Unless you are a scientist, you probably do not realize that analysts in forensic science exercise a massive amount of discretion with the reporting of data.

In a thought provoking article that appeared on www.SeparationsNow.com and in their “Lab Infomatics” section reads:

It’s the analytical scientist’s perennial dilemma: where to strike the balance between convenience and completeness. Do you analyse the full set of data produced by a mass spectrometer, even though that could well comprise millions of different data points, or do you try to simplify the data first, making the analysis easier but potentially losing information.

In other words, do we need to hear the entire song or will only a couple of opening bars of the song suffice?

Some call this exercise of discretion in analysis and in the corresponding editing and presenting of data an essential act. Some say that it is indeed part of the proper exercise of being a trained professional scientist.

In fact, I have blogged on it before:

• Exercise of Discretion: Sampling versus Sample Selection
• Which bag do you test? Do you test all bags? How much of each bag do you test?
• Why doesn’t your state crime laboratory use AMDIS?
• The case for raw data: “Integration” in Gas Chromatography: How to make an innocent person guilty in a DUI case by manipulating the software

Perhaps it is and it maybe even unavoidable, but shouldn’t the raw data be available to compensate? Isn’t science characterized by verification? What do you think?