Student Edition

classic Classic list List threaded Threaded
14 messages Options
Reply | Threaded
Open this post in threaded view
|

Student Edition

Jessica Pola
Do you offer a 3 month subscription of the SPSS Software.  I am currently taking a Social Statistics course and I need this program only for this fall semester. The price for the program is out of my price range.  Due to covid19 I'm struggling financially.  Please let me know if you offer an incentive or student discount.

=====================
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
Reply | Threaded
Open this post in threaded view
|

Re: Student Edition

Bruce Weaver
Administrator
Hello Jessica.  I assume that under normal conditions (i.e., no pandemic),
you would have access to SPSS in on-campus computer labs.  My own university
has set up virtual computer labs to deal with this problem:

https://www.lakeheadu.ca/faculty-and-staff/departments/services/helpdesk/computer-labs/virtual-computer-labs

You might want to check with your IT Dept to see if they are doing something
similar.  

HTH.


Jessica Pola wrote
> Do you offer a 3 month subscription of the SPSS Software.  I am currently
> taking a Social Statistics course and I need this program only for this
> fall semester. The price for the program is out of my price range.  Due to
> covid19 I'm struggling financially.  Please let me know if you offer an
> incentive or student discount.
>
> =====================
> 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.

--
Sent from: http://spssx-discussion.1045642.n5.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
--
Bruce Weaver
bweaver@lakeheadu.ca
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.
Reply | Threaded
Open this post in threaded view
|

Re: Student Edition

Kornbrot, Diana
Typical higher ed license enable students and staff to use off campus via VPN
But you might enjoy JASP which has a great user interface and does almost everything you might want. It is free
Can also do Bayes

Please leet me know if you find helpful; - just me not whole SPSS list
best
Diana

On 6 Sep 2020, at 02:10, Bruce Weaver <[hidden email]> wrote:

Hello Jessica.  I assume that under normal conditions (i.e., no pandemic),
you would have access to SPSS in on-campus computer labs.  My own university
has set up virtual computer labs to deal with this problem:

https://www.lakeheadu.ca/faculty-and-staff/departments/services/helpdesk/computer-labs/virtual-computer-labs

You might want to check with your IT Dept to see if they are doing something
similar.  

HTH.


Jessica Pola wrote
Do you offer a 3 month subscription of the SPSS Software.  I am currently
taking a Social Statistics course and I need this program only for this
fall semester. The price for the program is out of my price range.  Due to
covid19 I'm struggling financially.  Please let me know if you offer an
incentive or student discount.

=====================
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.

--
Sent from: http://spssx-discussion.1045642.n5.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

____________
University of Hertfordshire
College Lane, Hatfield, Hertfordshire AL10 9AB, UK
+44 (0) 208 444 2081
+44 (0) 7403 18 16 12
[hidden email]
http://dianakornbrot.wordpress.com/
skype:  kornbrotme
Save our in-boxes! http://emailcharter.org
 __________________







===================== 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
Reply | Threaded
Open this post in threaded view
|

Re: Student Edition

bdates
Jessica,

Aside from the recommendations of Bruce and Diana, there are alternatives for rental. You mention money is tight, and I'm not sure how tight that is, but OntheHub ($39.95), Hearne ($34.95), and StudentDiscounts.com ($34.95) are all student discount 6-month rentals. That's a whole lot less than a textbook.

Brian

From: SPSSX(r) Discussion <[hidden email]> on behalf of Kornbrot, Diana <[hidden email]>
Sent: Sunday, September 6, 2020 4:40 AM
To: [hidden email] <[hidden email]>
Subject: Re: Student Edition
 
Typical higher ed license enable students and staff to use off campus via VPN
But you might enjoy JASP which has a great user interface and does almost everything you might want. It is free
Can also do Bayes

Please leet me know if you find helpful; - just me not whole SPSS list
best
Diana

On 6 Sep 2020, at 02:10, Bruce Weaver <[hidden email]> wrote:

Hello Jessica.  I assume that under normal conditions (i.e., no pandemic),
you would have access to SPSS in on-campus computer labs.  My own university
has set up virtual computer labs to deal with this problem:

https://www.lakeheadu.ca/faculty-and-staff/departments/services/helpdesk/computer-labs/virtual-computer-labs

You might want to check with your IT Dept to see if they are doing something
similar.  

