A study by Anke Zimmermann and Richard Easterlin follows people from up to four years prior to their first marriage through several years after getting married. The basic finding is illustrated in the figure below, which shows the life satisfaction of people who got married during the study period relative to those who never got married during the study period. As we go from left to right, we see how the life satisfaction of a person changes over time as they first cohabitate with a partner, then get married, and continue that marriage for more than a year.

Compare the life satisfaction of people who have been married for a while to that of people who are not married but are living with their partner. Do you find this evidence supportive of or contrary to Gilbert's advice? Identify a confounder that this comparison suggests may have existed in the original correlation.