Lib/Con writeup
Yüklə 437,94 Kb.
|
|
Green philosophy endorsement tended to correlate with other variables in a manner similar to that of liberal endorsement. The independent endorsement tended to not correlate substantially with the other variables. In the discussion that follows, unless otherwise specified, all correlations are Pearson Product Moment correlations. Almost all of the scales correlate significantly at the .01 level or better and in the direction predicted with conservative and liberal political orientations. Specifically, of the 68 correlations, 64 (94%) are significant at the .05 level or better, most at the .01 level. This is interpreted to mean that the traits as measured represent psychological facets of the conservative and liberal worldviews, as hypothesized. To assess the collective power of the trait measures in divisions 1 though 10 to predict political orientation, the three traits in each cluster (a-c and d-f) were summed. Then a total score of the six in each division was created by reverse-scoring the three conservative scales and adding this to the three liberal scales. A linear regression between all ten of these dimension scores yielded a multiple correlation of .82 with a measure of the liberalism-conservatism orientation created by summing the liberalism endorsement self-rating and the conservative self-rating, reverse-scored (F = 7.815, sig. .000). When just the three strongest predictors were used, dimensions 1 (religion), 3 (government type preference) and 5 (foreign policy), the R value was .79 (F = 25.222, sig. .000). This is considered to be further support for the hypothesis that the traits as measured are meaningful psychological facets of the conservative and liberal political worldviews. In the third column are the correlations with gender, age, years of general education and high school grades, all as reported by subjects in responding to demographic questions at the beginning of each questionnaire. A negative value means males tend to be lower. Males tended to be slightly more likely than females to endorse 1a Religious Fundamentalism, 2a Social Disenfranchisement, 2c In-group Elitism, Male superiority traits (3a, b and c), 4c Militaristic philosophy, 6b Conservative economics, 8a Social Dominance Orientation, 8b In-group favoritism, 8c Power Oligarchy, 9c Leader Devotion, and Resource Consumption (10 a, b and c). They also tend to be higher than women on the cluster of miscellaneous conservative attitudes (11a). Females tended to be slightly higher than males on 1d Kindly religious beliefs, 3e Female politics, 4e Peace politics, 4f Peace promotion, 5f All citizen government, 6f Share economics, 8f Social egalitarianism, 9d Progressive government, 10e Green/clean/safe politics, and 11d Miscellaneous liberal values. While the author had made no initial predictions about how the measured traits would correlate with gender, these correlations seem consistent with the content of the traits. They are also consistent with the hypothesis that political worldviews may be shaped by evolution, as males are stronger and more aggressive than women and therefore more likely to serve in war activities and leadership based on physical strength and success in war. War has been necessary to protect groups from invaders, if not to invade. None of the traits correlate significantly with age, suggesting that these traits as measured may be relatively independent of culture and its influences on citizens over their lifetimes. Several of the traits on the liberal side of the matrix correlate positively and slightly to moderately with general education, including 7e Violence prevention (.29*), 3f Female Honoring (.29*) and 6e Liberal Economic Politics (.40**), as well as 8e. Anti-oligarchy (.28*), 9d. Progressive government (.29*), 9e. Liberal political agenda (.31*), 9f. Citizen authority government (.41**), 10c Sustainable communities (.33*), 10d Ideal Sustainability (.34*), 11d.Miscellaneous liberal attitudes (.30*), and 12d. Liberal tribal attitudes (.35*). Thus, persons in this sample with more education tended to hold liberal political attitudes as measured by these scales. In contrast, education consistently correlates negatively with traits in the conservative cluster: 7b Violence enabling (-.35*), 7c In-group self-defense (-.31*), 9a. Authoritarianism (-.29*), 10c Human resource use (-.28*), and 10b Government resource use (-.33*). Persons in this sample with less education tend to hold conservative attitudes as measured by these scales. Correlation doesn't document causal directions, so we can't conclude from this that education necessarily causes or shapes liberal political attitudes. It is possible that persons who have liberal attitudes are also more curious and open-minded and thus tend to seek educational experiences more than less open-minded persons do. The data cited by Jost, et al above documents that liberals do indeed tend to be higher on Openness than conservatives. Only a few traits correlated significantly with high school grade point average, with unclear implications, other than suggesting that political attitudes may be unrelated to intelligence. Table 3. Validity correlations.
The relationships between the traits in each of the divisions of political attitudes are presented in columns 4 and 5. The correlations all are significant and in the directions predicted. For each division, 1 through 10, each of the three traits in a conservative cluster correlate positively with each other, as do the traits in the corresponding liberal cluster (column 4). And the conservative traits correlate negatively with the liberal traits, as predicted (column 5). As stated earlier, an attempt was made for most divisions to create scales that measure content of basic, political and hypothetical content. However, for some divisions the correlations between traits in a cluster are so high (e.g. in the .80's) that it appears little meaningful differentiation may have been achieved. See for reference the three facets of religious fundamentalism (1a, 1b and 1c), and for violence-proneness and warmongering (4a, 4b and 4c). Another possible interpretation of these clusters of very high correlations is that for several divisions of human behavior there is indeed a very intimate relationship between basic beliefs or attitudes and corresponding political manifestations of them, and with related semi-conscious or private beliefs. In any case, the present data seems to document that there are many divisions of politically-relevant human beliefs and that these beliefs consistently fall into two “camps”, liberal and conservative, and that these two camps see the political world from consistently contrary perspectives. Regarding division 11, many items of rather disparate, miscellaneous content were included in the miscellaneous conservative and liberal clusters respectively. These items were included because they had interesting content, some of which has been suggested by prior research, as had music preferences, summarized above, and some of which didn't seem to fit in any of the prior 10 dimensions. Many had content that the author simply had intuitive hunches about for various conscious and subconscious reasons. Yüklə 437,94 Kb. Dostları ilə paylaş:
|