Editing Political spectrum (section)

==== Subsequent criticism of Eysenck's research ====
Eysenck's conception of tough-mindedness has been criticized for a number of reasons.
* Virtually no values were found to load only on the tough/tender dimension.
* The interpretation of tough-mindedness as a manifestation of "authoritarian" versus tender-minded "democratic" values was incompatible with the [[Frankfurt School]]'s [[single-axis model]], which conceptualized authoritarianism as being a fundamental manifestation of conservatism and many researchers took issue with the idea of "left-wing authoritarianism".<ref>{{cite journal |author=Stone, W.F. |title=The myth of left-wing authoritarianism |journal=Political Psychology |volume=2 |issue=3/4 |pages=3–19 |year=1980 |jstor=3790998 |doi=10.2307/3790998}}</ref>
* The theory which Eysenck developed to explain individual variation in the observed dimensions, relating tough-mindedness to [[extroversion]] and [[psychoticism]], returned ambiguous research results.<ref>{{cite journal |author1=Ray, J.J. |author2=Bozek, R.S. |title=Authoritarianism and Eysenck's P-scale |journal=Journal of Social Psychology |volume=113 |issue=2 |pages=231–4 |year=1981 |doi=10.1080/00224545.1981.9924374 }}</ref>
* Eysenck's finding that Nazis and communists were more tough-minded than members of mainstream political movements was criticised on technical grounds by [[Milton Rokeach]].<ref>{{cite journal |author1=Rokeach, Milton |author2=Hanley, Charles |title=Eysenck's Tender-Mindedness Dimension: A critique |journal=Psychological Bulletin |volume=53 |issue=2 |pages=169–176 |date=March 1956 |doi=10.1037/h0045968 |pmid=13297921}}</ref>
* Eysenck's method of analysis involves the finding of an abstract dimension (a factor) that explains the spread of a given set of data (in this case, scores on a political survey). This abstract dimension may or may not correspond to a real material phenomenon and obvious problems arise when it is applied to human psychology. The second factor in such an analysis (such as Eysenck's T-factor) is the second best explanation for the spread of the data, which is by definition drawn at right angles to the first factor. While the first factor, which describes the bulk of the variation in a set of data, is more likely to represent something objectively real, subsequent factors become more and more abstract. Thus one would expect to find a factor that roughly corresponds to "left" and "right", as this is the dominant framing for politics in our society, but the basis of Eysenck's "tough/tender-minded" thesis (the second, T-factor) may well represent nothing beyond an abstract mathematical construct. Such a construct would be expected to appear in factor analysis whether or not it corresponded to something real, thus rendering Eysenck's thesis [[Falsifiability|unfalsifiable]] through factor analysis.<ref>Wiggins, J.S. (1973) Personality and Prediction: Principles of Personality Assessment. Addison-Wesley</ref><ref>Lykken, D. T. (1971) Multiple factor analysis and personality research. ''Journal of Experimental Research in Personality'' 5: 161–170.</ref><ref>Ray JJ (1973) Factor analysis and attitude scales. ''The Australian and New Zealand Journal of Sociology'' 9(3):11–12.</ref>