![]() |
Visual Statistics Studio ♦ Outline of Visual Statistics ♦ Library |
David J. Krus and Auke
Tellegen
Summary.-Reich's
tripartite theory of social consciousness, as formulated in his book,
The Greening of America, was translated into Consciousness I, II,
and III Scales, based on content analysis of pertinent chapters in the
book, and quantified. The factor analysis of these scales
showed that Consciousness Scales load on two dimensions. Subsequent canonical
analysis of these scales, followed by rotation of canonical weights into
the simple structure, indicated that substantial amount of variance in the
Consciousness Scales can be accounted for by the normative-humanistic
constructs.
In the early seventies
the bestseller list was topped by
The
Greening of America
(Reich, 1971), a book eliciting a gamut of opinions. To Senator George
McGovern, the book was
'one of the
most gripping, penetrating and revealing analyses of American society I
have yet seen,'
whereas to Peter Marin of the New York Times Book Review, it was
'simplistic,
misleading, presumptuous.'
Reich's book attempted a description of the ideological
niveau
of the last three generations. The self-interest, aggressiveness, and
prejudice of the two older generations (Consciousness I and II) were
seen to be in the sharp contrast with the Consciousness III of the new
generation.
The present study
is an attempt to ascertain the degree of plausibility of Reich's
typology and its place within the family of other personality constructs. This involved a translation of Reich's original qualitative
and intuitively derived observations into quantitative scales and an
empirical exploration of the relationship of these scales to Tomkins'
(1964) Polarity Scale, Tellegen's (1972) scales of
Authoritarianism and
Absorption and subjects’ chronological age. Tomkins'
Polarity Scale was
designed to indicate ‘how humanistic
and how normative in orientation the individual is'
(Tomkins, 1964, p. 3).
Tellegen's (1972) third generation
Authoritarian Scale
was designed to relate this construct to the emerging 'big
two' personality
dimensions of introversion-extraversion and dominance-submission. Tellegen
& Atkinson’s (1974)
Absorption Scale
is characterized by items such as 'textures--such
as wool, sand, wood--sometimes remind me of colors or music,'
and
'I like to watch cloud shapes change in the sky,'
and is moderately related to hypnotic susceptibility.
METHOD
The descriptions of
Consciousness I, II, and III in
The Greening of America
(Reich, 1971, pp. 20-41, 62-90, 233-285) were converted into the
form typical of standard agree-disagree opinion inventories.
The
questionnaire was administered by students enrolled in an evening
statistics' class at the University of Minnesota. Subjects were their
friends or relatives (N = 58). Evening class students generally have
more varied backgrounds than full- time day class students, assuring better sampling than the traditional student
pools, particularly with respect to the age of the subjects (13 to 57 yr.,
M = 29 yr., SD = 123).
Scales included in the
analysis were intercorrelated as shown
in Table 1.
Table 1. Intercorrelations
between scales Included in the
analysis and the age of
the subjects
|
|
2 |
3 |
4 |
5 |
6 |
7 |
|
1. Consciousness
I+II |
.61 |
.59 |
-.15 |
-.21 |
-.43 |
-.02 |
|
2. Normative |
|
.32 |
-.09 |
.04 |
-.47 |
.08 |
|
3. Authoritarian |
|
|
-.01 |
.00 |
-.16 |
-.00 |
|
4. Consciousness
III |
|
|
|
.41 |
.56 |
-.40 |
|
5. Humanistic |
|
|
|
|
.37 |
-.08 |
|
6. Absorption |
|
|
|
|
|
-.31 |
|
7. Age |
|
|
|
|
|
|
This correlation matrix
was factor analyzed by the principal factors method with squared
multiple correlations used as the communality estimates. The first two
factors with eigenvalues equal to 2.19 and 1.80 accounted for 66.6% of the total
variance. The third factor with eigenvalue equal to .75 did not meet the
Kaiser's retention criterion and was deleted. Figure 1 shows the
resulting factorial structure plotted from the
Varimax rotated matrix
of factor loadings.