HTH.


Jessica Pola wrote
Do you offer a 3 month subscription of the SPSS Software.  I am currently
taking a Social Statistics course and I need this program only for this
fall semester. The price for the program is out of my price range.  Due to
covid19 I'm struggling financially.  Please let me know if you offer an
incentive or student discount.

=====================
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.

--
Sent from: http://spssx-discussion.1045642.n5.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

____________
University of Hertfordshire
College Lane, Hatfield, Hertfordshire AL10 9AB, UK
+44 (0) 208 444 2081
+44 (0) 7403 18 16 12
[hidden email]
http://dianakornbrot.wordpress.com/
skype:  kornbrotme
Save our in-boxes! http://emailcharter.org
 __________________







===================== 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
Reply | Threaded
Open this post in threaded view
|

Re: Student Edition

Kornbrot, Diana
Also recommend stats cloud https://statscloud.app/index.html
Designed for students & very simple to use

JMP aslo has cheap student version form OntheHub -its very good

On 6 Sep 2020, at 15:47, Dates, Brian <[hidden email]> wrote:

Jessica,

Aside from the recommendations of Bruce and Diana, there are alternatives for rental. You mention money is tight, and I'm not sure how tight that is, but OntheHub ($39.95), Hearne ($34.95), and StudentDiscounts.com ($34.95) are all student discount 6-month rentals. That's a whole lot less than a textbook.

Brian

From: SPSSX(r) Discussion <[hidden email]> on behalf of Kornbrot, Diana <[hidden email]>
Sent: Sunday, September 6, 2020 4:40 AM
To: [hidden email] <[hidden email]>
Subject: Re: Student Edition
 
Typical higher ed license enable students and staff to use off campus via VPN
But you might enjoy JASP which has a great user interface and does almost everything you might want. It is free
Can also do Bayes

Please leet me know if you find helpful; - just me not whole SPSS list
best
Diana

On 6 Sep 2020, at 02:10, Bruce Weaver <[hidden email]> wrote:

Hello Jessica.  I assume that under normal conditions (i.e., no pandemic),
you would have access to SPSS in on-campus computer labs.  My own university
has set up virtual computer labs to deal with this problem:

https://www.lakeheadu.ca/faculty-and-staff/departments/services/helpdesk/computer-labs/virtual-computer-labs

You might want to check with your IT Dept to see if they are doing something
similar.  

HTH.


Jessica Pola wrote
Do you offer a 3 month subscription of the SPSS Software.  I am currently
taking a Social Statistics course and I need this program only for this
fall semester. The price for the program is out of my price range.  Due to
covid19 I'm struggling financially.  Please let me know if you offer an
incentive or student discount.

=====================
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.

--
Sent from: http://spssx-discussion.1045642.n5.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

____________
University of Hertfordshire
College Lane, Hatfield, Hertfordshire AL10 9AB, UK
+44 (0) 208 444 2081
+44 (0) 7403 18 16 12
[hidden email]
http://dianakornbrot.wordpress.com/
skype:  kornbrotme
Save our in-boxes! http://emailcharter.org
 __________________







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

____________
University of Hertfordshire
College Lane, Hatfield, Hertfordshire AL10 9AB, UK
+44 (0) 208 444 2081
+44 (0) 7403 18 16 12
[hidden email]
http://dianakornbrot.wordpress.com/
skype:  kornbrotme
Save our in-boxes! http://emailcharter.org
 __________________







===================== 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
Reply | Threaded
Open this post in threaded view
|

Re: Student Edition

Mike
A question that needs to be asked is to what extent does the instructor for the
course require SPSS for the course because aspects of the course will focus on
SPSS specific features like syntax/menus, output, specific  analyses (e.g., reliability,
Cronbach's alpha, etc.), and so on.  Will the instructor require students to
hand in copies of syntax (even if one is using the menu to specify the analysis)
and/or output?  If so, you might be limited to using SPSS.

