[Blindmath] access to statistics for blind students

Sun Aug 29 20:33:32 UTC 2010

Hello Vince et al.,

I don't want the following to appear over critical, but I do need to 
test your ideas and assumptions because at present I can offer only 
qualified support.

I'm interested in knowing the factual basis of several of your 
claims. And then testing whether the facts as they might stand now 
are relevant for the time when your project is successful, which I do 
truly hope is the case.

First you say SPSS is the leader in industry. I wonder if this 
depends on which industry you mean, but outside the US, I doubt this 
claim holds water even if it does inside.

Second, I agree that the first time you do something, a GUI based 
system is faster than a command line based system, no matter what 
context. However, if you make a mistake, the whole process must be 
undertaken from the last most successful step. In a command line 
system you can re-run an entire day's work in minutes once the error 
has been corrected. I admit that the downside of using a GUI system 
is not well exposed in an introductory setting, but once you get into 
doing things that involve a multistage project where results from one 
action impact on what you do for the next action, a GUI is an 
inefficient way to operate. For example, I have written up the entire 
set of examples for the textbook used in one of my classes this 
semester. The output (both text and graphical) is then printed in a 
manual, including all the code used. Yes it took some time to prepare 
and fix the errors, but I can now renew the entire document after 
installing a new version of R, in less than 2 minutes on a 6-year old 
desktop PC. That's raw text files to beautiful presentation standard 
pdf. I could not do this on a GUI based system as I would need to do 
the examples one by one all over again.

I am now teaching an introductory class of Engineering students. I 
have given them the option of choosing a GUI system or R and offered 
to support both options. Around half took the easy option of using 
Minitab as it was "familiar" given their existing knowledge.

I now observe students exchanging help as R users in a way that is 
not possible to explain when two GUI users get together.

Pedagogically, I also find the command line system a better teaching 
option as it forces a student to ask different types of 
questions.  For example, a GUI user can go to the menus to see what 
options they have when wanting to display a set of say continuous 
data. They find the Graph menu and follow down the list, stopping 
(hopefully) at the histogram item. They then go into the dialogue and 
add the variable into the box as required. Often, all the variables 
that could not be graphed because they are not the right format are 
not available as options. They click OK and look at the graph to 
decide if it looks right. They do not need to wonder before hand if 
there were enough observations to bother with a histogram because the 
evidence is before them if there weren't.

A command line user could go through the same process. If you look at 
being a command line user emulating a GUI user then you see the 
inefficiency of being a command user. What I need to teach the 
students is to think about which graph is the most appropriate given 
the structure of the data. They may well have checked out how large 
the data set was on importing it. I encourage this in both settings 
actually. The user knows that typing out the commands for 
experimenting on which graph is going to be expensive in terms of the 
time taken so they have to think first, "Which one of the graphs I've 
seen before  is most likely to be useful?" Then get the graph.

Now let me run to the world of statistical software as it pertains to 
a blind user. First, without sighted assistance or access to tactile 
displays or the recent (and quite interesting) audible options, the 
blind user cannot work with the graphs. Even when we do have access 
to tactile or sonic options, we will always be slower than our 
sighted peers in terms of interpreting the graphs for ourselves (note 
not using expert systems). We must be as efficient as we can in terms 
of what we generate and then how we access it. Graphs are for the 
sighted reader. They are summary tools that are useful for the 
sighted. Every graph can be replaced by suitable textual analytic 
tools or processes, although we don't teach these tools as often as I 
would like.

Then having decided to work with text, we need to make sure we are 
not being bombarded with screens and screens of things that we do not 
need to read. Anyone who has worked with SAS knows what I mean here, 
but Minitab and SPSS among many programs churn out lots of lovely 
formatted text for the reader to process. As a blind user with more 
than enough to read, I want the necessary things read aloud and the 
frippery left silent. The sighted world filters the frippery out very 
quickly and I can't do this so easily using sound.

I do wish you every success in your endeavours. I suspect your 
results will be more use to introductory students than professionals however.

