Abstract:Several techniques have been proposed for displaying data certainty on maps, but few have been empirically tested for effectiveness. While it is important to make data certainty information easily accessible, the addition of such data should not unduly increase map complexity. Thus, it becomes important for cartographers to examine the available methods for displaying this aspect of metadata and to test each for its effectiveness. The focus of this study was the display of data certainty information on graduated circle maps. Four types of accuracy indicators were evaluated for their effectiveness in communicating data certainty information. Two were traditional accuracy indicators: reliability diagrams and legend statements. Two were bivariate in form, one using a value-size combination and the other mimicking the idea of focus by varying the line value of the graduated circles to suggest a fading of symbolization for least certain data. The study was designed to assess whether subjects could identify data certainty information on test maps, and evaluate how accurately and confidently they could extract and interpret both thematic and data certainty information. Mean accuracy and confidence rates were compared for maps using different accuracy indicators to evaluate their relative effectiveness. Results suggest that subjects had most difficulty identifying and extracting data certainty information using maps that employed legend statements. They were most successful when data certainty was wedded to thematic data on the map using the bivariate accuracy indicator that mimicked the concept of focus. Identification and extraction of thematic data values were not significantly affected by choice of accuracy indicator. Article: INTRODUCTIONMap accuracy is often equated with graphic quality. As noted by both Wright (1942) and McGranaghan (1993), well-drawn, precise maps are typically taken as scientifically authentic, regardless of the quality of their underlying data. Aesthetically pleasing maps, however, can conceal problems with the data and methods used in their creation. Wright (1942:527) provides perhaps the most interesting analogy on this subject: "A map may be like a person who talks clearly and convincingly on a subject of which his knowledge is imperfect." Always a problem cartographically, this particular issue has become even thornier as we have moved from manual, hand-drawn maps into the digital environment where nearly anyone who can master a software package can be a "mapmaker". Technology provides us with amazing capabilities in creating, editing, and displaying spatial data, capabilities that are offset by the fact that many of these maps are inappropriately used given the data upon which they are based. Since the validity of the underlying data is the key to making credible decisions, it makes sense that reporting and spatially depicting data certainty information should be addressed in a contemporary cartographic framework. Yet, as MacEachren (1994:67) points out: "The cartographic litera...