Also, It would  be good to know what types of statistical procedures will be covered. 
If the instructor doesn't care what package you use for the calculations, then
depending on what procedures are covered, it is possible that Excel's Data Analysis
toolpakplug-in might be able to handle most of the analyses (see the following website
for what is covered:
All students can get a copy of Microsoft Office 365 for free, so getting access to Excel is not a problem.

If more sophisticated statistical analyses are to be done, there are 3rd party add-ins
for Excel, like XLStat, that are available and with student pricing; see:
I believe Excel plus add-on is popular among economists (I continue to be surprised
by how often articles in economics make their data/results available in Excel).
So, the academic area and what is usual practice for statistical analysis in that
area is another consideration.

Another point to remember that a student is likely to have access to Excel long
after the course is over and probably access it anywhere in the world.  The same
cannot be said for the standard statistical packages -- except, of course, when we
reach a time when everyone is using R.

A couple of additional points:
(1)  As Bruce has pointed out, many schools have virtual computer labs that
can be accessed through the school's LMS.  At NYU, the virtual lab provides
access to SPSS and SAS but only to students (I don't think they have changed
that for this school year).  However, some colleges may not have the financial
resources for the licenses to make SPSS available to everyone but one would
assume that a person taking a statistics course would be granted access. 

(2)  David's suggestion of using OntheHub, etc., to purchase a rental license
is probably the way to go if (1) your instructor really, really wants you to use
SPSS, and (2) your school doesn't provide free online access.

(3)  Again, what one may need for the course depends upon the analyses
being covered.  Once upon a time most statistical analyses in the undergrad
intro statistics course only required one to have a calculator and knw how to lay out
one's calculations in order to keep track of intermediate steps and to aid in
double-checking.  But if the course relies a lot on graphical displays/visualization,
simulations, bootstrapping, other computationally intensive procedures, then
you will need something like SPSS or, if your instructor permits, some other
package. 
NOTE:  if your instructor insists on SPSS but you have to use another package,
say JMP, your instructor probably will not be able to help you if you have
problems with JMP or whatever.  They may also be somewhat peeved that
you are using another package, especially one that they are not familiar with,
because there may be departmental requirements that a particular package
be used, e.g., the public course description says SPSS will be used. 

If the instructor doesn't care how the statistical analyses are done, you
have a lot of options.  If they do care, your options are limited.

-Mike Palij
New York University


On Sun, Sep 6, 2020 at 11:06 AM Kornbrot, Diana <[hidden email]> wrote:
Also recommend stats cloud https://statscloud.app/index.html
Designed for students & very simple to use

JMP aslo has cheap student version form OntheHub -its very good

On 6 Sep 2020, at 15:47, Dates, Brian <[hidden email]> wrote:

Jessica,

Aside from the recommendations of Bruce and Diana, there are alternatives for rental. You mention money is tight, and I'm not sure how tight that is, but OntheHub ($39.95), Hearne ($34.95), and StudentDiscounts.com ($34.95) are all student discount 6-month rentals. That's a whole lot less than a textbook.

Brian

From: SPSSX(r) Discussion <[hidden email]> on behalf of Kornbrot, Diana <[hidden email]>
Sent: Sunday, September 6, 2020 4:40 AM
To: [hidden email] <[hidden email]>
Subject: Re: Student Edition
 
Typical higher ed license enable students and staff to use off campus via VPN
But you might enjoy JASP which has a great user interface and does almost everything you might want. It is free
Can also do Bayes

Please leet me know if you find helpful; - just me not whole SPSS list
best
Diana

On 6 Sep 2020, at 02:10, Bruce Weaver <[hidden email]> wrote:

Hello Jessica.  I assume that under normal conditions (i.e., no pandemic),
you would have access to SPSS in on-campus computer labs.  My own university
has set up virtual computer labs to deal with this problem:

https://www.lakeheadu.ca/faculty-and-staff/departments/services/helpdesk/computer-labs/virtual-computer-labs

You might want to check with your IT Dept to see if they are doing something
similar.  

