What Anna Karenina teaches us about u seful n egative r esults Daniel S Katz Senior Fellow Computation Institute University of Chicago amp Argonne National Laboratory Affiliate Faculty CCT LSU ID: 270343
Download Presentation The PPT/PDF document "Scientific research:" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Scientific research: What Anna Karenina teaches us about useful negative results
Daniel S. Katz
Senior Fellow, Computation Institute (University of Chicago & Argonne National Laboratory)
Affiliate Faculty, CCT (LSU)Slide2
"Happy families are all alike; every unhappy family is unhappy in its own way.” -- Leo Tolstoy. Anna KareninaSlide3
Anna Karenina principleDescribes an endeavor in which a deficiency in any one of a number of factors dooms it to failureConsequently, a successful endeavor (subject to this principle) is one where every possible deficiency has been
avoided
In statistics, the term
is
used to describe significance tests: there are any number of ways in which a dataset may violate the null hypothesis* and only one in which all the assumptions are
satisfied
-- https://
en.wikipedia.org
/wiki/
Anna_Karenina_principle
*The null hypothesis usually refers to a general statement or default position that there is no relationship between two measured phenomena, or no difference among
groups
-
-
Everitt
, Brian (1998). The Cambridge Dictionary of
StatisticsSlide4
Scientific research processThe goal of research is provable (reproducible) knowledgeExpressed descriptivelySomething an
observer can read, i.e., in a paper, book, algorithm, etc
.
Expressed
physically
S
omething an
observer can touch, measure,
etc.,
i.e., an
experiment
Expressed
digitally
S
omething an
observer can run on a computer, i.e., simulation or analysis software and required
data
Both
physical and digital results usually also require descriptive
analysisSlide5
The value of research resultsHow do we decide if there is value in a result?Results can be positive or negativeMust be new (and arguably provable
)
Positive results that show something
novel
Positive results that show something already
in
a novel
waySlide6
The value of negative research resultsMany results can show that something doesn't workFirst demonstration that something doesn’t work is novel
May be sufficient for publication by itself
May require understanding of why
Otherwise, want to
provide new understanding of why it doesn't
work
Potentially leading to possible paths to make it workSlide7
Back to Anna Karenina"Happy families are all alike; every unhappy family is unhappy in its own way.” -- Leo Tolstoy. Anna
Karenina
Lesson
To sell a story, it needs to be novel, perhaps by explaining why the family is unhappy