6.3 The Real World

Identifying Assumption: An assumption you make about your causal diagram or data that, if incorrect, means that your approach to identifying the answer to your research question doesn’t work.

Police Presence and Crime
Police Presence and Crime
Some Assumptions
  • LaggedCrime doesn’t cause LawAndOrderPolitics
  • PovertyRate isn’t a part of the DGP
  • LaggedPolicePerCapita doesn’t cause PolicePerCapita (or anything else for that matter)
  • RecentPopularCrimeMovie doesn’t cause Crime
Typst Code 
= Figure 6.6: Police Presence and Crime
#diagram(bezier: true,
  node(<Crime>, fill: blue.lighten(50%)),
  node(<ExpectedCrimePayout>, fill: gray.lighten(75%)),
  node(<LaggedCrime>, fill: blue.lighten(50%)),
  node(<LawAndOrderPolitics>, fill: gray.lighten(75%)),
  node(<PolicePerCapita>, fill: blue.lighten(50%)),
  node(<SentencingLaws>, fill: blue.lighten(50%)),
  
  edge(<ExpectedCrimePayout>, <Crime>, stroke: gray.lighten(50%)),
  edge(<LaggedCrime>, <Crime>),
  edge(<LaggedCrime>, <PolicePerCapita>),
  edge(<LaggedCrime>, <SentencingLaws>),
  edge(<LawAndOrderPolitics>, <PolicePerCapita>, stroke: gray.lighten(50%)),
  edge(<LawAndOrderPolitics>, <SentencingLaws>, stroke: gray.lighten(50%)),
  edge(<PolicePerCapita>, <ExpectedCrimePayout>),
  edge(<PolicePerCapita>, <Crime>),
  edge(<SentencingLaws>, <ExpectedCrimePayout>),
)