back to the forum! Many people may be interested in the additional information. Also, if the person answering your question is not entirely sure about the answer, post back saying whether it worked or not. That helps for when the question comes up again. Forums work well because everybody can see all the information.
 <SNIP MVP /M$ pay to play spiel>
------------------------------------------
AND!!!!
DMM1.  Don't hijack previous threads with an OT/different topic.  There is a NEW THREAD option and replying to a "how do I mutilate my data using the SPSS ginzu option" with a "My table isn't coming out right" or "Python sucks/Macros rule" is bound to bury your question in the bowels of something completely irrelevant to your cause and the right people won't see it.  Personally I use Nabble to view this group and Changed subject lines end up threaded under the original topic.

DMM2.  Don't expect people to do your thinking and your job for you.  If you have been struggling to solve a problem for days or weeks maybe you should hire a competent consultant or find training.  Ultimately it may save you money and a lot of hair.  SPSS is a complex program.  Don't expect t learn it in 2 days (or even 2 years).  I have used it for 30 years and still look at the manual occasionally.

DMM3. RTFM (read the fine manual).  Not literally "read", but BROWSE it for X's sake to get the lay of the land.

DMM4.  Browse the archives to see if anything remotely similar has been asked in say the past 20 years?
OOPS the archives have been truncated to 1996 so 16 years.
------
ADDED 9/12: After the fact.

DMM5.  If you are attempting to perform some sort of unnatural act with strange data please provide some sort of reasonable representation of the existing structure of the data and an example of what you wish to achieve as a result of the data transformation.  Rather than attempting to describe the data in human language provide an actual snippet or mock up of the relevant fields and their role in the problem.

DMM6.  If in the rare event you should receive a cryptic error message or warning from the Oracle of Cthulhu http://en.wikipedia.org/wiki/Cthulhu_Mythos disguised as coming from your SPSS session PLEASE provide the exact error message or warning so you at least will have some hope of receiving a reply.

DMM7.  If you are attempting to have people here do your homework you would be advised to disguise that fact and not quote the homework question verbatim.  Opacity in such matters is critical.

eg Obvious homework question(s):
1.  If a variable is distributed normally with mean 0 and SD 1.  What is the probability that the mean of 10 numbers drawn from this distribution exceeds 1.9645 .

2. What is the probability the mean of sample size 20 lies between -2 and 1.98.

PLONK!!!

(opacity added)
I am new to this list and was wondering how one computed probabilities for a normal distribution in SPSS.
--

[Art Kendall]

Please clarify your post.
On 1/17/2013 11:51 PM, David Marso wrote:

After reviewing recent activity!!!
BUMP!!!!
Need I say more?
About 3 posts away from a 3 mo LOA!!!
Maybe this is TLTR and I should do a Cliff Notes version for the ADD peeps?
----

David Marso wrote

Modified from:
http://word.mvps.org/findhelp/posting.htm
<Edited 09/12/2012>
Original in regular font my mods in italic and bold.
-------
1. The first thing to remember is that nobody is paid to answer questions
in the forums.

/

Despite the FUBAR economy most

/

 have a day job doing something else. Those who answer do so because they
enjoy it

/

or are incorrigible masochists

/

. If you respect that, and try to make it easy and enjoyable

/

or painful

/

 for people to answer you, then all the other tips here will seem obvious.
<SNIP>
  Please make sure you mention which version of SPSS you are using when
you put a question (it's often impossible to answer your question if you
don't). It also helps if you mention which operating system you are using.

/

OTOH:  My solutions work for

*

 all

*

versions.

/

*

Please do not send people direct emails

*

 unless they ask for them: it's very rude. And you probably won't get an
answer: With

/

??

/

-million users of SPSS out there, no-one would live long enough to answer
all the emails that would result if each Word user sent them just one
email in their whole lifetime.

/

In fact I would reply with a 5M file of my cat in drag to every one of
them

/

.

