Action Theories Flashcards
What are the core assumptions of RCT?
Focus: Utility-maximizing proposition
- Expectations/beliefs enter as a third determinant of behavior (besides preferences and constraints) in situations of risk
1. sometimes the consequences are very clear then exp/beliefs are not as important (consumerism for example, customers know the value of sth and just pick and choose given their budget constraints, but very limited simplistic model
2. usually in social sciences, we face situations were there is risk involved (i.e. migration, marriage, criminal behavior - we hope for the best, but only have beliefs) - Some advocates of RCT stress utility maximizing as human law, true for every individual, everywhere, in every time period (satisfying the Hempel-Oppenheimer scheme)
- Critique: very void, very general, in reality, compatible with all kinds of behavior, how much does it actually buy sociologists?
Example: Classic rational choice theory of crime (Becker 1968)
Equation
EU[S] = q×B – p×C
i. EU(S) = expected utility of crime
ii. B = benefit
iii. q = probability of success
iv. C = severity of sanctions
v. p = sanction risk
Example: Classic rational choice theory of crime (Becker 1968)
hypthesis of deterrence
Gives a motivational, more psychological explanation of deterrence
Hypothesis: The perceived severity of sanctions should affect the willingness to engage in an offence stronger, the higher the perceived risk of getting caught
So, if risk of getting caught is getting higher, people - even if highly motivated - will abstain from it bc too dangerous
Example: Classic rational choice theory of crime (Becker 1968)
Why overly simplistic?
- Different crime propensities of individuals (risk-seeking, morality, impulsivity)
- Parameters include momentary/situational effects? Interesting whether in the heat of the moment you can take cost vs benefit into account - Can model account for that?
- Costs are much broader than the severity of sanctions (loss of job, friends, etc.) -> consequently “sanction risk” can vary drastically
- Overall, very narrow model of crime (and RCT) = just egoistic preferences & only tangible consequences are relevant (i.e. money or sentence)
Narrow vs Wide
- Iannaconne & Becker -> quite narrow
- but there is different version of RCT = wide version, where all kinds of additional soft or psychological incentives (i.e. shame, bad conscience, thrill of risk) are possible -> you can incorporate basically everything
- Extremely flexible but you lose a lot of control -> utility maximizing becomes completely empty if everything can become utility maximizing rational behavior even i.e. “fun, a kick or rush” (Thomas, 2022) which seem to be by nature quite irrational behavior
Rational Choice as an approach
- Start with core assumptions -> RCT model of X (i.e. religion, crime, etc.)
- But there need to be answered some question first, bc theory itself is so general due to its degree of abstraction, to apply it to empirical phenomena we therefore need to specify them in form of auxiliary assumptions (i.e. what are their preferences, the constraints?)
- Big question: What is or can be the source of those?
1. Background knowledge (experts)
2. Previous research, literature
3. Qualitative data (i.e. if interested in motivation behind crime, you can conduct interviews with prisoners) - What is the problem with this? Difficult to refute (maybe cannot be refuted at all) bc we never know if just our auxiliary assumptions were wrong or we forgot to include sth important or core assumptions (especially in social sciences) true for all general theories
-> Rational choice = approach to construct testable models rather than a testable approach itself, too broad too general itself
RCT = Catch-22?
- Wide RCT -> metatheoretical: anything goes here - so not restrictive enough
1. Critique: measurement, tautology, circularity, emptiness, etc arguments
2. Opp’s defending (1999): “correct that infinite number could be taken into account, but given a certain explanation, RCT provides clear guidelines for their selection” weather could be important for some explanations (voting-behavior) but not for others (tax-fraud)
a. BUT what are those clear guidelines?