Jonathan

At 06:31 a.m. 30/08/2010, you wrote:
>My advisor has assigned a Post doc and a couple of more people in the
>sonnification lab at Georgia Tech to start working with me on an Accessible
>Statistics project.  What we need to determine is exactly how to go about
>approaching the entire problem.  The ultimate goal is to create both an
>formal and informal approach towards the topic, such as just presenting and
>teaching Statistics to blind high school and college students.  The other is
>a way to make the output from SPSS and other calculating packages
>accessible.
>
>Since SPSS is by far the leader in the industry, we would like to make their
>program accessible to screenreading programs.  We know that "R" can be
>utilized by using command lines from the terminal prompt, but that still
>requires blind students to do way more than their sighted peers just to do
>their calculations.  What we really would like input from the community
>about is what is the most important thing to attack first.  I have a STATS
>class this semester and we are using SPSS for the lab portion and also to
>complete any class project.  This is now the standard in most universities
>in the country, whether it is undergratuate or graduate level.  Since I will
>have another year and a half of classess in Statistics and then research on
>a dissertation, making SPSS is the priority for me.
>
>
>
>
>
>-----Original Message-----
>From: blindmath-bounces at nfbnet.org [mailto:blindmath-bounces at nfbnet.org] On
>Behalf Of blindmath-request at nfbnet.org
>Sent: Sunday, August 29, 2010 1:00 PM
>To: blindmath at nfbnet.org
>Subject: Blindmath Digest, Vol 49, Issue 22
>
>Send Blindmath mailing list submissions to
>         blindmath at nfbnet.org
>
>To subscribe or unsubscribe via the World Wide Web, visit
>         http://www.nfbnet.org/mailman/listinfo/blindmath_nfbnet.org
>or, via email, send a message with subject or body 'help' to
>         blindmath-request at nfbnet.org
>
>You can reach the person managing the list at
>         blindmath-owner at nfbnet.org
>
>When replying, please edit your Subject line so it is more specific
>than "Re: Contents of Blindmath digest..."
>
>
>Today's Topics:
>
>    1. Fw: [Nfb-science] Fw: Canadian scientists envision        synthetic
>       corneas (Christine Szostak)
>
>
>----------------------------------------------------------------------
>
>Message: 1
>Date: Sat, 28 Aug 2010 13:57:38 -0400
>From: "Christine Szostak" <szostak.1 at osu.edu>
>To: "Discussion list for NABS,  National Alliance of Blind Students."
>         <nabs at acb.org>,         "Blind Math list for those interested in
>mathematics"
>         <blindmath at nfbnet.org>
>Subject: [Blindmath] Fw: [Nfb-science] Fw: Canadian scientists
>         envision        synthetic corneas
>Message-ID: <5BEDF5F3F5F54D009DDDDB6D26C43B37 at cs11>
>Content-Type: text/plain; format=flowed; charset="iso-8859-1";
>         reply-type=original
>
>Hi All,
>   I just thought some hear might be interested. I apologize as this is not
>100% related to list topics, but  it may prove of interest to many here.
>many thanks,
>Christine
>Christine M. Szostak
>Graduate Student
>Language Perception Laboratory
>Department of Psychology, Cognitive Area
>The Ohio State University
>Columbus, Ohio
>szostak.1 at osu.edu
>----- Original Message -----
>From: "qubit" <lauraeaves at yahoo.com>
>To: <nfb-talk at nfbnet.org>; <nfb-science at nfbnet.org>
>Sent: Friday, August 27, 2010 11:06 PM
>Subject: [Nfb-science] Fw: Canadian scientists envision synthetic corneas
>
>
>I recently got the following article from a friend and thought it was quite
>exciting if it can be perfected.  I am sending it along for those interested
>
>in corneal transplants.
>Enjoy!
>--le
>
>----- Original Message -----
>From: Gianfranco Di Cosmo
>To: Undisclosed-Recipient:;
>Sent: Thursday, August 26, 2010 3:13 AM
>Subject: Canadian scientists envision synthetic corneas
>
>
>
>Canadian researchers may have made a breakthrough in the treatment of
>blindness caused by damaged corneas.
>
>They've developed a biosynthetic cornea that can actually help the eye
>repair its own damaged eye tissue and restore vision. And with further
>research, they say their approach could help restore sight to millions of
>people around the world who have lost their sight from diseases that lead to
>
