[Hidden-tech] suggestions needed for a relational database
Chris Hoogendyk
hoogendyk at bio.umass.edu
Thu Mar 11 15:07:15 EST 2010
Keri Heitner wrote:
>
> Thank you for all of the suggestions. Here is additional information
> about what will be tracked and who will be entering the data:
>
>
>
> The purpose is to track student information to see if the program is
> helping with GPA, test scores, etc. Instructors collect data on: GPA
> pre- and post-course, math test scores, which students dropped out
> during the school year, which students went on to a 4-year college,
> attendance, graduation rates, infraction rates.
>
>
>
> Staff running the course at another college use Efforts to Outcomes
> software. The problem is that many teachers stopped uploading the
> above information because the process was very tedious. So the
> database must support multiple users and be easy to use.
>
>
>
> Keri
>
> · Keri L. Heitner, Ph.D.
> All Aspects Research
> 413 549-0154 ph/fax (ET)
> kheitner at earthlink.net <mailto:kheitner at earthlink.net>
>
So, as some others have said, this is a multi-level problem.
You need to assess where you have resources, what can be done, where to
allocate the effort, and who's going to have to supply the effort.
You're the research person who wants the data and wants to work with it.
The teachers need a really easy mechanism for entering data, and need
any roadblocks removed, because there basically is little motivation for
them given all their other responsibilities.
You could apply effort on the user interface at the teacher's end, and
this would make life easier for you, because it would deliver data into
your database in a ready to use form. You just analyze. However, this
then becomes a software development project and has to address possibly
multiple desktop platforms and versions, depending on your teacher
population. That might require an experienced professional developer and
might cost a pretty penny.
Alternatively, you could develop a web interface to your database and
have teachers navigate to that, log in, etc. However, that then gets you
into web development, security issues, systems administration, and,
again, a developer (web this time) with some experience, and that would
cost something. And then you need to host that web site somewhere, which
leads to ongoing costs. Here you might end up in what is referred to as
the LAMP environment (Linux, Apache, MySQL and PHP). There are lots of
developers out there who can deal with this and lots of packages that
can provide a starting point. Make sure whoever does it understands
security.
Another alternative might be that you give the teachers a record format
that you want, and have them email data to you (either in-line or
attached) from whatever means they have. Some of them might have a
grading program, some might have a spread sheet, some might have a paper
grade book. You would want a flat file of one line per student with
specified fields. Since it's going over email, you might want to encode
student identification to protect that information as it traverses the
internet. Then, the labor falls on you, the motivated researcher, to
transfer their data from the emails to your database and analyze it.
This is closer to the old fashioned way of dealing with data. But the
costs are different. You don't have the up front developer costs. You
probably have more cost for your own time. Your teachers might find it
easier to get data to you. So the value of your research might be
higher. For updates, you could email them a report of their data with
requests to fill out certain additional fields, in line in the email,
and return it (reply).
With the last option, just about any means of managing the data that you
are comfortable with would work. Then, how are you going to analyze it.
What kinds of tests? Are you going to construct those yourself? Or are
you going to use a statistical package? Depending on your data design, a
spread sheet might give you more analytical capabilities. But you could
use Access or FileMaker and generate summary reports that you export and
analyze in a spread sheet.
--
---------------
Chris Hoogendyk
-
O__ ---- Systems Administrator
c/ /'_ --- Biology & Geology Departments
(*) \(*) -- 140 Morrill Science Center
~~~~~~~~~~ - University of Massachusetts, Amherst
<hoogendyk at bio.umass.edu>
---------------
Erdös 4
More information about the Hidden-discuss
mailing list