Big Science …
and Little Statistics
Keith Crank, Assistant Director for Research and Graduate Education
W
hat is Big Science? From reading the news- multidisciplinary. This fits well with the sta-
paper, one can readily identify space tistics discipline. A broad-based proposal
exploration as Big Science. Pictures connecting statistics to a number of
from the Hubble Space Telescope and activi- fields, or a narrower concentration
ties at the International Space Station often on one or two other disciplines,
appear on news broadcasts or in the could be viable. STCs are not
newspaper. Occasionally, one hears single institution activities.
about the Large Hadron Collider or While a few of the current
the Laser Interferometer Gravitational- STCs represent a relatively
Wave Observatory (LIGO). And small geographic area, most
there is a lot more that one almost are spread across a large
never hears about, at least in the part of the United States.
general news. The choice of institutions
Why should it matter to statistics to include in a proposal
and statisticians? Big Science means should be based on the sci-
big money. It also means big(ger) rec- ence to be done, not geo-
ognition, which is something statistics graphical proximity.
needs. The discipline suffers from this There are limits to the number
lack of recognition, often getting blamed of STC proposals that can be sub-
for faulty analysis done by nonstatisticians. mitted by an institution (as the lead
The ASA recently hired a director of science
A Hubble Space
institution). If your institution already
policy to help educate members of Congress, their
Telescope photograph
has plans to submit the maximum number
staffs, and other federal officials. But, one person of the Whirlpool Galaxy of proposals, it can still be a partner institution.
will not solve the problem. Much needs to be done (M51). Courtesy of
nasa.gov This allows the inclusion of the appropriate institutions
at all levels to make people aware of statistics and for the proposed center, but may restrict the choice of
what it has to offer. a lead institution. The solicitation for the STC competi-
So, where does Big Science fit in? Statisticians don’t tion is not yet available, but it is expected to be available in
need big machines or instruments to do their work. We the next month or two. Additional information about the STC
have nothing to propose that could compete with projects such as competition is available at
www.nsf.gov/od/oia/programs/stc/index.jsp.
LIGO and the Hubble Space Telescope. There is no Big Statistics This site also has a link to currently funded STCs.
that requires massive amounts of funding before we can make any Not everyone can (or should) be involved in an STC proposal,
more progress. But, that doesn’t mean we should be content with and submission of one or more proposals for a statistics STC is
the status quo. Before there was Big Science, there was big science no guarantee of funding. We need to continue and expand other
and then bigger science. We need to move from little statistics to activities, as well. There should be another competition for math
big statistics. institutes soon. Funding should begin in 2010, which means a
What do I mean by big statistics versus little statistics? At NSF, sta- solicitation is likely within the next few months. The statistics
tistics is concentrated primarily within the Division of Mathematical community should be ready. Multiple proposals are needed to
Sciences (DMS). This creates a model for funding patterned after ensure statistics proposals are reviewed by statisticians, not math-
that for mathematicians, typically one or two investigators on a ematicians. Five to 10 separate proposals from the statistics com-
grant, possibly with a graduate student. We have made some prog- munity would be appropriate.
ress in moving away from little statistics with the establishment of I am not suggesting we abandon individual research in sta-
NISS (the National Institute for Statistical Sciences) and SAMSI (the tistics. That should continue to be the primary form of research
Statistical and Applied Mathematical Sciences Institute). This is a activity. But, we should do more. Math institutes and STCs are
step in the right direction, and it should be continued and expanded. two ways to expand our activities, our recognition, and our fund-
But, it still follows the model of mathematics. ing. Eventually, we may find we do need Big Statistics. And that’s
Let’s move to big statistics. In 2009, NSF plans to have anoth- not just a bunch of BS.
er competition for Science and Technology Centers (STCs). To contact me, send email to
keith@amstat.org. Questions or
Statistics should get involved and be ready to submit one or comments about this article, as well as suggestions for future arti-
more proposals to this competition. STCs are expected to be cles, are always welcome. n
20 AMSTAT NEWS APRIL 2008
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76