
To: cse5800@yahoogroups.com
From: Ibrahim El-Far  <ielfar@fit.edu>
MIME-Version: 1.0
Mailing-List: list cse5800@yahoogroups.com; contact cse5800-owner@yahoogroups.com
Date: Thu, 7 Feb 2002 14:58:20 -0500 (EST)
Subject: [cse5800] Next week's reading assignment
Content-Transfer-Encoding: 7bit

Greetings everyone,

This is a reminder of the reading assignment for next week. There are two
articles whose titles are

1) Theoretical Comparison of Testing Methods
2) An Empirical Comparison of Seven Programming Languages

The first article is a critique of how researchers used to compare
software testing methods along with a few recommendations for future work.
The second article compares several programming languages along a number
of guidelines and it presents "hard data"

(The descriptions above may very well be inaccurate. I only skimmed
through these recently, and it has been a very long time since I last read
either).

I would like you to analyze the first paper and look for what kind of
threats to validity he discussed and what kinds did he miss. Mind you, the
paper itself is not a study. It is merely a critique (well, it's almost a
rant) of studies that aim to compare testing methods.The second paper is a
study, and I would also like you to look for discussions of validity: what
is there and what is missing?

The structure of yor critique can be as follows:
- Would you characterize the study as quantitative or qualitative?
- Are the inferences, arguments, conclusions backed by some kind of data?
- What is the nature or type of this data?
- What instrument was used to collect this data?
- Was evidence presented for external, instrument, and internal validity?
- Was there any specific threats or potential obstacles that were not
addressed?

Here's a list of what to look for:
* Threats to External Validity:
- Results only apply to the sample 
- Small or narrow sample size
- Purposive or convenience samples
- Non-random sample of instrument administrators
- Information is missing on a sizeable portion of the initial sample
* Threats to Instrument Validity
- Is the instrument content an adequate sample of content domain it should
be representing?
- Is the format of the instrument appropriate enough so as to be able to
obtain valid results?
* Threats to Internal Validity:
- Subject characteristics
- Loss of subjects
- Location
- Instrument decay
- Data collector characteristics
- Data collector bias
- Testing
- History
- Maturation
- Attitude of subjects
- Regression
- Implementation
* Was there a discussion of:
- Predictive validity
- Concurrent validity

In case you cannot find a discussion of a certain item in the list:
- was the item irrelevant?
- was it ignored justifiably (trivially obvious, for example)?
- etc etc etc

If you have some trouble with the definitions, you can look them up in the
slides. If the slides only contain headings, then you will find the proper
reference (where to find the definitions and related text) in the text
book and on the website that I mentioned in class.

In any case, you may have to end up reading Chapter 6 (only section on
external reliability) Chapter 8 (only section on validity) and Chapter 9
(all of it). This will provide you with most of the background information
that you need. If you feel that you need a different perspective or a
different wording of the same concepts discussed in the book, I have
included the list of sections on the website that are relevant to next
class's discussion in the slides (Background Reading slide).

If you have any questions or concerns, let me know.

Kind Regards,

Ibrahim


------------------------ Yahoo! Groups Sponsor ---------------------~-->
Get your FREE credit report with a FREE CreditCheck
Monitoring Service trial
http://us.click.yahoo.com/ACHqaB/bQ8CAA/ySSFAA/BcXolB/TM
---------------------------------------------------------------------~->

To unsubscribe from this group, send an email to:
cse5800-unsubscribe@yahoogroups.com

 
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/