Let me suggest some additional factors that should be
considered on whether data has "aged out" or is still
relevant:
(1) What was the research question addressed by the
research and has the theoretical and/or empirical basis
for the research question changed since the research
was first done? For example, today we know that the
initial formulation of the Levels/Depth of Processing (LoP)
theory is incorrect, that is, sensory memory representations
are not more "fragile" than semantic (abstract propositional)
representations. The work on Douglas Nelson on his
Sensory-Semantic model and other results such as the Picture
superiority effect show under which condition sensory based
memories are either remembered as well as semantic based
memories or are superior to them. One important factor here
is whether interference effects (e.g., similarity of stimulus
items) are induced by the stimuli used. Recognition memory
for picture is typically significantly higher than recognition
memory to equivalent word EXCEPT when the pictures are
highly similar to each other -- recognition memory for such
pictures is reduced. So, if one has data that one collected
in support of LoP, one has to determine whether it actually
provides either new information that current theoretical
explanations need to take into account or whether it is more
of the "same-old, same-old" results. In the latter case, the
data and results have reduced/little relevance.
(2) Do the data come from an experimental study or
nonexperimental (i.e., quasi-experimental, correlational,
survey, observational, etc.) research? If one has data
from an experiment with high internal validity and addresses
a still interesting question (how interesting is a relevant
parameter), I don't see why the research can't be published
especially if not much experimental research has been done
in the area. I feel somewhat less sure about nonexperimental
research unless it provides insight into some phenomenon
at some earlier point in time that has not been adequately
reviewed or analyzed (e.g., attitude studies of U.S. citizens
toward fascism during the 1930s and prior to the Pearl
Harbor attack). But if the research has low internal validity
and other types of validity, I think age may not be the most
important problem or objection to the research.
(3) Clinical vs non-clinical research: my own feeling is that
research in traditional areas of experimental psychology
(i.e., learning, perception, psychophysics, problem solving,
reasoning, etc.) may provide support publication of an old
research study. I am less sure about clinical studies but that
may just be my own bias as an experimental/cognitive psychologist
who admits to having committed clinical research at various
points in time of my career (I'm trying to stick to the straight
and narrow). Again, one has to determine if the research is
still relevant or not, have subsequent research gone beyond
the questions addressed in the original research.
So, I think one would need to know a lot more about the research,
where it falls in a research tradition, and whether it still has
relevance to the field.
Of course, one should also ask WHY one wants to publish
old research data. Does one just want to get another entry
into their publications list (for tenure/promotion/academic
reasons)? If so, go ahead -- there are more than enough
publishers out there who will publish stuff especially if you
pay for it.. If one thinks that the research really does make
a significant contribution to an area and is still relevant, then
the fact that it is old research is not really relevant. If the
research is good from a design and analysis perspective,
if it addresses questions that are interesting and still relevant,
then I think one should try to publish the research. Just
make sure that the results you want to present are not contradicted
by subsequent research because you will have to explain
this discrepancy ("well, i did this before so-and-so did their
study and found the opposite result which has be widely
replicated -- I thought that my research still should become
part of the knowledgebase" -- don't be surprised if people
are not convinced by this argument).
-Mike Palij
New York University
[email protected]
On Wed, 13 Jan 2016 09:23:47 -0800, Miguel Roig wrote:
Dap, you raise a very interesting question and I look forward
to others' opinions. I suppose that whether the data are deemed
worthy of publication will depend on several factors, such
as the degree to which the data are relevant/applicable
(e.g., attitudes toward same-sex marriage in the US). I did
some searching and found a couple of discussions that may
be of relevance.
https://www.researchgate.net/post/How_late_is_too_late_to_publish_clinical_research
A related question is whether authors should be required to
disclose when their data were collected.
On Wednesday, January 13, 2016 10:33 AM, Dap Louw wrote:
Tipsters
I'm looking for guidelines on when gathered data is regarded
as too old to be used for an article in a scientific journal. I
specifically
refer to Psychology. I fully realize that it's relative and will be
determined
by several factors, but I would still appreciate your opinions. Maybe
a few examples could also help.
---
You are currently subscribed to tips as: [email protected].
To unsubscribe click here:
http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=47851
or send a blank email to
leave-47851-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