2.    Please do not send attachments unless we request them. We normally
do not need an attachment to answer your question:

/

a reasonably concise thorough well thought out

/

 description is sufficient.

*

<SNIP>

*

3.    Check to see which forum is most relevant to your problem. Read the
questions and answers, and see if they are on a topic similar to your
question. It may even be that your question has already been asked and
answered recently! If you have made an honest effort to find an
appropriate forum, people will do their best to answer you, even if the
question is at the edge of the main subject area of the forum.

/

Well, the only other forum other than the IBM/SPSS forums is the SPAM
raped google forum (good luck with that.

/

4 omitted...(just stuff about not posting to GT 1 forum).

5.    Say which version of SPSS

/

god forbid PASW snicker...

/

 you are using. ..
<SNIP>
 see 1.
EXCEPT "we can't see your screen from here..." and eSPSS and
InterNeTelepathy are in short supply

 6.   Describe your problem as fully

/

(and concisely -THAT IS

*

NOT

*

PARADOXICAL-)

/

 as possible, including where appropriate what you have tried so far when
attempting to fix it. This is NOT the place to try to be brief: that's for
Twitter, and this is not Twitter!  Where possible, use the terms used
within SPSS itself to describe what is happening. Describe what you are
trying to achieve overall. Sometimes people concentrate so hard on a
particular way of solving the problem, that they don't notice that they
are using a hammer to drive screws into the wall!

/

In other words:  Please take at least as much time formulating and posting
as you expect others to take to resolve your issue.  My personal limit on
time commitment is about 3-5 minutes!!  If it is going to take me longer
then it will take you hours and you should pay me my 70 per hour rate for
my consulting time.

/

.

7.    Make the message subject heading descriptive of the question. Saying
“SPSS problem” or “Help!!!” isn't going to catch the eye of somebody who
might know the answer. “

/

Can't install RCI device in ether orifice

/

” or “

/

Error message when running GLM with

*

500

*

variables”

/

is much more informative and useful.

/

In fact I purposely ignore these types of questions even if I know the
answer.

/

8.    Use your real name, or at least something that looks like a real
name. It makes us feel a little stupid to start a reply by saying "Hi
MadDog". The forums will automatically anonymise your email: we cannot
respond to you by email because we do not have your email address

/

(Sure we do.. Are you paranoid yet?)

/

.

/

Unless your name is Black Rock City Hottie then see 8.

/

 9.   If nobody answers your question, it is probably because nobody knows
the answer.

/

Of course we do.  we are just sadists and holding out on you because the
question is hazier than the brown acid at Woodstock .

/

9b. There is no guarantee that anyone will know the answer, though most
questions do get answered. There is also no guarantee regarding response
times – an answer will appear when somebody who knows the answer chances
to read the question. Putting “URGENT” into a message header will make no
difference as to when that will happen.

/

Putting URGENT in the header is a great way to be completely ignored!

/

 10.    If your English is not very good, don't worry – nobody is going to
laugh at you

/

(but you will not get an answer in yout mother tongue so you had best be
able to read English or use Babblefish)

/

 . Do your best to explain your problem in English, but also include your
question in your native language: chances are one of us can read it. If we
have trouble understanding something, we will ask you to explain again

/

 (or maybe not).

/

10a.    On the other hand, if English is your mother tongue, please try to
remember to run the spellchecker before posting – and especially, try to
ensure there are no errors in your subject line. In a long thread,
especially, these can become irritating for other people.

/

ABOVE ALL DON'T SPEAK IN FREAKING IM!!! and don't use all CAPS it is
considered *SHOUTING* and is really freaking annoying!!!!!!!

/

11.    If you have a follow-up question, or a contribution to make to an
answer post it back to the forum! Many people may be interested in the
additional information. Also, if the person answering your question is not
entirely sure about the answer, post back saying whether it worked or not.
That helps for when the question comes up again. Forums work well because
everybody can see all the information.

<SNIP MVP /M$ pay to play spiel>
------------------------------------------
AND!!!!

