In policy contexts valuations may be required not only for the most inclusive good considered but also the relevant sub-areas or components which are nested within the broader good. Valuation of differing levels of quantity/quality, referred to here as scope, provides many challenges to the researcher. Through considering the sensitivity of welfare measures to scope and the valuation sequence used, this thesis was undertaken to investigate the difficulties in and prospects for using CV to estimate the non-market value of nested goods. Using a mixed methodology survey design, the case study focuses on the use of the contingent valuation method to estimate non-market benefits from a saline flood alleviation scheme in Broadland, located in Eastern England. A scheme to protect the `whole' area from saline flooding was compared to five separate `part' schemes which would only protect nested sub-areas of Broadland. Insensitivity to scope was observed within both a split sample comparison of `whole' and `part' valuations and a same sample comparison where the `part' scheme was valued first. The prior valuation of the `whole' scheme did induce sensitivity to scope between `whole' and `part' schemes, but this effect can be interpreted in terms of either economic theory, consistency pressures or the use of a more natural valuation sequence. However, the prior valuation of the `whole' scheme also induced split sample sensitivity to good characteristics between the `part' schemes. Overall the explanation for the results given cannot be monopolised by either contingent valuation supporters or their critics. In the case of Broadland flood alleviation, the validity of the valuations presented remains unclear but some interesting findings suggest future directions for research.