HTH.


Jessica Pola wrote
Do you offer a 3 month subscription of the SPSS Software.  I am currently
taking a Social Statistics course and I need this program only for this
fall semester. The price for the program is out of my price range.  Due to
covid19 I'm struggling financially.  Please let me know if you offer an
incentive or student discount.

=====================
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.

--
Sent from: http://spssx-discussion.1045642.n5.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

____________
University of Hertfordshire
College Lane, Hatfield, Hertfordshire AL10 9AB, UK
+44 (0) 208 444 2081
+44 (0) 7403 18 16 12
[hidden email]
http://dianakornbrot.wordpress.com/
skype:  kornbrotme
Save our in-boxes! http://emailcharter.org
 __________________







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

____________
University of Hertfordshire
College Lane, Hatfield, Hertfordshire AL10 9AB, UK
+44 (0) 208 444 2081
+44 (0) 7403 18 16 12
[hidden email]
http://dianakornbrot.wordpress.com/
skype:  kornbrotme
Save our in-boxes! http://emailcharter.org
 __________________







===================== 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
Reply | Threaded
Open this post in threaded view
|

Re: Student Edition

Art Kendall
Good summary.

Excel can be a great way to share the data itself if one check the recent
archives to see how to include the data definition.  From the perspective
that the purpose of statistics is to make reasoned arguments, knowing what
the data does and does not say is critical.

I may have missed it but there were very substantial numerical analysis
objections to using Excel for more than a spreadsheet.

Some reasons to have students use the same software are:
to reinforce the habit of quality assurance by checking each other's work.
to have them see that there can be different valid ways to do analyses.

Jon recently alerted me about exporting to a web format, which is a useful
way to share results.



-----
Art Kendall
Social Research Consultants
--
Sent from: http://spssx-discussion.1045642.n5.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
Social Research Consultants
Reply | Threaded
Open this post in threaded view
|

Re: Student Edition

Jon Peck
In reply to this post by bdates
I am far removed from pricing and licensing issues, but looking on the web, I see Base Student version 27 offered for $34.95 for six months.  But the Standard version might be needed depending on the course content.  That goes for $49 on the web.  There is a more expensive Premium version, but that is unlikely to be needed for an intro course.  

A little off topic: I first encountered empirical statistics in an advanced economics course.  (The probability and statistics courses in the math department, my major, were entirely theoretical.)  Predating the arrival of the first digital computer on campus, we were required to compute a regression by hand (three variables).  Even electronic calculators were not available at the time.  We only computed the coefficients, since the SEs required even more calculation.

I would have been very happy to pay the equivalent of $34.95 (less than $5 at the time) to avoid this if I had had access to a computer to run it on!

The next year an IBM 1620 arrived, and I was quickly hooked.

On Sun, Sep 6, 2020 at 8:47 AM Dates, Brian <[hidden email]> wrote:
Jessica,

Aside from the recommendations of Bruce and Diana, there are alternatives for rental. You mention money is tight, and I'm not sure how tight that is, but OntheHub ($39.95), Hearne ($34.95), and StudentDiscounts.com ($34.95) are all student discount 6-month rentals. That's a whole lot less than a textbook.

Brian

From: SPSSX(r) Discussion <[hidden email]> on behalf of Kornbrot, Diana <[hidden email]>
Sent: Sunday, September 6, 2020 4:40 AM
To: [hidden email] <[hidden email]>
Subject: Re: Student Edition
 
Typical higher ed license enable students and staff to use off campus via VPN
But you might enjoy JASP which has a great user interface and does almost everything you might want. It is free
Can also do Bayes

Please leet me know if you find helpful; - just me not whole SPSS list
best
Diana

On 6 Sep 2020, at 02:10, Bruce Weaver <[hidden email]> wrote:

Hello Jessica.  I assume that under normal conditions (i.e., no pandemic),
you would have access to SPSS in on-campus computer labs.  My own university
has set up virtual computer labs to deal with this problem:

https://www.lakeheadu.ca/faculty-and-staff/departments/services/helpdesk/computer-labs/virtual-computer-labs

You might want to check with your IT Dept to see if they are doing something
similar.  

