Re: Stats course geared towards Program Evaluation

Posted by Bruce Weaver on Mar 06, 2015; 11:29pm
URL: http://spssx-discussion.165.s1.nabble.com/Stats-course-geared-towards-Program-Evaluation-tp5728871p5728909.html

In a slightly different (but I think related) vein, I once heard an education researcher talking about a study that involved some intervention to improve student performance in a particular course.  Somewhat surprisingly, the results showed no improvement in the target course.  However, there was a substantial improvement in another course.  It was a situation where each course was populated by the same students.  So the researchers speculated that students were able to maintain their grades in the target course with less time and effort, and that they gave that extra time and effort to the other course, where it was apparently more important to them to improve their performance.  But if the researchers had not been paying attention to what was going on in the other non-target course(s), they might easily have concluded that the intervention was not effective.  

I can't remember who the researcher was.  But if anyone is dying to know, I can probably track it down.

Cheers,
Bruce

Mike Palij wrote

Just one more addition to this thread, a cautionary tale.
Susan Reverby (a researcher best known for her work on
the Tuskegee syphilis nontreatment study; see:
http://books.google.com/books?hl=en&lr=&id=qm5X5gW7qNIC&oi=fnd&pg=PR7&dq=%22susan+reverby%22+tuskegee&ots=b5Q3bpmUif&sig=GasL1Eq2uegPmHuaZD9uCyNX1pM#v=onepage&q=%22susan%20reverby%22%20tuskegee&f=false )
who is on another mailing list that I'm a member of, provided a
link to an analysis by HIV researcher Ida Susser of a program
intervention study that was published in the New England Journal
of Medicine.  The program was to promote preventive measures
to stop the spread of the HIV but, as implemented, appears to
have failed to do so.  Susser identifies some possible factors for
why the intervention failed which highlights some of the problems
the exist between the people implementing a program and the
recipients of the program.  Susser had her analysis published
on the Aljazeera America website (just a news website, not
a jihadi outlet) and it shows that when doing program evaluation
one cannot simply do statistical analysis but one also should do
an analysis of how the program was implemented.
NOTE: even if the program was implemented as planned, the
plan that was used may not have been the best approach to use,
consequently, it should not come as a surprise that there were
negative results, however, it failed because the wrong implementation
was used not that the implementation does not work.
Susser's article can be read here:
http://america.aljazeera.com/opinions/2015/3/blame-research-design-for-failed-hiv-study.html#

-Mike Palij
New York University
[hidden email]

  ----- Original Message -----
  From: Dates, Brian
  To: [hidden email] 
  Sent: Monday, March 02, 2015 10:19 AM
  Subject: Re: Stats course geared towards Program Evaluation


  Ryan,

   

  The Evaluators’ Institute at George Washington University has a full cadre of courses for evaluators.  Here’s the current listing:

   

  Analytic Approaches

    a.. Applied Regression Analysis for Evaluators
    b.. Applied Statistics for Evaluators
    c.. Hierarchical Linear Modeling
    d.. Intermediate Cost-Benefit and Cost-Effectiveness Analysis
    e.. Intermediate Qualitative Analysis
    f.. Introduction to Cost-Benefit and Cost-Effectiveness Analysis
    g.. Needs Assessment
    h.. Practical Meta-Analysis: Summarizing Results Across Studies
    i.. Qualitative Data Analysis
  The link to the Evaluators’ Institute is: http://tei.gwu.edu/course-listing-category 

   

  I’d also recommend the session by Stephanie Evergreen, Presenting Data Effectively: Practical Methods for Improving Evaluation Communication . She’s done amazing work with data visualization. We’ve had her at my organization for a full day, and her stuff is really good.

   

  Hope this all helps.

   

  Brian

   

  Brian Dates, M.A.
  Director of Evaluation and Research | Evaluation & Research | Southwest Counseling Solutions
  Southwest Solutions
  1700 Waterman, Detroit, MI 48209
  313-841-8900 (x7442) office | 313-849-2702 fax
  [hidden email] | www.swsol.org

   

  From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Ryan Black
  Sent: Monday, March 02, 2015 10:05 AM
  To: [hidden email]
  Subject: Stats course geared towards Program Evaluation

   

  OT:

   

  Is anyone familiar a grad level stats course that is geared towards program evaluation?

   

  Thanks,

   

  Ryan

  ===================== 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