Thirteen tests designed to measure aspects of configurational knowledge were compared and contrasted using a repeated measures, multi-data collection and multi-data analysis approach. Respondents consisted of 279 first-year undergraduates newly resident to the study area (Swansea, Wales). Respondents completed four tests, drawn from four different groupings which were consistent in nature. Tests were varied in spatial cueing (the amount of information supplied) and location cueing (the amount of information requested). In addition, the data produced were subjected to systematic filtering (a set number of places were removed from the analysis) and data-defined filtering (guesses and unfamiliar places were removed from the analysis). It is demonstrated that tests produce differing results and that task demands explain the variances. It is recommended that multiple, mutually supportive tests should be used to assess individuals' configurational knowledge.