Ten Propositions about Data Analysis *


1) Data analysis ability has the character of a skill that can be
acquired and developed with practice.  Like a skill it cannot be fully
explicated, or absolutely established.

2) Transfer of skill-like knowledge is capricious.

3) Skill-like knowledge travels best (or only) through accomplished
practitioners.

4) Data analysis ability is invisible in its passage and in those who
possess it.

5) "Proper working" of a data analysis, parts of the analysis,
_and_the_analyst_ are defined by their ability to take part in
producing the "proper" analysis result.  Other indicators [of
"correct" analysis/analyst] cannot be found.

6) Scientists and others tend to believe in the responsiveness of data
to manipulation directed by sets of algorithm-like instructions.  This
gives the impression that carrying out data analysis is, literally, a
formality.  This belief, though it may occasionally be suspended at
times of analysis difficulty, re-crystallizes catastrophically when a
proper result is obtained from the data analysis.

7) When the normal criterion -- a proper analysis result -- is not
available, scientists disagree about which data analyses are
competently done.

8) Where there is disagreement about what counts as a competently
performed analysis, the ensuing debate is coextensive with the debate
about what the proper analysis result is.  The closure of debate about
the meaning of competence is the 'discovery' or 'non-discovery' of a
new phenomenon.

9) Decisions about the existence of phenomena are coextensive with the 
'discovery' of their properties.

10) In the long term, phenomena with radical ["considerable departure
from the usual or traditional"] properties can exist only within
forms-of-life and sets of institutions which overlap minimally with
science as a whole.  Otherwise, either the phenomena, or science, must
change.



 - - -
* These have been adapted from the "Ten propositions about
experiments" on p.129-130 of "Changing Order: Replication and
Induction in Scientific Practice" by H.M. Collins,
1992 edition (1985 originally), University of Chicago Press.
-------