This monograph utilizes a 26 socioeconomic variable correlation matrix from 17 nations to develop propositions “that will indicate when and under what conditions the military becomes dominant” (p. 1). While employing this data to move from “reductionist explanations” (p. 24), the work is flawed fatally by its own “reductionist” (simplistic) methodology:
The author makes naive assumptions about his data. Nowhere are we told that the variable relationships are linear, linearity being an assumption of the correlations used. At one point he speaks of a curvilinear relationship between two variables, but one cannot tell if he has corroborated this relationship visually.
The author uses static correlations at a single point in time to advance causal or process propositions, such as “Increasing trade and industrialization will decrease the need for aggressive military forces . . .” (p. 10). This is an inexcusable methodological lapse. Moreover, he advances the above proposition without having an industrialization variable.
The author molds his findings to fit his assumptions. At one point a .34 correlation is a “high positive relationship” (p. 23), while at another point a .48 correlation shows “little relationship” (p. 16). This is either sloppy work or deception.
These lapses, plus poor writing (seeming to refer to Chile as underdeveloped and non-urban [p. 6]) render this monograph totally inadequate.