In this paper, we compare single and joint evaluation (JE) of competing public sector programmes in a contingent valuation exercise. Using survey data aimed at evaluating WTP for cancer interventions (n = 2628), we disantangle two types of effects of JE: informational effects and sequence effects. By the former, we
mean: by presenting different programmes to respondents, they will acquire more information on each programme than they would if each programme was valued in isolation. Sequence effects are underisable and induced by the JE exercise itself: changing the order of the valuation sequence induces different WTP values.
Our results show that there are informational effects but no sequence effects. We therefore argue that JE approaches can be added to the armoury of techniques aimed at designing better survey instruments in a way that induces informational effects without incurring problems of sequencing.