A Belief and Belonging Paradox? Americans with less formal education tend to express greater certainty about their belief in God. Fifty-nine percent of Americans without a college education say they are certain God exists. Only 44 percent of those with post-graduate education say the same, although they are slightly more likely to express certainty that God does not exist. Despite the fact that college graduates are more uncertain in their religious beliefs, they are actually more likely to belong to a church or congregation, a divide that has grown larger in recent years.

The Diverging Trajectories of Same-sex Marriage & Abortion. In 1996, Americans were far more supportive of legal abortion than allowing gay and lesbian couples to marry. Only 31 percent of Americans favored same-sex marriage in the mid-1990s while nearly six in 10 (57 percent) said abortion should be legal in at least some circumstances. Today, nearly seven in 10 (69 percent) Americans believe gay and lesbian people should be allowed to marry. Views of abortion have remained relatively unchanged—60 percent say abortion should be legal. The massive 38-point increase in public support for same-sex marriage was due to a variety of unique factors, including the widespread belief that it was inevitable.

The Importance of Work Friends. Americans who never attended college are experiencing a rapid social decline—they have fewer close friends and considerably less social support. Part of the reason may have to do with the fact that Americans without a college education have fewer workplace friends. Sixty-two percent of college graduates have a close friend they met through work, compared to 47 percent of those without a college education. But why do Americans without a college degree have fewer friends? Lower labor force participation and greater economic instability may be to blame. For the growing number of Americans who are unmarried and unattached to religion, the workplace has become an incredibly important source of social connection and belonging.