MAGNITUDE BIAS:
When interpreting a finding, the selection of a scale of measurement may markedly affect the interpretation (e.g. $1.000.000 may also be 0.0003% of the national budget)1. The choice of measurement scales, whether it be a difference measurement (such as risk difference) or a ratio of odds/risks measure (such as an odds or risk ratios), or a frequency measure (such as percent improved etc) may sway readers into believing the relationship between the exposure and disease is stronger or weaker than it actually is, even when using the same set of data. For example, some randomized clinical trials of new treatments may report a percent improved measurement because the ratio of risks is less impressive for the new treatment. Similarly, a randomized clinical trial may present odds ratio estimates instead of a risk ratio estimate, even if the outcome is common (see Rare Disease Assumption2), and they have prospective temporality, because the odds ratio may inflate the relationship between the treatment and outcome (makes the treatment look more impressive, without altering the data collected in the study). In some cases, medical journals who publish scientific reports, will require that the results be presented on different scales to prevent misinterpretations of the true magnitude of the effects observed. Also see: Information Bias.
1. Sackett DL. Bias in analytic research. J Chronic Dis. 1979;32 (1-2):51-63. (Link to Reference)
2. Porta M, ed. A Dictionary of Epidemiology. Sixth ed. New York, NY: Oxford University Press 2014. (Link to Reference)