A person’s choice to abandon a norm depends on their sensitivity to the norm, their risk sensitivity, their risk perception and how many others are following the norm.
One of the requirements of successful norm change is first-movers willing to spark the change. What makes a person a first mover? First movers are people with very low thresholds who are willing to defy the established norm. They: have low sensitivity to the norm (i.e. does not care about the norm); are autonomous in their decision making, and; have a low perception of the risk.
There exists a network of social relations, and the structure of these networks may determine whether deviant behaviour will or will not produce a social change. With increasing abandonment, increasingly more people are likely to abandon the norm. This is because the perception of the risk involved decreases. This continues until a tipping point is reached. However, if there is a big gap between the trendsetter threshold and other peoples’ thresholds, failed diffusion will occur and the new behaviour dies out.
Coordination mechanisms for empirical expectations to induce behaviour change include the media, trusted government mandate and a public commitment.
For change to occur, having shared reasons that are independent of social expectations is a necessary precondition. This is accompanied by a change of personal normative beliefs. Reasons why a norm is accepted includes: (1) people not seeing the norm as a problem; (2) people may not know that there are alternatives, and; (3) due to pluralistic ignorance. Interventions to change beliefs must be suitable to the underlying causes.
However, deviating from the norm invites negative consequences, thus people will not act alone. This implies that there must be a collective change of expectation, and a coordinated change in behaviour.
Tools for change include media, legislative interventions, economic incentives, educational campaigns and group deliberations. All the tools make people aware that they are being exposed to the same message, but collective discussions has several advantages. During the discussion, the general acceptance of certain arguments becomes visible. It helps to change our personal normative and factual beliefs and to observe that other beliefs are changing too.
Pluralistic ignorance occurs when many people privately condemn a behaviour but wrongly believe that their peers endorse it. This can occur because the behaviour or its consequences are visible but people do not voice their disagreement, leading to a lack of transparent communication. In such cases, updating normative expectations can help to change behaviour.
A different but related situation is when people misperceive the frequency of a behaviour (e.g. prevalance of corruption or drinking). Correcting misperceptions can change descriptive norms but not social norms.
To measure norms, you must check empirical expectations, normative expectations and whether there is a conditional preference.
Behaviour can be measured through monitors or self-reports. Personal normative beliefs can be measured through surveys, and it is important to distinguish between prudential and non-prudential beliefs.
If empirical and normative expectations are mutually consistent, it is PROBABLE that there is a norm, but it could also signal a moral rule. Therefore it is necessary to check whether preferences are conditional. This is done by varying empirical and normative expectations and checking whether behaviour changes, through direct hypotheticals or vignettes.
You can also measure how stable a norm is by assessing which expectation has more weight. For example, when empirical and normative expectations diverge, empirical expectations usually dominate.
Thus, through systematic measure, you can diagnose whether a behaviour is a custom, moral rule or true social norm, and whether the social norm has causal relevance.
Preference is a disposition to act in a certain way. It is different to liking or endorsing something.
Conditional preferences means that a choice is dependent on empirical expectations (descriptive norms) or both empirical and normative expectations i.e. if different expectations lead to different behaviour, then people have conditional expectations. It is important to note that consistency between social expectations and behaviour does not mean that the behaviour is influenced by expectation.
Whether people have conditional preferences influences the intervention that is designed. If people have conditional preferences, the intervention should focus on changing social expectations.
A social norm is a rule of behaviour such that people prefer to conform to the norm conditional on empirical and normative expectations. Under social norms, there are usually informal social sanctions that motivate compliance. Hence the presence of sanctions indicate that there are normative expectations. However, some people might comply without sanctions if they believe that other people’s normative expectations are legitimate.
(If empirical expectations were sufficient to motivate compliance, this would be a descriptive norm. )
Motivations for behaviour differ from community to community. Bicchieri provides a useful visual summary for diagnosing whether a behaviour is a custom/rule, descriptive norm or social norm.
Besides empirical expectations, the second type of social expectation is expectations of what people think we should do. It is a second-order belief that are often paired with an expectation of a positive or negative sanction (e.g. shaming). Depending on the circumstances, different people will matter to our decisions. Or, they may not matter because we do not care. For instance, normative expectations do not matter for customs, descriptive norms and moral norms.
Personal normative beliefs
Personal normative beliefs are what I think I should, or people in general, should do. Normative beliefs overlap with normative expectations that the lecturer (Bicchieri) states that while normative beliefs matter to choice, normative expectations carry more weight.
Bicchieri also brings up the measurement of attitudes in surveys. Attitudes are very wide (i.e. personal normative beliefs is a narrower category) and just measuring attitudes might not provide a clear understanding of underlying beliefs and expectations. Thus she recommends making a clearer distinction in surveys.
When people engage in a collective behaviour, it is important to understand whether the action is interdependent. This is because interdependent actions are influenced by what a person’s reference network – the set of individuals who matter to me when I have to make a particular decision – does or think we should do.
In contrast to an interdependent action, a custom is a pattern of behaviour that individuals prefer to conform to because it meets their needs. For example, people open defecate because it requires the least amount of work – it is not conditional on how others defecate or what other people think of you when you open defecate.
Changing customs can be difficult because the alternative behaviour may require collective action that entails introducing interdependencies. Thus, understanding the interdependence of a collective behaviour helps us decide what type of intervention offers the best chance of success.
Empirical expectations refer to situations when the expectations of what other people do guide my actions. They can be:
Unilateral (when others don’t have expectations of me)
When we want to imitate someone
When what others do is informative (social proof)
Multilateral (when others also have expectations of me)
To coordinate with others
Empirical expectations are important to the lecturer’s (Bicchieri) definition of descriptive norms, which she describes as a pattern of behaviour that we prefer to engage in because we believe that others follow it. (This differs from the definition from social psychology.)