If there are multiple dimensions of trust, then particularized and generalized trust ought to be distinct from each other empirically as well as analytically. If generalized trust is a value, it ought to be stable. In this chapter, I make no bold theoretical claims. Instead, I take on some methodological issues. First, I discuss the measurement of generalized and particularized trust. Then I consider methodological issues in establishing that trust is indeed a value. If trust does indeed reflect moral concerns, it should be stable over time. After all, people don't change their minds on their core values. And people should think about trust as something that is general, not simply reflecting their day-to-day experiences. I find support for both claims: Trust is stable, or at least as stable as most predispositions in political and social attitudes. And when people “think aloud” about trust, they discuss it in general terms. I also consider some reservations expressed about the standard trust question. There are methodological problems that won't go away, but they are not severe.

Rarely is there a survey that asks a wide range of trusting questions, such as the Pew Center survey discussed in Chapter 2.