This idea just cropped up on reading Mano’s latest blog and some of the comments. One of the questions that I’m interested in is does behavior change because of certain events. For instance, let us consider an analogy from one of my favourite subjects, political science. Can we ask the question did US foreign policy change after 9/11? If the answer is yes then how do we measure it? For measurement we must resort to causal empiricism (anecdotal observations vs. scientific evidence). For this example there are hardly any practical difficulties. We can collect data on foreign policy and statistically analyze it to provide scientific evidence (which is the harder part).
Now to the issue of interest. Does the behavior (defined as broadly as possible) of couples change when they have children? Again if we are to resort to causal empricism as our basis, what would the answer be? While anecdotal evidence suggests that this should be the case, is there scientific evidence for the same? How do we measure if behavior actually changed or not? Difficult questions but could prove to be useful. For instance consider a nested question within the broad theme. Has risk taking behavior of couples changed after they have children? If the scientific evidence is yes, then insurance companies can use targeted advertising on this group to their benefit (just thinking aloud and it is quite possible that they are already doing this). But it is quite possible that risk taking behavior did not change. It could just be that while this might be common perception, in reality, there might be no change. What if there was a change but it was in the opposite direction? What if people wanted to take greater risks (theoretically possible).
To me what is interesting is how do we generate scientific evidence to answer this question. Can we design an experiment to observe the changes in behavior if any (that should be scientific evidence for sure).