>clouding of the cornea.
>
>The research, published today in the journal Science Translational Medicine
>is a small one, involving just 10 people. But scientists say they were
>surprised that the treatment worked in the majority of patients who had
>scarred corneas, helping to restore the sight of nine of the 10 patients.
>
>The cornea is the thin, transparent layer of collagen and cells that acts as
>
>a window on the eyeball. In most cases of corneal damage, only a transplant
>can restore sight.
>
>But in this research, Dr. May Griffith of the Ottawa Hospital Research
>Institute, the University of Ottawa and Link?ping University in Sweden
>created corneas using biosynthetic collagen produced in the lab that was
>moulded into the shape of a cornea, much like a contact lens.
>
>After first testing the corneas on pigs (who have eyes similar to humans),
>they recruited 10 Swedish patients with advanced keratoconus, or central
>corneal scarring. Each patient underwent surgery in 2007 to remove damaged
>corneal tissue. That was then replaced with corneas made from synthetic
>human collagen, which were sewn onto the eyes.
>
>For two years, researchers watched what happened. Over time, the implants
>acted as scaffolding to help the eye restore normal corneal cell and nerve
>growth.
>
>"You put the material in the eye and it becomes almost an integral part. It
>allows the natural cells of the person treated to go into the material and
>become part of it," co-author Rejean Munger of the Ottawa Hospital Research
>Institute told CTV News.
>
>The corneas even became sensitive to touch and started making tears to keep
>the eyes oxygenated.
>
>Griffith says while the study was intended only to test the safety of the
>new corneas, her team found that nine of the 10 patients saw their vision
>improve, though some needed to wear contact lenses. The 10th patient is
>improving though much more slowly.
>
>"We were actually very surprised and happy that we saw improvement in the
>vision," Griffith said.
>
>"After surgery, patients didn't have perfect vision, but they could see
>better," she says. "One patient had almost perfect vision; others had
>slightly less than perfect vision."
>
>The research was published Wednesday in the journal Science Translational
>Medicine.
>
>
>Page 2.
>Dr. Keith Gordon of the CNIB notes that there are millions of people in
>Canada and around the world who need new corneas, but there is a dire
>shortage of donor corneas. A synthetic cornea from a lab would be an
>important new way of treating patients.
>
>"If these transplants are as effective as they appear to be, we have got a
>winner. And it will be exciting and useful for people with vision loss due
>to corneal disease," he says.
>
>Researchers think it will take another five years to further improve the
>implants, and to test them in other eye conditions.
>
>With a report from CTV's medical specialist Avis Favaro and producer
>Elizabeth St. Philip
>
>
>CTV.ca News Staff
>
>
>Posted at 2010/08/25 18:51:13
>
>If you want to visit my website,   http://dicosmo.ifreepages.com/
>
>Gianfranco
>
>_______________________________________________
>Nfb-science mailing list
>Nfb-science at nfbnet.org
>http://www.nfbnet.org/mailman/listinfo/nfb-science_nfbnet.org
>To unsubscribe, change your list options or get your account info for
>Nfb-science:
>http://www.nfbnet.org/mailman/options/nfb-science_nfbnet.org/szostak.1%40osu
>.edu
>
>
>
>
>------------------------------
>
>_______________________________________________
>Blindmath mailing list
>Blindmath at nfbnet.org
>http://www.nfbnet.org/mailman/listinfo/blindmath_nfbnet.org
>
>
>End of Blindmath Digest, Vol 49, Issue 22
>*****************************************
>
>
>
>_______________________________________________
>Blindmath mailing list
>Blindmath at nfbnet.org
>http://www.nfbnet.org/mailman/listinfo/blindmath_nfbnet.org
>To unsubscribe, change your list options or get your account info 
>for Blindmath:
>http://www.nfbnet.org/mailman/options/blindmath_nfbnet.org/a.j.godfrey%40massey.ac.nz

_____
Dr A. Jonathan R. Godfrey
Lecturer in Statistics
Institute of Fundamental Sciences
Massey University
Palmerston North

Office: Science Tower B Room 3.15
Phone: +64-6-356 9099 ext 7705
Mobile: +64-29-538-9814
Home Address: 22 Bond St, Palm. Nth.
Home Phone: +64-6-353 2224 (Just think FLEABAG) 