Chang said she’s not surprised by the influence numbers have on behavioral decision-making, but what stands out to her is the robustness of the effect, which was replicated across 21 experiments involving 23,000 randomly selected participants. Despite the significant sample size, the predilection for numbers never wavered, except when numbers were presented in ways that were harder to process. Chang and her co-authors describe the mechanism underlying quantification fixation as “comparison fluency,” or the ease of judging numerical values compared with non-numbers, such as words and pictures.
Katherine Milkman, one of the coauthors, is a scholar I follow pretty closely, and this article seems pretty insightful and I’ll be reading the research this week. Our field has some difficulty here, none no more so in the mixed legacy of Deming on the subject, mostly misinterpretations if you ask me. Mark Graban wrote a great post on that last year.
Risk can be associated with a number of different types of consequences, impacting different objectives. The types of consequences to be analyzed are decided when planning the assessment. The context statement is checked to ensure that the consequences to be analyzed align with the purpose of the assessment and the decisions to be made. This can be revisited during the assessment as more is learned.
Methods used in analyzing risks can be qualitative, semiquantitative, or quantitative. The decision here will be on the intended use, the availability of reliable data, and the decision-making needs of the organization. In ICH Q9 this is also the level of formality.
Risk
Is….
The combination of the probability of the occurrence of the harm
and the severity of that harm.
The effect of uncertainty on objectives
Often characterized by reference to the potential event and
consequences or combination of these
Often expressed in terms of a combination of the consequences of
an event (including in changes in circumstances) and the associated
likelihood of the occurrence
Qualitative assessments define consequence (or severity), likelihood, and level of risk by significance levels, such as “high,” “medium,” or “low.” They work best when supporting analysis that have a narrow application or are within another quality system, such as change control.
Qualitative
Below is a good way to break down consequences and likelihood for a less formal assessment.
Investigations of a Dog 