アメリカ主要地理学教室における計量革命のその後

抄録

This paper is concerned with the impact of the Quantitative Revolution on leading graduate departments of geography in the United States through two indices: proximity of graduate departments of geography in terms of PhDs granted; and fundamental dimensions of departmental specialties. Data sources are 1) Directory of College Geography of the United States, Academic Year 1962-1963; 2) Guide to (Graduate) Departments of Geography in the United States and Canada 1976-1977, 1984-1985, 1985-1986; and 3) AAG Directory 1982. In addition to those analyses, the recent development of the Scientific Geography movement was briefly described on the basis of Scientific Geography newsletters.
First, in order to examine the supply of quantitative geographers, Smallest Space Analysis-II was applied to PhD transaction matrices. The resulting two-dimensional column solution is termed “proximity space”, revealing an outflow pattern of graduates. Less than thirty departments were selected for study, most of which appeared in Morrill's (1980) department ranking (Fig. 2). The faculty members included were associate professors and full professors.
Proximity space for 1962-1963 is shown in Fig. 3. Major departments are grouped into two clusters: 1) Washington, Northwestern and Chicago, and 2) Wisconsin and UC Berkeley. According to Raup (1961), the former group in partcular consists of economic geography-oriented departments. The proximity space clustering shows a pattern in which, while major departments specializing in a single research field or emphasizing synthetic research occupy the central part, others employing graduates from these departments surround them.
Proximity space for 1976-1977 (Fig. 4) shows a pattern in which departments producing many geographers are located in the center, and others employing graduates from these departments occupy the periphery. In contrast with 1962-1963, when major departments were divided into two groups, they form one group as a whole across the 1976-1977 proximity space. This may be due to the fact that personnel interchange was accelerated among major departments. Considering the space in detail, such departments famous for their quatitative research as Washington, Chicago, Northwestern, Iowa and Ohio State form a sub-group in the lower rightward portion. This suggests that the Quantitative Revolution is no longer a revolution, and that departments exclusively producing quantitative geographers have come to form a cluster.
Proximity space for 1984-1985 is shown in Fig. 5. Its fundamental character is the same as in 1976-1977: departments producing large numbers of geographers are located in the central part. The “quantitative”-oriented group occupies the Left portion; this group is in turn sub-divided into the “long-established” departments of Washington, Northwestern and Iowa, and the “emerging” departments of Ohio State, Toronto and McMaster. This is related to the fact that the content of quantitative geography is beginning to be differentiated to some extent, though departments exclusively producing quantitative geographkrs still form a cluster.
Second, in order to ascertain whether quantitative geography has already taken root in American geography, the fundamental dimensions of departmental specialities were elicited. A binary data matrix indicating the presence of 37 specialties in 29 departments were standardized so that the sum of squares of each row vector is equal to 1.0. The resulting data matrix as an input data was subjected to factor analysis. Two major factors were extracted, accounting for 65% of the total variance (Table 3). In the case of Factor I, loadings are highly correlated with the number of specialties of the depertment (r=0.917^**). Scores are completely correlated with the number of departments offering the specialty concerned (r=0.999^**).