liveBook · Manning

Process assumption

As explained in Grokking Statistics

An implicit or explicit belief about how a data-generating or analytical process operates.

A process assumption is a belief—stated or unstated—about how data arise or how an analysis works. Questioning it helps separate a real effect from a convenient story. section 1.4

When evaluating a statistical claim, inspect:

  • Whether sampling bias, ignored variables, or aggressive outlier removal changed the result. section 1.4
  • Whether incentives or funding could favor a particular conclusion. section 1.4

The pricing-system success story

A director attributes $70 million in extra revenue to a new pricing system because the analysis is statistically significant. A competitor’s retreat from several markets offers another explanation: newly acquired customers may account for the increase. The significance calculation does not establish that the pricing system caused it. section 1.4

How Grokking Statistics covers it

How deep it goes
Contrasts process assumptions with sampling bias, ignored variables, outlier removal, and funding, but does not formalize or test them.
Examples
Uses conflicting egg studies and a pricing claim potentially explained by competitor retreat.
What you can do after
Diagnose statistical claims by checking alternative explanations, bias, omitted variables, outliers, incentives, and funding.
How it connects
Links process assumptions to sampling bias, statistical significance, ignored variables, funding, incentives, and research ethics.
Cover of Grokking Statistics
liveBookmanning.com
Read Grokking Statistics on liveBook Get the book at manning.com

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Grokking Statistics. The section links point to the parts of the book it draws on.