HTH.


Jessica Pola wrote
Do you offer a 3 month subscription of the SPSS Software.  I am currently
taking a Social Statistics course and I need this program only for this
fall semester. The price for the program is out of my price range.  Due to
covid19 I'm struggling financially.  Please let me know if you offer an
incentive or student discount.

=====================
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.

--
Sent from: http://spssx-discussion.1045642.n5.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

____________
University of Hertfordshire
College Lane, Hatfield, Hertfordshire AL10 9AB, UK
+44 (0) 208 444 2081
+44 (0) 7403 18 16 12
[hidden email]
http://dianakornbrot.wordpress.com/
skype:  kornbrotme
Save our in-boxes! http://emailcharter.org
 __________________







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


--
Jon K Peck
[hidden email]

===================== 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
Reply | Threaded
Open this post in threaded view
|

Re: Student Edition

Mike
In reply to this post by Art Kendall
From my reading of the stat literature on the accuracy of Excel (Bruce McCullough
has run a cottage industry on testing statistical software) one had to be really
careful with versions of Excel up and including Excel 2010.  By the mid-20teens
it seems that Microsoft finally realized that it should correct problems/errors in
Excel, some dating back to the 1990s.  I can't find references for the 2016 and
later versions but I do remember that these have increased in general accuracy. 
One problem that appears to persist has to do with floating point arithmetic; one
article that covers this issue is the following:

I'll leave it to Jon Peck to say how significant this problem is (it seems to
be relevant when dealing with very large numbers or very small numbers).

My own experience with using Excel in teaching undergrad psych statistics has
not revealed any problems -- I also use SPSS in the course so Excel's
calculations are double-checked for accuracy (during graduate school
I was repeatedly reminded that any important analyses should always
be conducted by two different programs -- one instance that stands out
in my memory is a physicist emphasizing this in a presentation on
neuromagnetism research; back then I compared results with BMDP and SPSS). 
For some assignments, like using the Anscombe quartet dataset, Excel has
the benefit of producing both the statistics and graphs easily and quickly.

-Mike Palij
New York University

On Sun, Sep 6, 2020 at 1:34 PM Art Kendall <[hidden email]> wrote:
Good summary.

Excel can be a great way to share the data itself if one check the recent
archives to see how to include the data definition.  From the perspective
that the purpose of statistics is to make reasoned arguments, knowing what
the data does and does not say is critical.

I may have missed it but there were very substantial numerical analysis
objections to using Excel for more than a spreadsheet.

Some reasons to have students use the same software are:
to reinforce the habit of quality assurance by checking each other's work.
to have them see that there can be different valid ways to do analyses.

Jon recently alerted me about exporting to a web format, which is a useful
way to share results.



-----
Art Kendall
Social Research Consultants
--
Sent from: https://urldefense.proofpoint.com/v2/url?u=http-3A__spssx-2Ddiscussion.1045642.n5.nabble.com_&d=DwICAg&c=slrrB7dE8n7gBJbeO0g-IQ&r=A8kXUln5f-BYIUaapBvbXA&m=aZxdZYSryUeJ9A5rfcduJUI0IOzyt7L2L8NOnQgK0kc&s=vc8OTjg-HnoNeezjQ_-YqaNxEaBOAm_YMxBZOCmpta0&e=

=====================
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
Reply | Threaded
Open this post in threaded view
|

Re: Student Edition

bdates
In addition to the Stats Pak, there's a free program for Excel at Real-Stats.com. It does a lot of stuff that SPSS basic does not. However, no matter how much greater the accuracy Excel gets, I have difficulty with the data set-up. Unless some of you know a way around this, most of the databases I have used allow a variable(s) for the independent variables, e.g., 1 and 0 for treatment groups. It's simply a matter of "telling" the program what the independent variable(s) are, and in some cases like the SPSS t-test, what the codes are. In Excel, I've not found that facility. Each dependent variable has to be separated into two variables, one for the subjects in the treatment group and one for those in the comparison group. I've had students who otherwise would have arrived at the right answers fail to respond correctly simply because they cut and pasted incorrectly. I don't what that as a 'variable' in their grade. So I prefer programs that allow separate variables for independent variables.

