| Studies over the past fifty years have reported that United States citizens scored poorly on geographic knowledge tests. It has been the perception of the author that college students from different parts of the United States vary in their geographic knowledge base. Thus the problem of this study was to determine how well college geography students scored on a researcher developed test instrument of geographic knowledge and to examine the resulting regional test score patterns. To do so the author developed a geography-based test instrument and distributed it nationwide. Responses from 1,004 students in 39 different states were analyzed. Spatial analysis of scores was performed based upon the eight geographic regions of the United States, as defined in the 1997 Rand McNally classroom atlas.; Scores ranged from 3 to 26 correct answers, with an overall mean of 17.88 and a standard deviation of 3.622. Of the 1,004 respondents, 98.8% scored within two standard deviations of the mean. Using the total number of possible correct answers, the mean percentage correct was 68.8%.; A one-way analysis of variance (ANOVA) compared the mean scores for the continuous/dependent variable, (number of correct answers) with the independent/categorical variable (state/region). When a statistically significant difference was found, a Scheffe post-hoc was performed to determine which regions accounted for the difference(s). Post-hoc analysis eliminated three regions, the Great Plains, Mid Atlantic, and Midwest, from further review due to the small number of responses. As a result of this study statistically significant differences in basic geographic knowledge were found to exist between the Northeast, Pacific, Rocky Mountain, South, and Southwest regions. |
PDF Full Text Request