/

DMM1.  Don't hijack previous threads with an OT/different topic.  There is
a NEW THREAD option and replying to a "how do I mutilate my data using the
SPSS ginzu option" with a "My table isn't coming out right" or "Python
sucks/Macros rule" is bound to bury your question in the bowels of
something completely irrelevant to your cause and the right people won't
see it.  Personally I use Nabble to view this group and Changed subject
lines end up threaded under the original topic.

DMM2.  Don't expect people to do your thinking and your job for you.  If
you have been struggling to solve a problem for days or weeks maybe you
should hire a competent consultant or find training.  Ultimately it may
save you money and a lot of hair.  SPSS is a complex program.  Don't
expect t learn it in 2 days (or even 2 years).  I have used it for 30
years and still look at the manual occasionally.

DMM3. RTFM (read the fine manual).  Not literally "read", but BROWSE it
for X's sake to get the lay of the land.

DMM4.  Browse the archives to see if anything remotely similar has been
asked in say the past 20 years?
OOPS the archives have been truncated to 1996 so 16 years.

/

------
ADDED 9/12: After the fact.

DMM5.  If you are attempting to perform some sort of unnatural act with
strange data please provide some sort of reasonable representation of the
existing structure of the data and an example of what you wish to achieve
as a result of the data transformation.  Rather than attempting to
describe the data in human language provide an actual snippet or mock up
of the relevant fields and their role in the problem.

DMM6.  If in the rare event you should receive a cryptic error message or
warning from the Oracle of Cthulhu
http://en.wikipedia.org/wiki/Cthulhu_Mythos disguised as coming from your
SPSS session PLEASE provide the exact error message or warning so you at
least will have some hope of receiving a reply.

DMM7.  If you are attempting to have people here do your homework you
would be advised to disguise that fact and not quote the homework question
verbatim.  Opacity in such matters is critical.

eg Obvious homework question(s):
1.  If a variable is distributed normally with mean 0 and SD 1.  What is
the probability that the mean of 10 numbers drawn from this distribution
exceeds 1.9645 .

2. What is the probability the mean of sample size 20 lies between -2 and
1.98.

PLONK!!!

(opacity added)
I am new to this list and was wondering how one computed probabilities for
a normal distribution in SPSS.
--

-----
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me.
--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Tips-on-Posting-questions-to-the-SPSS-list-tp5715038p5717492.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

===================== 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

[Bruce Weaver]

I'll attempt to translate.  ;-)

After reviewing recent activity!!!
BUMP!!!!
Need I say more?

A lot of recent posters have obviously not considered the advice in the "Tips on Posting questions to the SPSS list" thread.  BUMP = see http://spssx-discussion.1045642.n5.nabble.com/Tips-on-Posting-questions-to-the-SPSS-list-td5715038.html.

About 3 posts away from a 3 mo LOA!!!

LOA = Leave of Absence

Maybe this is TLTR and I should do a Cliff Notes version for the ADD peeps?


ADD = Attention Deficit Disorder.
I haven't worked out what TLTR is yet.  

HTH.

Art Kendall wrote

Please clarify your post.
      Art Kendall
Social Research Consultants
      On 1/17/2013 11:51 PM, David Marso wrote:
   
   
      After reviewing recent activity!!!
BUMP!!!!
Need I say more?
About 3 posts away from a 3 mo LOA!!!
Maybe this is TLTR and I should do a Cliff Notes version for the ADD peeps?
----

David Marso wrote

     
        Modified from:
http://word.mvps.org/findhelp/posting.htm 
<Edited 09/12/2012>
Original in regular font my mods in italic and bold.
-------
1. The first thing to remember is that nobody is paid to answer questions
in the forums.