Brian

From: SPSSX(r) Discussion <[hidden email]> on behalf of Michael Palij <[hidden email]>
Sent: Sunday, September 6, 2020 6:02 PM
To: [hidden email] <[hidden email]>
Subject: Re: Student Edition
 
From my reading of the stat literature on the accuracy of Excel (Bruce McCullough
has run a cottage industry on testing statistical software) one had to be really
careful with versions of Excel up and including Excel 2010.  By the mid-20teens
it seems that Microsoft finally realized that it should correct problems/errors in
Excel, some dating back to the 1990s.  I can't find references for the 2016 and
later versions but I do remember that these have increased in general accuracy. 
One problem that appears to persist has to do with floating point arithmetic; one
article that covers this issue is the following:

I'll leave it to Jon Peck to say how significant this problem is (it seems to
be relevant when dealing with very large numbers or very small numbers).

My own experience with using Excel in teaching undergrad psych statistics has
not revealed any problems -- I also use SPSS in the course so Excel's
calculations are double-checked for accuracy (during graduate school
I was repeatedly reminded that any important analyses should always
be conducted by two different programs -- one instance that stands out
in my memory is a physicist emphasizing this in a presentation on
neuromagnetism research; back then I compared results with BMDP and SPSS). 
For some assignments, like using the Anscombe quartet dataset, Excel has
the benefit of producing both the statistics and graphs easily and quickly.

-Mike Palij
New York University

On Sun, Sep 6, 2020 at 1:34 PM Art Kendall <[hidden email]> wrote:
Good summary.

Excel can be a great way to share the data itself if one check the recent
archives to see how to include the data definition.  From the perspective
that the purpose of statistics is to make reasoned arguments, knowing what
the data does and does not say is critical.

I may have missed it but there were very substantial numerical analysis
objections to using Excel for more than a spreadsheet.

Some reasons to have students use the same software are:
to reinforce the habit of quality assurance by checking each other's work.
to have them see that there can be different valid ways to do analyses.

Jon recently alerted me about exporting to a web format, which is a useful
way to share results.



-----
Art Kendall
Social Research Consultants
--
Sent from: https://urldefense.proofpoint.com/v2/url?u=http-3A__spssx-2Ddiscussion.1045642.n5.nabble.com_&d=DwICAg&c=slrrB7dE8n7gBJbeO0g-IQ&r=A8kXUln5f-BYIUaapBvbXA&m=aZxdZYSryUeJ9A5rfcduJUI0IOzyt7L2L8NOnQgK0kc&s=vc8OTjg-HnoNeezjQ_-YqaNxEaBOAm_YMxBZOCmpta0&e=

=====================
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
===================== 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
Reply | Threaded
Open this post in threaded view
|

Re: Student Edition

Robert L
In reply to this post by Jessica Pola

From my perspective working as a statistician supporting healthcare staff, the way people  use Excel for statistical calculations is often quite depressing. The system is used for a lot of tasks of course, so statistics is only one area. But all these self-taught Excel wizards who learn to solve their specific problems have such an unprincipled approach that the resulting structures so often are such a mess: no variable names or other documentation, functions which noone can debug or adapt, datasets organized both vertically and horizontally, data for subgroups spread everywhere,…

 

