Understanding factor analysis

Thousallds of variables have been proposed to explain or describe the complex variety and i~lterco~lnectiolls of social and illternatio~lal relations. Perhaps an equal number of hypotheses and theories linkillg these variables have been suggested. The few basic variables and propositions central to ullderstalldillg remain to be determined. The systematic dependencies and correlations among these variables have been charted only roughly, if at all, and many, if not most, call be ~lleasurecl only on presence-absence or rank order scales. And to take the data on any one variable at face value is to beg questions of validity, reliability, and comparability. Confronted with entangled behavior. unkno interdependencies, nlasses of qualitative and cluantitative variables, and bad data, many social scientists are turning toward factor analysis to uncover major social and internatio~lal pa t terns. Vactor hundred variables, conlpellsate for random error and invalidity, and disentangle complex interrelationships into their major and distinct regularities. Factor analysis is not without cost, ho\\-ever. It is lnathe~llatically complicated and entails diverse and numerous consideratio~~s in application. Its technical vocabulary includes strange terms such as e i g e ~ z ~ n l u e s, iv iate, dim, e~zsions, orthogonal, londingr, and co~rrn11~nality. Its results usually absorb a dozen or so pages in a given report, leaving little room for a methodological iiltroductio~l or explanation of terms. Add to this the fact that students do not ordinarily learn factor allalysis in their formal training, and the stun is the major cost of factor analysis: most laymen, social scientists, and policy-makers find the nature ant1 significance of the results incomprehensible. The problem of cornn~unicati~lg factor analysis is especially crucial for peace research. Scholars in this field are clraw~l An invitec1 paper for The Jolrrncll of Conflict frolll clisciplines alld professions, alld Resolt~tion. Prepared in connection with research supported by the National Sciencc Fotunf? ~ them are acquainted with thr dation. GS-1230. For many helpful comments method. made on a previous draft. I wish to thank As our empirical kllowledge of conflict Henry Kariel, hlichael Haas, Robert Hefner: processes, behavior, conditions, and pattelns Pitts, and J. David Singer. Portions of the article are taken from Rummel (1968). beconle illcreasi~lgly expressed in factoror a bibliography of applications of factor analvtic terms, those who need this knowlanalysis in the social sciences (esclucling psyedge lnost in order to make illforlned chology), see Rummel (1968). A bibliography of a ~ ~ l i c a t i o n s to conflict and inte~national decisions may be those who are most A A