Rebuttal to Prof. Gilat-Schmidt’s response

Discussion in 'Aviation Passenger Security in the USA' started by TSA News Blog, Jun 17, 2012.

  1. TSA News Blog

    TSA News Blog News Feed


    In rebuttal to Professor Gilat-Schmidt’s rebuttal, I would like to summarize my view of her response. And yes, it’s a bit strong:
    “Garbage in, garbage out.”
    While I sympathize with her position in having her findings misrepresented in the LA Times and appreciate that she supports additional testing of these devices, her rebuttal fails to adequately address the flaws in her study.
    Though her paper may have been “peer reviewed,” it can only be reviewed for the analysis of potentially flawed, if not fraudulent, data. While she may have gotten the math right, the underlying test information cannot be duplicated, failing the first requirement of scientific method, i.e. the test must be replicable. Since the base tests are not replicable and the data are not readily available, the study fails to meet two requirements of scientific method principles and is thus invalid. The presence of a peer review only meets one of the three requirements and is insufficient.
    Peer review of the analysis of a flawed data set does not validate the base claims of the data, only the calculations and techniques used in analyzing that data. This review may include such things as the researcher’s selection of the correct statistical test for sample. For instance, a reviewer may be arcane as to whether the researcher chose to use a Chi-squared test or a Two-proportion z-test in analyzing the results. These issues are germane to the data interpretation and usually only of interest to academics and scientists.
    Gilat-Schmidt’s rebuttal also ignores the issue of a statistically significant sample, which I conservatively estimate to be around 60, assuming the number of backscatter (as opposed to millimeter wave) units is only 150 of the 700 scanners in service. Many believe there are far more backscatter scanners than that, which would require that the sample be larger than 60.
    But we don’t know how many of the scanners at airports are backscatter scanners, because the TSA refuses to tell us. That lack of information adds yet more doubt to any study on backscatter safety or performance.
    I recognize that her research was focused on how deeply the x-rays penetrate internal organs and to what degree, but the base data remains in question. Therefore, the analysis of those results remains an exercise in futility, especially without independent testing of the scanners in service.
    Gilat-Schmidt also does not mention why she doesn’t allow her children to use the backscatter scanners. Is it because of the radiation exposure or because the backscatter units still produce a naked picture of her child that is being viewed by some anonymous TSA agent in a back room?
    (Photo: Flickr Creative Commons/Sean MacEntee)
  2. barbell

    barbell Coach Coach

    Excellent piece, as usual, Fisher!
  3. Mike

    Mike Founding Member Coach

    But completely off base.

    The purpose of her study was to model (given the previously published radiation levels, the ones TSA keeps saying are "safe", less than 2 minutes' exposure to in-flight radiation, etc.) just what that type & amount of radiation can reach in the human body.

    Up to now people have claimed that it only reaches the skin. She showed that to be incorrect, noting that her modeling shows that the radiation which TSA has admitted is produced by the machines actually reaches 29 internal organs. That pokes a really big hole in TSA's safety claims.

    If she could redo her research after an honest outside evaluation of the x-ray scanners emissions, the results could be dynamite.
  4. Lisa Simeone

    Lisa Simeone Original Member

    Great comment, Mike. Worth leaving at the post at TSA News.
  5. Fisher1949

    Fisher1949 Original Member Coach

    Good point, wish I had thought of that.

    Nonetheless, if we can't verify the data we don't know if it reaches internal organs or to what degree. Can't prove anything one way or another with bad data.
  6. Fisher1949

    Fisher1949 Original Member Coach

  7. Lisa Simeone

    Lisa Simeone Original Member

Share This Page