3. Becker: starts with the assumption that behavior is rational self-confirmation bias
Difficulty of testing theories
- Very long chain, a lot of possibilities for error
1. Theory needs to be translated into testable implications
2. Research design (i.e. experiments) that don’t give you headaches about causal identification, validity
3. And even if you have a data matrix with certain numbers, to apply rational choice model we need a statistical model to summarize data matrix into a certain coefficient and then we can look at coefficient, is it significant? - Therefore: We cannot test theories as such, only the combination of a theory and all auxiliary assumptions (incl. measurement hypotheses) -> makes “theory testing” a difficult task for researchers (and continues to keep philosophers of science busy)
Test strategies
Direct
- So we try to measure the utility of A and B, i.e. fear of being caught, severity of sanctions, etc. -> than compute the equation
- Example: political participation being measured by several items which are computed into an index
- Critique: not very reliable, today I might say 2 tomorrow 3, not high quality of measurement - some degree of measurement error, for some it is just noisy for others unacceptable + sometimes, direct strategy are just not available
Test strategies
Indirect - How to?
- Engage in bridge assumptions, utility for action A is greater than for action B, i.e. depending on the neighborhood, in one neighborhood perceived risk is higher as more police control but just related by assumption not by measurement!
- Could also be that one lives in a neighborhood but spends all of his time in a certain other neighborhood partying, much more time in criminogenic areas than other people from my neighborhood
- Further engage in auxiliary assumption: young male persons are more likely to be exposed to these criminogenic areas gets more and more indirect
So, by specifying the theory with auxiliary assumptions, it becomes measurable
Test strategies
Indirect - Advantages & Disadvantages
Advantages
- Easier measurement
- Objective factors of intrinsic sociological interest: measurement is confined to the behavior of interest and other objective factors, such as group size, family type, or social class.” (p. 77) –> this is what is sociologically more interesting bc we emphasize the macro-level (oppositely, we don’t care about people’s answers on a questionnaire - too psychological)
- Lower data requirements (wider applicability)
Disadvantages
- Greater reliance on auxiliary assumptions –> So, you ask yourself: is the specific implication from RCT or could any theory predict this? Direct test strategy allows for more specific testing, here you rely on many assumptions but sometimes there are just no direct indicators
- Theoretical ambiguity: Sometimes, objective factors affect parameters of actors’ cost-benefit calculus that work in opposite directions
(Ex: greater income greater costs of being convicted of tax fraud, but also greater potential savings through tax fraud = difficult to tell what the average effect of income will be)
Hypotheses are often consistent with other theories as well
Lindenberg’s goal framing theory
starting point
Assumption of bounded rationality: Humans are unable to take into account the full set of alternatives, preferences, and constraints - we only take in a limited amount, so starting point is RCT again but we don’t look at full set of alternatives, preferences, constraints - we simplify our situation, concentrate on specific goals
How does goal-framing work?
logic 1
Goal frame = a general goal orientation that can be situationally activated and selectively guides behavior in a situation, so can switch from situation to situation
- Three goal frames:
1. Hedonic goal: Seeking pleasure here and now (humans just like animals)
2. Gain goal: Investing in long-term goals (less present-focused)
3. Normative goal: Doing the right thing / acting in line with norms - One goal frame dominates, the others work in the background
- The default activation level is greatest for the hedonic goal, followed by the gain goal and then, lastly, normative goal -> which means, as a society, to ensure order, investments in future and norm conformity, humans are in need of institutions (i.e. rule of law, educational system), social control (i.e. friends, neighbors), and self-control that strengthen the gain goal and normative goal
Injunctive vs Descriptive Norm
- Injunctive Norm: Rule / beliefs of how one should behave (i.e. don’t steel, don’t smoke inside)
- Descriptive norms: Behavioral regularity / how people behave (i.e. don’t speak during a lecture but students are tuscheling)
Field experimental test of the cross-norm inhibition effect
Setup
- Alley in Groningen located in a shopping area and commonly used to park bicycles. Participants (N = 77 in each condition) were all people who came to collect their parked bicycles.
- “In their absence, a flyer with an elastic band had been attached to the handlebar of their bicycle. The flyer was white and thus very noticeable. It read: “We wish everybody happy holidays,” signed with the name of a nonexistent sportswear shop. The flyer had to be removed by the participant to easily use the handlebar. Because there were no trash cans in the alley, “not littering” meant taking the flyer with them. We counted throwing the flyer on the ground or hanging it on another bicycle as littering.”
