Why do diners tip: rule-of-thumb or valuation of service?

When diners decide how much to tip, is the decision based on social convention or on conscientious appraisal of server productivity? Previous researchers in economics and social psychology are generally inconclusive on this question. A common finding in the literature is that tip size and service quality are unrelated, a result usually obtained from OLS regressions. OLS is only appropriate if service quality is exogenous. It is argued that service quality is very likely endogenous in any regression of tip size; good quality encourages good tips, but server expectations of good tips encourage good quality. This simultaneity is accounted for by jointly estimating percentage tips and customer rankings of service quality on a sample of 247 diners in a Central Minnesota restaurant. Included are explanatory variables consistent with both the social psychology and economic views of tipping. In contrast to previous studies, it is found that service quality significantly affects tip size and when servers expect higher tips, customers rank service quality higher. Also it is found that patronage frequency and coupon redemption have no effect on percentage tips, but server gender influences quality significantly. It is concluded that the results are generally supportive of an economic hypothesis of tipping.