There would appear to be no way to answer this question for quite some time. The statisticians are currently analyzing poll results along with exit polls, to see if there is such evidence. A quick search of Lexis Academic / Guided News Search,"general news, " "major papers" limiting to past 6 months "headlines & lead paragraph..." < election! and fraud! and statistic! > gets 25 hits - 3 of them since the election. < voting machin!, fraud! > gets 26 hits, 17 of them later than Nov 2. ( the ! in the searches is truncation; statistic! gets statistic, statistical, statistically....). Many of these are opinion pieces, but some are reports of what people are investigating.

So it is certainly possible to summarize the latest theories, for a brief report or paper - or indeed to START on a very ambitious research project, complete with a survey of books on polls and statistical analysis in voting trends. This would require being close to a good sized library. For example, here at University of Maryland, a "words anywhere" search of our library catalog at catalog.umd.edu gets 31 hits on the search < polls statistic* > (no need to use AND in our catalog - and very many others. Lots of places are adopting the standard that google.com uses - space means AND, double quotes means exact phrase).

From your 734 area code, I would guess you are reasonably close to U of Michigan. The same search < polls statistic* > gets 37 hits in the U. Michigan Catalog.

A number of typical call number patterns show up in this list of 37. The call number for subject "Presidents -- United States -- Election -- Statistics" seems to be JK524. For subject "Social Sciences--Statistical Methods" HN29 seems to be the call number; for "Public Opinion Polls", try HM261. The Ann Arbor Public Library catalog is down tonight, but Lansing got one hit on < polls and statistic* >, and since their call number is around 324.973, I would guess that's a good area to look in if Ann Arbor also uses Dewey classification. Browsing the stacks this way can sometimes be a very helpful way to start a project - or even to do a casual search.

It will no doubt be a number of months before statistical and sociological studies start showing up in scholarly journals. Possibly a good first pass would be Academic Search (Ebsco), since it also covers some more popular magazines, where articles will show up first. Even if this is a serious research project, popular magazines can sometimes mention names of statisticians or other experts, and help you find out where they are working.

Also, the web itself can be very useful for a "hot" and recent topic like this. For example, Google.com Directory (under "MORE") can help locate more precise topics. Examples are:
directory.google.com/Top/Society/Politics/Campaigns_and_Elections/Electronic_Democracy/
and
directory.google.com/Top/Society/Politics/Campaigns_and_Elections/Voting_Systems/.
You can also limit web searches to college and university sites by using "site:edu" or "site:ac.uk" for British universities, etc: <"voting machines" "election fraud" site:edu> gets 196 hits in Google, and is just one of many possible searches.

Finally, many other issues come into this - and will be debated for many years if not generations. Just one of them is the charge that far too few voting machines and polling places were available in some areas, leading to hours-long lines. This would suggest a search of Lexis or Academic Search, (or even Magazine Index in public libraries), using the words "discrimination" or "unequal access", etc. along with the terms "polling places", "precincts", etc; or trying phrases like "minority precincts" along with "delays", "long lines" etc. If this charge turns out to be widely accepted, there will no doubt be court cases in coming years. The actual cases would not show up in legal databases like Lexis / Legal Research for quite some time - but TALK of legal action may show up very soon in newspapers and magazines.

Please let me know if this is confusing, or if we need to try something else.

Jim Miller

Senior Reference Librarian

College Park Patent & Trademark Depository Library (PTDL) Representative

Engineering & Physical Sciences Library

University of Maryland

College Park MD 20742-7011

tel: 301-405-9152

fax: 301-405-9164

jmiller2@umd.edu