CLPEX Chat Board

I am considering doing a PhD after finishing my degree.

I have been working on microscopic analysis of fingerprints and examining whether details such as the pores are repeatable through different inked impressions of the same finger but I would like to extend this to examination between multiple subjects.

I have been reading works by Simon Cole and I was wondered what would be constituted as a sufficient sample size that would be approprate in a court of law to consider fingerprints unique? As much of what I have read has stated that there has been no "scientific" analysis of the uniqueness of fingerprints.

I am hoping to pursuade each subsequent student from forensic courses at my university to donate their fingerprints. Each one would go into the records with a sex, year of birth and ethnic origin but no names. These can be then analysed on a number of levels.

Then I was hoping to network with other universities around the country and get samples from them.

I just wondered how many samples would be enough?

Thanks for any insights, or funding (I can wish can't I?)

Emma,
The fly in the ointment on something like this is always going to be the word "Scientific".

Millions of fingerprints have already been compared and analyzed over the last 100 years and all have been found to be unique. Dr. Cole will tell you that this does not constitute scientific research. I believe he calls it anecdotal evidence. So the actual number of subjects you compare or the length of time you conduct your research may not be the issue. It's going to be just how "scientific" was the research. Also, if the research is to be considered truly scientific, you will not be able to state your conclusion in any type of an absolute.

You should be prepared to have your work criticized by the "science" people if the give a conclusion of absolute uniqueness. Even a conclusion of 1 in 10 to the 97 power will be deemed unscientific. I believe the FBI is doing this same sort of thing right now on the permanence of fingerprints. But you might want to check around a little bit more, I'm sure there must have been some "scientific research" somewhere on fingerprint uniqueness.


Cheers, Steve