     
      /

     
        Despite the FUBAR economy most

     
      /

     
         have a day job doing something else. Those who answer do so because they
enjoy it

     
      /

     
        or are incorrigible masochists

     
      /

     
        . If you respect that, and try to make it easy and enjoyable

     
      /

     
        or painful

     
      /

     
         for people to answer you, then all the other tips here will seem obvious.
<SNIP>
  Please make sure you mention which version of SPSS you are using when
you put a question (it's often impossible to answer your question if you
don't). It also helps if you mention which operating system you are using.

     
      /

     
        OTOH:  My solutions work for

     
      *

     
         all

     
      *

     
        versions.

     
      /

     
       

     
      *

     
        Please do not send people direct emails

     
      *

     
         unless they ask for them: it's very rude. And you probably won't get an
answer: With

     
      /

     
        ??

     
      /

     
        -million users of SPSS out there, no-one would live long enough to answer
all the emails that would result if each Word user sent them just one
email in their whole lifetime.

     
      /

     
        In fact I would reply with a 5M file of my cat in drag to every one of
them

     
      /

     
        .

2.    Please do not send attachments unless we request them. We normally
do not need an attachment to answer your question:

     
      /

     
        a reasonably concise thorough well thought out

     
      /

     
         description is sufficient.

     
      *

     
        <SNIP>

     
      *

     
       
3.    Check to see which forum is most relevant to your problem. Read the
questions and answers, and see if they are on a topic similar to your
question. It may even be that your question has already been asked and
answered recently! If you have made an honest effort to find an
appropriate forum, people will do their best to answer you, even if the
question is at the edge of the main subject area of the forum.

     
      /

     
        Well, the only other forum other than the IBM/SPSS forums is the SPAM
raped google forum (good luck with that.

     
      /

     
        4 omitted...(just stuff about not posting to GT 1 forum).

5.    Say which version of SPSS

     
      /

     
        god forbid PASW snicker...

     
      /

     
         you are using. ..
<SNIP>
 see 1.
EXCEPT "we can't see your screen from here..." and eSPSS and
InterNeTelepathy are in short supply

 6.   Describe your problem as fully

     
      /

     
        (and concisely -THAT IS

     
      *

     
        NOT

     
      *

     
        PARADOXICAL-)

     
      /

     
         as possible, including where appropriate what you have tried so far when
attempting to fix it. This is NOT the place to try to be brief: that's for
Twitter, and this is not Twitter!  Where possible, use the terms used
within SPSS itself to describe what is happening. Describe what you are
trying to achieve overall. Sometimes people concentrate so hard on a
particular way of solving the problem, that they don't notice that they
are using a hammer to drive screws into the wall!

     
      /

     
        In other words:  Please take at least as much time formulating and posting
as you expect others to take to resolve your issue.  My personal limit on
time commitment is about 3-5 minutes!!  If it is going to take me longer
then it will take you hours and you should pay me my 70 per hour rate for
my consulting time.

     
      /

     
        .

7.    Make the message subject heading descriptive of the question. Saying
“SPSS problem” or “Help!!!” isn't going to catch the eye of somebody who
might know the answer. “

     
      /

     
        Can't install RCI device in ether orifice

     
      /

     
        ” or “

     
      /

     
        Error message when running GLM with

     
      *

     
        500

     
      *

     
        variables”

     
      /

     
        is much more informative and useful.

     
      /

     
        In fact I purposely ignore these types of questions even if I know the
answer.

     
      /

     
       
8.    Use your real name, or at least something that looks like a real
name. It makes us feel a little stupid to start a reply by saying "Hi
MadDog". The forums will automatically anonymise your email: we cannot
respond to you by email because we do not have your email address

     
      /

     
        (Sure we do.. Are you paranoid yet?)

     
      /

     
        .

     
      /

     
        Unless your name is Black Rock City Hottie then see 8.

     
      /

     
       
 9.   If nobody answers your question, it is probably because nobody knows
the answer.

     
      /

     
        Of course we do.  we are just sadists and holding out on you because the
question is hazier than the brown acid at Woodstock .

