I would be interested in whether this is static or longitudinal data.
If it's cross-sectional data, the data could be due to aging (all 30 year olds go through this), or it could be generational (THESE 30-year-olds go through this, as a function of the cultural milieu in which they grew up/find themselves). You could expect this pattern if, say, sexual norms became more liberal over time, peaked in the late 70s / early 80s, then something (*coughAIDScough*) caused them to start swinging back toward the conservative end of the spectrum.
no subject
If it's cross-sectional data, the data could be due to aging (all 30 year olds go through this), or it could be generational (THESE 30-year-olds go through this, as a function of the cultural milieu in which they grew up/find themselves). You could expect this pattern if, say, sexual norms became more liberal over time, peaked in the late 70s / early 80s, then something (*coughAIDScough*) caused them to start swinging back toward the conservative end of the spectrum.