Bilz, Psychology and the law, Fall 2013
· Clinical Psychology- Goal is to heal. Find dysfunction and fix it.
o Cognitive and social psychologists- What to know who normal people work. Empirical testing.
· Three Theorists:
o Classical Law Economics (conventional)
§ Humans are rational.
§ Humans value the maximization of utility (single unitary function)
o Behavioral Law Economist
§ People are irrational (mistakes are random)
§ Humans value the maximization of utility
o Expressive Law Economist
§ People are rational
§ Human have plural values (no single unitary function)
· Muller v. Oregon (bakers and working hours)
o Real world empirical data was cited. Problems:
§ Making arguments in legal terms obscures normative values
§ “General” Knowledge
§ Institutional competence (judges, adversary system)
· Brown v. Board of Education (different from Muller)
o Martian coming to earth. He/She would need social science to find out why they are unequal
o Trial Level- Barbie experiment
o Researchers might be cherry picking research
· Roper v. Simmons- Death sentence to juvenile?
o Reifies the differences (lack maturity, easily influenced) and therefore it translates into a legal conclusion, which is slight of hand. Making normative conclusions.
· Limits of Social Science and the Law
o Legal academics are not institutional competent to conduct this research
o Cherry picking evidence for cases
o Couching contestable beliefs as if they are facts
o Normative question: What should we do?
· Value of Empiricism (John McGinnis)
o Theory 1: Politics are less contentious (less fistfights). McGinnis: stats are more convincing to people and make us agree.
§ Problem:1 Assumption: computers and training in research methods will make us agree. (Many empiricists have different research methods)
§ Problem 2: He characterizes the question to get a specific answer (do you like safety)
· Does the death penalty deter?
o Debate started in the 60s and no one agrees till this day
o Take race out of the equation and determine it based on priors.
o Regression analysis is a statistical process for estimating the relationships among variables
o Looked at cases from US. Want only cases from the same state.
· Attacking a Regression Analysis (SAW, NES, CU)
o Sample (relevance). You need to have a theory attached to it.
§ Sample size
§ Type: is the population different than the one you want?
o Correlation does not mean causation
§ Fix the problem by doing a controlled experiment
o Unknown third variable may exist
o Why certain variables were chosen and others left out (cherry picking)
o Statistical Significance (how real is it)
§ Significance of the findings and the effects side
§ Significance reveals whether an effect is real (coefficients)
§ It can either be real or fake. If the number drops, it is still significant
o Effect size
§ How big is the effect (death penalty, etc.)
o Noise (when subjects come in, they are noisy)
§ Need many cases for noisy data
§ Captured by error term
§ Noise consist of errors that are not predictable with the information you have (they are random).
o Are you measuring what you say you’re measuring?
· What is an experiment?
o Experimenter manipulates the independent value.
o Big regression analysis (10,000 people)
o If it is real, you will eventually find it (more impressive with 40 people, because you easily found it without big numbers)
· Underlying assumptions of Economists
o Conventional Law Economists
§ Law of supply and demand (makes items more or less expensive.
§ A rational actor doesn’t care who or how offer is made
§ Utility function is a single metric
· Expected utility function- how much it is valued to the likelihood of how its going to come about
§ Marginal utility- Utility curve for any good isn’t straight
· Discount rate example (1 dollar off is the same across the board)
§ Human is a rationalizer and wants to maximize
o Objections to CLE
§ People are bad with math. Criminal not thinking about the numbers
§ Assume people are selfish (helping others may be part of the utility)
§ Right objections- The freer you are, the less distinctive economics become as a discipline
§ The more you add in (in utility), the harder you can predict people’s behavior.
· First instinct is to say “add it to utility curve,” its
value the extrinsic good that the extrinsic good affects (wearing ugly bracelet because it represents us)
o Rationality is measured by asking. “Are their actions expressing their values,” instead of whether it is maximizing the utility.
o You look back and intentions
o Expressivist theories are backward looking. Other core features:
§ Cannot just enrich utility curve
o CLE- No place in rational thought (useless)
o Expressive- It has a place because we have multiple values that cannot all be expressed at the same time
· Which explains people’s values? (Depends on situation)
o When the benefits are obvious, personal, and immediately, sizeable, then conventional economists have a better argument. (Evaluating policies, employment decisions)
o Danger: for both, it becomes untestable if you over enrich the utility curve.
· Who is right in regards to why people support or oppose gay sodomy laws?
o Hardwick- law is okay for arrest. Took a conventional approach
o Lawrence- Reversed Hardwick. Took an expressivist approach
· Why do we commit crimes?
o CLE- Assume people obey because of the consequences
o BLE- Whenever we do something out of the ordinary, you think somebody is watching you.
o Grogger- Portion of population does react to incentives, which effects if people commit crimes.
· Why do we obey?
o Authoritism (impulse)- we are hard wired. Impulse to obey.
o Milgram- Why do people obey Hitler? Electroshock case
§ DV is level of obedience. IV (has to be comparable) are the teachers/learners
§ Worst experiment. It can’t be revealing unless your comparing it to something.
§ Questioning experiments: You say the problem is generalizability. What about the real world? Nazis would get killed if they did not comply.