Following the topic "Concept or Calculation", I would like to ask what
advice colleagues offer in the following situation. As commonly happens,
I have been asked by a student who had completed his data collection,
analysis, and has a late draft report, for advice on reporting the tests
of association for tables. I noted that he had devoted most of a page to
reproducing a table (with counts, percentages, margins, shading, boxes or
double boxes) followed by a text that reported no significant difference
between the groups. I asked why he thought this had to occupy so much
space, and got the impression he thought he just had to include everything
that had been asked in the survey. The draft had been seen previously by
a supervisor, who was presumably content with the display.
So, I would like colleagues' opinions or thoughts on how we are now
training researchers:
Should they be drilled to report all results using a standard layout,
regardless of whether the data show any features or not?
OR
Should they be encouraged to show judgment, and include tables only
when the reader needs to see the tabulated counts?
My own view, as must be obvious, leans to the latter. In this case, what
seemed worth reporting was that the overwhelming majority (over 95% of N
which I've forgotten) agreed with the statement, and there was almost no
difference between young and old respondents (chi-squared test, p>0.4).
That information seemed lost in the detail of the table and his text,
which reported that "most" (cf survey that showed "most" can be
interpreted as anything over 60%) repondents agreed and that a chi-squared
test between the groups was not significant.
Does the advice differ according to whether the client is a PhD student,
a member of staff writing a paper, or someone writing a report for a
commercial client? If so, why?
Comments to me please, and I'll summarise. I am, OF COURSE, aware that in
this situation I can offer advice, but responsibility for the student's
progress lies with the supervisor and the department.
R. Allan Reese Email: [log in to unmask]
Associate Manager GRI Direct voice: +44 1482 466845
Graduate School Voice messages: +44 1482 466844
Hull University, Hull HU6 7RX, UK. Fax: +44 1482 466436
====================================================================
The management here were SO impressed with W Edward Deming's "Out of
the crisis" that they are working flat out to create THE BEST crisis,
so we can start implementing the 13 points.
|