The chart can be misleading, because it's missing a crucial data point: the % of the heterosexual population who are in a couple. My understanding is that this

The chart can be misleading, because it's missing a crucial data point: the % of the heterosexual population who are in a couple. My understanding is that this % has also declined over the years, which would make the decline in "met through friends", "met in school", and "met through family" look even worse. Also, given that we don't know the % of the coupled population, the chart can't tell us whether the % of the population who met their significant other in a bar or restaurant has increased, stayed the same, or even decreased. It might be all well and good if other forms of meeting were simply replaced by online dating with no loss, but I think there has been a significant decline in couplehood.