     
      /

     
        9b. There is no guarantee that anyone will know the answer, though most
questions do get answered. There is also no guarantee regarding response
times – an answer will appear when somebody who knows the answer chances
to read the question. Putting “URGENT” into a message header will make no
difference as to when that will happen.

     
      /

     
        Putting URGENT in the header is a great way to be completely ignored!

     
      /

     
       
 10.    If your English is not very good, don't worry – nobody is going to
laugh at you

     
      /

     
        (but you will not get an answer in yout mother tongue so you had best be
able to read English or use Babblefish)

     
      /

     
         . Do your best to explain your problem in English, but also include your
question in your native language: chances are one of us can read it. If we
have trouble understanding something, we will ask you to explain again

     
      /

     
         (or maybe not).

     
      /

     
       
10a.    On the other hand, if English is your mother tongue, please try to
remember to run the spellchecker before posting – and especially, try to
ensure there are no errors in your subject line. In a long thread,
especially, these can become irritating for other people.

     
      /

     
        ABOVE ALL DON'T SPEAK IN FREAKING IM!!! and don't use all CAPS it is
considered *SHOUTING* and is really freaking annoying!!!!!!!

     
      /

     
       
11.    If you have a follow-up question, or a contribution to make to an
answer post it back to the forum! Many people may be interested in the
additional information. Also, if the person answering your question is not
entirely sure about the answer, post back saying whether it worked or not.
That helps for when the question comes up again. Forums work well because
everybody can see all the information.

<SNIP MVP /M$ pay to play spiel>
------------------------------------------
AND!!!!

     
      /

     
        DMM1.  Don't hijack previous threads with an OT/different topic.  There is
a NEW THREAD option and replying to a "how do I mutilate my data using the
SPSS ginzu option" with a "My table isn't coming out right" or "Python
sucks/Macros rule" is bound to bury your question in the bowels of
something completely irrelevant to your cause and the right people won't
see it.  Personally I use Nabble to view this group and Changed subject
lines end up threaded under the original topic.

DMM2.  Don't expect people to do your thinking and your job for you.  If
you have been struggling to solve a problem for days or weeks maybe you
should hire a competent consultant or find training.  Ultimately it may
save you money and a lot of hair.  SPSS is a complex program.  Don't
expect t learn it in 2 days (or even 2 years).  I have used it for 30
years and still look at the manual occasionally.

DMM3. RTFM (read the fine manual).  Not literally "read", but BROWSE it
for X's sake to get the lay of the land.

DMM4.  Browse the archives to see if anything remotely similar has been
asked in say the past 20 years?
OOPS the archives have been truncated to 1996 so 16 years.

     
      /

     
        ------
ADDED 9/12: After the fact.

DMM5.  If you are attempting to perform some sort of unnatural act with
strange data please provide some sort of reasonable representation of the
existing structure of the data and an example of what you wish to achieve
as a result of the data transformation.  Rather than attempting to
describe the data in human language provide an actual snippet or mock up
of the relevant fields and their role in the problem.

DMM6.  If in the rare event you should receive a cryptic error message or
warning from the Oracle of Cthulhu
http://en.wikipedia.org/wiki/Cthulhu_Mythos  disguised as coming from your
SPSS session PLEASE provide the exact error message or warning so you at
least will have some hope of receiving a reply.

DMM7.  If you are attempting to have people here do your homework you
would be advised to disguise that fact and not quote the homework question
verbatim.  Opacity in such matters is critical.

eg Obvious homework question(s):
1.  If a variable is distributed normally with mean 0 and SD 1.  What is
the probability that the mean of 10 numbers drawn from this distribution
exceeds 1.9645 .

2. What is the probability the mean of sample size 20 lies between -2 and
1.98.

PLONK!!!

(opacity added)
I am new to this list and was wondering how one computed probabilities for
a normal distribution in SPSS.
--

     
     




-----
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me.
--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Tips-on-Posting-questions-to-the-SPSS-list-tp5715038p5717492.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


   
   
 


=====================
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