Why do we persist in parsing a dead study--"data dredging
," as it's pejoratively known?
By treating the whole modeling process interactively and exploiting high-level knowledge in the form of an ontology, the framework is able to aid the user in a number of ways, including in helping to avoid pitfalls such as data dredging
. Prudence must be exercised to avoid these hazards as certain conclusions may only be supported if, for example, there is extra knowledge that gives reason to trust a narrower set of hypotheses.
, bias, and confounding: they can all get you into the BMJ and the Friday papers.
By doing a lot of data dredging
with the aim of getting some publicity.
It is increasingly apparent that problems such as publication bias, selective analysis and outcome reporting, and data dredging
affect disciplines as remote as clinical medicine, omics, animal studies, economics, social sciences, psychology, and neurosciences.
But Caruana exhibits an uncritical understanding of Imre Lakatos's methodological position, on which the idea is based, when he contends that data dredging
is how science normally works, so Jesuits are off the hook.
This type of data dredging
violates statistical protocol and results in nonsensical, often "significant" results that have little biological meaning relative to proposed hypotheses.
He also cautioned against "data dredging
," in which multiple analyses are likely to eventually find something that is statistically significant.
The most egregious flaw involves data dredging
: Many of the statistical associations lose power when confounders are taken into account.
Such careful thinking and a priori model specification renders unnecessary data mining (and its pejorative cousin, data dredging
), which is now widely advertised by purveyors of statistical software.
Henry states that we (Baroudi, Olson, and Ives) "have fallen into the common trap of thinking that path analysis can aid in determining the direction of causality in a multivariate model." If I understand Henry correctly, he is saying that path analysis is not a data dredging
Clinical trial investigators were offended by what they viewed as post hoc data dredging
on the part of CMS officials concerned more with costs than what the data showed.