This relates to the above annotation. I also agree that people's perceptions of others behaviors are unreliable; but this is an interesting point that in this case, perception may be more important than reality. Of course, that would need to be tested for us to know for sure. I think an interesting future study would be to use SNA and have both the egos and their alters actually track their substance use activities day by day. This would address the perception vs reality issue, as well as the underestimation of own behavior issue. There could still be some social desirability bias though.

This reminds me of analysis by Yang et al. They suggest that often social network that includes many other drinkers or even one drinker, an individual's risk of relapse increases. I wonder how this analysis may apply to egocentric network. They also mention how Alcoholics Anonymous (AA) help drinkers with their social networks, they help reshape their network with those who may be role models. They try to reduce pro-drinking ties and increase pro-abstinent social ties. I feel like applying SNA in such networks can really produce positive results on both individual and community level.

I think this week's readings by Yang et al were really helpful in understanding how addictive behavior can spread through the network. A study by Cohen and Lemay (2007) suggested that there is a link between having less diverse social networks and getting influenced into drinking and smoking. It will be really interesting to see the results of this study. Especially, how the network dynamics might change since this is a egocentric network.