Linking scientific and economic data potentially has many benefits for natural resource management and policy analysis. This paper compares a sophisticated method for bringing these data together in farm surveys using detailed farm boundary data collected during an interview, with a less onerous approach requiring only a single point from each surveyed farm. A comparison is made by modelling land values using regression analysis in central New South Wales. It was found that there were only minor additional benefits from using the more sophisticated approach in this case. Additionally, there are significant costs in collecting and processing the detailed georeferencing data that make it considerably less practical than the simpler approach.