[MUSIC] Let us look again at the diagnostic diagram. We have measured the Empirical Expectation and Normative Expectation. And by measuring conditional preferences, we have come to the conclusion that social expectation influence behavior. We have a Social Norm. Further important question to ask is whether people normative expectations are correct. That is, do normative expectation correctly reflect people personal normative beliefs? If they do not perceived and objective consensus will diverge. How can this happen? Remember the lecture on pluralistic ignorance. Whenever behavior, or its consequences, are visible, but transparent communication is difficult or impossible, we may experience pluralistic ignorance. People may wrongly assume that other's behavior is consistent with their beliefs, and that everyone else endorses the observed behavior. Therefore if it is prudent to conceal what one really thinks, one will conform to the prevalent behavior. Let me give you an example. Say a father observes frequent child beating and assumes parents hit their children, because they believe it is the right way to discipline them. In fact, many of the parents think that corporal punishment is inappropriate and does not benefit the child. But, they do not dare to openly express their opinion for fear of being judged poor parents. So, when a child misbehaved, the probably will probably hit him especially, if someone watches. The practice of child beating continues unchallenged. Measurement of personal normative beliefs and normative expectation helps determine if in fact there is pluralistic ignorance. So if we find out that perceived and objective consensus diverge, we can conclude that the social norm that is supported by those wrong normative expectations is not collectively endorsed. Transgression will be concealed and the norm that most people dislike will survive. This norm however will be fragile because coming to know what people really think may use may potential transgressors to openly challenge it. Now, let us listen to an example of pluralistic ignorance. I think that there's probably ideas of masculinity that fit into this issue, right? So even if a man doesn't personally believe it's right to beat your wife, it's this something that he would tell his friend about or would he speak out when his friend's talking about beating his own wife. I don't know. I think that's something that we're looking at in terms of prevention programming. How can we work with men who maybe hold different ideas about relations as women to speak more publicly or to be influences of other men? Let me conclude this lecture on measurement by stressing the elements that make a norm sustainable. First, let us remember that a norm exists, and is followed in a group if most of its member have high levels of social expectation, which means they expect most members of their group to follow the shared role of behavior and to approved and support it. Moreover the social expectation influenced their behavior. This does not mean that is sustainable since empirical or normative expectations maybe wrong, If people misperceive how frequent the behaviour is or how much support it has sooner or later they will come to realize their mistake. A norm therefore will be sustainable only when expectations are consistent with behavior and beliefs. Fortunately, we can measure all these components and decide whether a norm exists, and how stable or sustainable it is.