Still, I don’t think Excel will be replaced. So for stats calculations, I usually suggest using the so called ”pivot tables” in Excel to set up tables, crosstabs, calculate descriptive measures, split the results with respect to categorical variables. When there is need to calculate p-values the results can be plugged into web based systems such as OpenEpi (http://www.openepi.com). Fairly simple, trustworthy routines (as far as I can tell). And the aspect I find perhaps most important is that the use of pivot table forces users to set up their data with columns/variables in a way which makes it possible to export to the data to SPSS or other systems. The pivot table approach also is similar to how we have to think in data analysis. What are the variables? What can be done with them? What _should_ be done? So I do think users learn some more general skills with the pivot tables than what they normally do in Excel: cut-and-paste data into small chunks, adapt some obscure function to rectangles somewhere in the worksheet and the like.

 

If students would learn to use a more principled approach, then the mess out in the workplaces could perhaps eventually improve.


Robert

===================== 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 Lundqvist
Reply | Threaded
Open this post in threaded view
|

Re: Student Edition

Jon Peck
In reply to this post by Mike
I don't think the extreme values in the examples in that reference are of any practical concern.  The article also illustrates the well known precision issues that can occur with floating point arithmetic even with typical data due to the inability of floating point hardware to represent exactly repeating decimals such as 1./3.  This can make mathematically identical expressions such as (a/b) * c and (a*c)/b give different results, but, again, the difference is rarely of interest.  Exact tests for equality in those situations should be avoided, however.

In choosing to use Excel or another package, I think the important issue is whether it provides the statistics beyond the initial results that would allow an investigator to determine their validity and alternatives to address such issues.

On Sun, Sep 6, 2020 at 4:02 PM Michael Palij <[hidden email]> wrote:
From my reading of the stat literature on the accuracy of Excel (Bruce McCullough
has run a cottage industry on testing statistical software) one had to be really
careful with versions of Excel up and including Excel 2010.  By the mid-20teens
it seems that Microsoft finally realized that it should correct problems/errors in
Excel, some dating back to the 1990s.  I can't find references for the 2016 and
later versions but I do remember that these have increased in general accuracy. 
One problem that appears to persist has to do with floating point arithmetic; one
article that covers this issue is the following:

I'll leave it to Jon Peck to say how significant this problem is (it seems to
be relevant when dealing with very large numbers or very small numbers).

My own experience with using Excel in teaching undergrad psych statistics has
not revealed any problems -- I also use SPSS in the course so Excel's
calculations are double-checked for accuracy (during graduate school
I was repeatedly reminded that any important analyses should always
be conducted by two different programs -- one instance that stands out
in my memory is a physicist emphasizing this in a presentation on
neuromagnetism research; back then I compared results with BMDP and SPSS). 
For some assignments, like using the Anscombe quartet dataset, Excel has
the benefit of producing both the statistics and graphs easily and quickly.

-Mike Palij
New York University

On Sun, Sep 6, 2020 at 1:34 PM Art Kendall <[hidden email]> wrote:
Good summary.

Excel can be a great way to share the data itself if one check the recent
archives to see how to include the data definition.  From the perspective
that the purpose of statistics is to make reasoned arguments, knowing what
the data does and does not say is critical.

I may have missed it but there were very substantial numerical analysis
objections to using Excel for more than a spreadsheet.

Some reasons to have students use the same software are:
to reinforce the habit of quality assurance by checking each other's work.
to have them see that there can be different valid ways to do analyses.

Jon recently alerted me about exporting to a web format, which is a useful
way to share results.



-----
Art Kendall
Social Research Consultants
--
Sent from: https://urldefense.proofpoint.com/v2/url?u=http-3A__spssx-2Ddiscussion.1045642.n5.nabble.com_&d=DwICAg&c=slrrB7dE8n7gBJbeO0g-IQ&r=A8kXUln5f-BYIUaapBvbXA&m=aZxdZYSryUeJ9A5rfcduJUI0IOzyt7L2L8NOnQgK0kc&s=vc8OTjg-HnoNeezjQ_-YqaNxEaBOAm_YMxBZOCmpta0&e=

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


--
Jon K Peck
[hidden email]

===================== 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
Reply | Threaded
Open this post in threaded view
|

Re: Student Edition

Rich Ulrich
Here is an Excel horror story, from real life.