Figure 1. Factorial
structure of the Consciousness Scales
The Consciousness I and II Scales together with the Normative and Authoritarian Scales define the first factor while the Consciousness III Scale together with the Humanistic and Absorption Scales define the second factor.
Canonical Analysis
The scales were also analyzed by the canonical analysis. As the Consciousness I and II Scales reflect a similar construct, they were combined into a single scale and, together with the Consciousness III Scale, defined the criterion set of variables. The other scales involved in the analysis comprised the predictor set of variables. The extracted eigenvalues and their corresponding canonical correlations are shown in the Table 2.
Table 2. Eigenvalues and Canonical Correlations for the
Canonical Analysis of the Consciousness Scales.
|
Eigenvalues |
Canonical Correlations |
|||
| Canonical Components |
1 |
2 |
1 |
2 |
|
1 |
.626 |
.000 |
.791 |
.000 |
|
2 |
.000 |
.399 |
.000 |
.632 |
The results of the unrotated canonical analysis indicated that within the first canonical component, the predictor set of variables accounted for 63% of variance of the first component of the Consciousness Scales. Within the second canonical component, the predictor set of variables accounted for 40% of variance of the second component of the Consciousness Scales. As these components are orthogonal, the explained variances are not additive and to speak of the total amount of variance explained by the predictor set of variables is nonsensical. Next, the canonical weights were graphically rotated, as shown in Figure 2.

Figure 2. Graphical Rotation of Canonical Weights.
The extracted eigenvalues and their corresponding canonical correlations for the rotated solution are shown in the Table 3.
Table 3. Eigenvalues and Canonical Correlations for the
Rotated Canonical Analysis of the Consciousness Scales.
|
Eigenvalues |
Canonical Correlations |
|||
| Canonical Components |
1 |
2 |
1 |
2 |
|
1 |
.576 |
.004 |
.759 |
-.064 |
|
2 |
.004 |
.442 |
-.064 |
.665 |
The results of the rotated canonical analysis indicated that within the first canonical component, the predictor set of variables accounted for 58% of variance of the first component of the Consciousness Scales. Within the second canonical component, the predictor set of variables accounted for 44% of variance of the second component of the Consciousness Scales. The obtained and the rotated canonical weights are reported in Table 4.
Table 4. The Obtained and
the Rotated Canonical Weights
|
|
Consciousness I+II |
Consciousness III |
||
|
Canonical Weights |
Obtained |
Rotated |
Obtained |
Rotated |
| Consciousness III | -.276 | .193 | .973 | .993 |
| Consciousness I + II | .921 | .999 | .418 | -.043 |
|
Normative |
.4 |
.6 |
.497 |
.241 |
|
Authoritarian |
.50 |
.588 |
.31 |
.054 |
|
Absorption |
-.30 |
-. |
.65 |
. |
|
Humanistic |
-.30 |
-.183 |
. |
.3 |
|
Age |
-.036 |
-.231 |
-.44 |
-.378 |
Canonical weights
were graphically rotated about 260 in the counterclockwise direction.
The rotation of canonical weights suggested that, as in factor analysis,
rotation is a
conditio sine qua non for meaningful interpretation of results of
the canonical analysis.
The structure of the rotated canonical weights, isomorphic with the obtained factor analytic structure, are in congruence with intuitive understanding of these constructs.
The Reich's book is an incisive analysis of the social niveau of the sixties and the early seventies, as reflected by the Consciousness Scales. Results of the factor analysis and results of the canonical analysis suggest that even though the dominant issues of the post-war generation appeared unique, they reflected the enduring humanistic-normative facets of human affairs.
REFERENCES
Abrams, M. H. (1958)
The mirror and the lamp.
New York: Norton.
Reich, C. A. (1971)
The greening of America.
New York: Bantam.
Tellegen, A., & Atkinson,
G. (1973) Openness to absorbing and self-altering experiences
(‘absorption’), a trait related to hypnotic susceptibility.
Journal of Abnormal Psychology,
33, 268-277.
Tomkins, S. S. (1964)
Polarity scale.
New York: Springer.
Accepted
October 30, 1974
|
|
Visual Statistics Studio ♦ Outline of Visual Statistics ♦ Library |