- Difference = Wall without and with graffiti ( conjunction between descriptive and injunctive norm due to graffiti not allowed sign)
Explain: Model of Frame Selection (graphic)
Activation weight (AW) is greater and spontaneous decision more likely
1. the more clearly the situation could be defined (mi )
2. the more accessible the script (behavioral repertoire) is (aj|i)
3. the more strongly the script is internalised (aj )
4. the more strongly it regulates action (ak|j)
Voting example
- (Civic duty) Norm = participate in elections (Only relevant if there is an election + fair election)
- Model predicts: if situation is clear (election day) + script is accessible and strongly internalized, it won’t matter if it’s raining or whether the ballot is close or the election is very clear counter to RC bc they would say everybody weighs in the factors above
3 additional factors important for AW:
FMS vs RCM
- The first 4 are how strongly a behavior is situationally activated BUT even if you have a strongly activated script that regulates your choice, you might still be motivated to think carefully whether to migrate or not (opportunity to reflect) -> in high-cost situation people tend not to act spontaneously which led some RC scholars to believe that RCT is for high-cost situations
- Also, RC for initial decision: we once weighed cost & opportunity for a situation (i.e. way to university) but then it became habitual and no we don’t think about it anymore (standing decisions)
- Dual process model could also be interpreted as a RC model to decide when to engage in RC decision (meta decision), if we can avoid thinking about something we avoid time and energy - routines are efficient
Theorizing culture
How do Frames/Repertoires connect to Social Structure & Social Behavior?
Problem (1): The link between structure and culture
- There are certain social structures that affect your frames and repertoires –> how much do ur frames/repertoires depend on where you come from socially? And how do they govern your behavior?
- Reductionist view: We don’t have to think about culture independently, it’s all structure, culture just reflects structure (e.g., Marx)
Problem (2): Action theories
- So, can we just take RCT or do we need to account for spontaneous impulses, goal frames, etc.
- Reductionist view: cultural influences = just another form of incentives and opportunities (rational choice approach)
Theorizing culture: Dynamic view
(1.) Periods of structural change
In periods of a society where there is a lot of structural change (i.e. modernization), you’re established ways of doing/seeing things ergo frames and repertoires (i.e. gender roles) may no longer fit the reality of society (i.e. labor market) actors tend towards more reflected decision making, start to ask what is optimal? (i.e. reduced number of children) = more incentive-driven, Changed incentives and opportunities shape behavior to a greater extent
Theorizing culture: Dynamic view
(2a.) Period of greater stability
BUT as soon as there is greater stability, these thoughts and decisions become encoded in frames and repertoires Formation of new mental models, institutionalization
Theorizing culture: Dynamic view
(2b.) Period of greater stability
- However, in this situation different people/bases, some more incentive-based others might be more frame-based very hard to distinguish between incentive-based decision or decision based on cultural programs
- Encoded cultural models become an independent determinant of social behavior. There is heterogeneity:
i. Taken-for-granted exclusion of action alternatives
ii. Value/norm-based exclusion of action alternatives
iii. Reflected deliberation on a greater set of action alternatives
Theorizing culture: Dynamic view
(3.) New period of structural change
- ONLY if situation changes again - exogenous shock - cultural influences becomes: Actor do not (immediately) respond to changed incentives and opportunities, will hold on to norms for certain period of time and then maybe think about it again
- Thesis of cultural rigidity / “cultural lag” / “hysteresis” (e.g., Marx, Bourdieu)
- If society is stable, you might think of everyone maximizing their utility by weighing cost and benefits, however, they are often just following norms that have proved to be rather beneficial to actors - only visible that they are normative-based actions once the situation changes, then you see people don’t optimize for the best (i.e. changed bus schedule, gender roles) - takes time to change the frames and repertoires they have been raised with