IIRC, it was in the wake of Bush's 2008 Deregulation-Recession that
two credible economists published results showing that countries
were in trouble if their national debt exceeded their national, annual
income.  Many people pointed to the study to justify a tight-money policy.

A couple of years later, a grad student who tried to replicate their results
from their original data, graciously provided to him by the authors, found
that the analysis was wrong. In their Excel spreadsheet, they had
accidentally left off the last dozen (in the arbitrary order of their data) of
their 75 or so countries.  Including them did not reverse the results, but it
wiped out the conclusions.  Oops. 

It is nice to be able to read the code and see exactly what has been done;
Excel makes that harder.  At one time the Excel code was not guaranteed
to provide the N of cases for an analysis, which (I suspect) contributed to
their error - That's something that always should be checked.

--
Rich Ulrich





From: SPSSX(r) Discussion <[hidden email]> on behalf of Jon Peck <[hidden email]>
Sent: Monday, September 7, 2020 12:43 PM
To: [hidden email] <[hidden email]>
Subject: Re: Student Edition
 
I don't think the extreme values in the examples in that reference are of any practical concern.  The article also illustrates the well known precision issues that can occur with floating point arithmetic even with typical data due to the inability of floating point hardware to represent exactly repeating decimals such as 1./3.  This can make mathematically identical expressions such as (a/b) * c and (a*c)/b give different results, but, again, the difference is rarely of interest.  Exact tests for equality in those situations should be avoided, however.

In choosing to use Excel or another package, I think the important issue is whether it provides the statistics beyond the initial results that would allow an investigator to determine their validity and alternatives to address such issues.
...
===================== 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
Reply | Threaded
Open this post in threaded view
|

Re: Student Edition

Jon Peck
Excel is legendary for wrong results created by user error.  It's tremendous flexibility comes with tremendous possibilities for errors made more likely by that flexibility in contrast to the data discipline enforced by database and statistical software.  Not to say that errors can't occur there.

Using Excel may not be much of an issue for an intro stat class, but it would be a bad habit to get into for life.

On Mon, Sep 7, 2020 at 11:09 AM Rich Ulrich <[hidden email]> wrote:
Here is an Excel horror story, from real life.

IIRC, it was in the wake of Bush's 2008 Deregulation-Recession that
two credible economists published results showing that countries
were in trouble if their national debt exceeded their national, annual
income.  Many people pointed to the study to justify a tight-money policy.

A couple of years later, a grad student who tried to replicate their results
from their original data, graciously provided to him by the authors, found
that the analysis was wrong. In their Excel spreadsheet, they had
accidentally left off the last dozen (in the arbitrary order of their data) of
their 75 or so countries.  Including them did not reverse the results, but it
wiped out the conclusions.  Oops. 

It is nice to be able to read the code and see exactly what has been done;
Excel makes that harder.  At one time the Excel code was not guaranteed
to provide the N of cases for an analysis, which (I suspect) contributed to
their error - That's something that always should be checked.

--
Rich Ulrich





From: SPSSX(r) Discussion <[hidden email]> on behalf of Jon Peck <[hidden email]>
Sent: Monday, September 7, 2020 12:43 PM
To: [hidden email] <[hidden email]>
Subject: Re: Student Edition
 
I don't think the extreme values in the examples in that reference are of any practical concern.  The article also illustrates the well known precision issues that can occur with floating point arithmetic even with typical data due to the inability of floating point hardware to represent exactly repeating decimals such as 1./3.  This can make mathematically identical expressions such as (a/b) * c and (a*c)/b give different results, but, again, the difference is rarely of interest.  Exact tests for equality in those situations should be avoided, however.

In choosing to use Excel or another package, I think the important issue is whether it provides the statistics beyond the initial results that would allow an investigator to determine their validity and alternatives to address such issues.
...


--
Jon K Peck
[hidden email]

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