The relative effectiveness of noncontiguous cartograms, in presenting data were tested over a continuum of tasks ranging from general to specific, using university students as subjects. A total of three tasks were examined, one task required subjects to judge cartograms showing a very general distribution, the second task required a moderate amount of information to be retrieved, and the third task required very specific information to be obtained from the cartogram. Two sets of non contiguous cartograms of the United States were used; one set had international and internal boundaries the second set had only international boundaries. In general students performed task one and two with a great deal of accuracy, proving non contiguous cartograms are a useful method of displaying geographic information. The third task was less successful than the first two, although this fact is not surprising, we learned that when displaying and trying to retrieve very specific kinds of information, the non contiguous cartogram is not a viable option. / Master of Science

Although highway maps are commonly used, little research has focused on the effectiveness with which they are utilized. Route following and route planning are the two main uses of highway maps, and road symbolization is the information most desired by highway map users. Therefore, testing was conducted to determine relative effectiveness of different road symbolization systems in route following and route planning tasks. One black and white and two color road surface symbol hierarchies were employed on maps tested with 162 participants. It was hypothesized that both color hierarchies, one consisting of a part-spectral progression of hues (red-orange-yellow) and the other a full-spectral order (red-green-blue), would prove to be more effective than the black and white hierarchy, and that the part-spectral hierarchy would prove similarly superior to the full-spectral hierarchy. Overall results did not conclusively support either hypothesis. There was, however, significant variation in performance according to the purpose of use and among subgroups of the test population. These findings indicated that designing an optimal map for all purposes and map users may be an impossible task. The results also demonstrate that it is critical for the highway mapmaker to identify who will use the map he creates and the purpose(s) for which it will be consulted. / M. S.