Characterising human behaviour is challenging, and datasets about people often suffer from issues of misrepresentation. To account for misrepresentation, researchers have turned to data synthesis. Here, we implement a straightforward data synthesis approach that does not rely upon knowledge of dataset uncertainty and use it to parametrise predictors used in an agent-based model (ABM) to estimate visits by people to greenspaces in Glasgow. The predicted visits follow expected patterns, with more visits on weekends, during daylight, and to popular tourist destinations. The approach is easy to implement and allows researchers to combine datasets of varying veracity to predict human behaviour.