Lack of significant associations at lower exposures may be attributed to methodologic issues (e.g., inadequate statistical power, exposure misclassification), or to differences in the dose response relationship at high versus low exposures.
The presence of this kind of radiation has primarily natural reasons (the statistical annual exposure dose is around 2.4 mSv).
In statistical literature, exposure misclassification and miss-measurement are known as a measurement error problem and a plethora of approaches exist to correct for biases that arise from it under certain assumptions [ 4, 5].
Although false-negative findings are a possibility, whether due to insufficient statistical power, exposure misclassification, confounding, other biases, or chance, the consistency of null findings suggests that the absence of association is real.
These require experience constructing and criticizing models of biological phenomena and an understanding of probability theory, as the underpinning of inferential statistics, and exposure to statistical methods used to establish confidence and test hypotheses.
For statistical analysis, exposures were dichotomized into "ever" versus "never," and the "number of days exposed" was calculated.
The independent action (IA) model depends on statistical independence between exposures, i.e. each exposure acts independently of the other exposures but they all contribute to the outcome.
For statistical analysis, the exposure time of the CCD was fixed, yet for display the exposure time of the camera was adapted to keep a high SNR.
"A lot of what we see in our injury data is almost certainly due to a statistical measure called exposure hours," Dr. Lawrence says.
Statistical comparison between exposure group(s) and control group(s) was made individually for each time of day (large circles), for pair-wise comparison between time points (oval shape between large circles), and over all time points (center circle).
