Post date: Jan 30, 2014 9:12:34 PM
The statistical analyses discussed here, as well as quantitative elicitation in general, are important components of modernized health care and communicating their relevance is of no small significance [1]. Indeed, even routine medical screenings can require an interpretation of relatively complex statistical statements [2]. Otherwise well trained doctors are often times left ill-equipped or ill-educated in the ways of communicating or understanding these data [3] but are nevertheless charged with disseminating the information contained within. Risk communicators have developed a plethora of methods for communicating risk(s) and other quantitative and probabilistic information [4], but little progress has been made in terms of communicating complex statistical statements containing fundamentally different types of uncertainties [5]. We must therefore consider whether or not these methods are likely to succeed in informing decisions made by professionals and lay people alike.
The natural frequency is one of the more notable communication techniques(objects?) used for communicating frequency and/or probabilistic information. The natural frequency sidesteps many known cognitive biases that can muddle the information contained in probability or frequency statements [6]. Reference class ambiguity, overconfidence biases, the conjunction fallacy as well as other cognitive biases largely disappear when a frequency representation of probabilistic information is used. Some have gone so far as to suggest that humans have evolved in an environment where mentally representing probabilistic information as frequencies is even expected [7].<<maybe expand here>>
[In spite of this] natural frequencies are not free of all cognitive biases. Indeed, the denominator of a natural frequency can bias the interpretation of a probability and/or frequency value depending on its associated magnitude (citation)<<need to add here>>. It is, perhaps counter intuitively, the communicative characteristics of the natural frequency which allows it to act as conduit for communicating complex statistical inferences derived from confidence structures <<need to expand here, maybe generalize for all methods mentioned>>. The confidence structure-derived binomial inference encodes information about the uncertainties relevant to sample size and success rate into its result and can likewise transmit this uncertainty information into a natural frequency <<overly simplified, perhaps error(s) of omission or just plain wrong>>. <<Using some math wizardry>> we can fit an equivalent binomial count to its associated confidence structure and still capture a complex account of the existing uncertainties. This “transformation” facilitates the communication and understanding of complex forms of statistical information <<need to greatly expand here>>.
<<It is for this general reason why we must consider how useful these methods will be when they are charged with informing decisions that are to be made by professionals and lay people alike.>>
<<Otherwise well trained doctors, of which whom are charged with disseminating the information in the resulting analyses, are often times ill-equipped or ill-educated in the ways of communicating[[,]] or [[even themselves]] understanding these data (citation).>>
1. Schapira, M.M., A.B. Nattinger, and C.A. McHorney, Frequency or Probability? A Qualitative Study of Risk Communication Formats Used in Health Care. Medical Decision Making, 2001. 21(6): p. 459-467.
2. Kerlikowske K, G.D.B.J.S.E.A.E.V., Likelihood ratios for modern screening mammography: Risk of breast cancer based on age and mammographic interpretation. JAMA: The Journal of the American Medical Association, 1996. 276(1): p. 39-43.
3. Gigerenzer, G., et al., Helping Doctors and Patients Make Sense of Health Statistics. Psychological Science in the Public Interest, 2007. 8(2): p. 53-96.
4. Tucker, W.T. and S. Ferson, Strategies for Risk Communication. Annals of the New York Academy of Sciences, 2008. 1128(1): p. ix-xii.
5. Spiegelhalter, D., M. Pearson, and I. Short, Visualizing Uncertainty About the Future. Science, 2011. 333(6048): p. 1393-1400.
6. Gigerenzer, G., How to Make Cognitive Illusions Disappear: Beyond “Heuristics and Biases”. European Review of Social Psychology, 1991. 2(1): p. 83-115.
7. Cosmides, L. and J. Tooby, Are humans good intuitive statisticians after all? Rethinking some conclusions from the literature on judgment under uncertainty. Cognition, 1996. 58(1): p. 1-73.