dimensions of adolescent alienation - university of minnesota · dimensions of adolescent...

219

Dimensions of Adolescent AlienationJames Mackey and Andrew AhlgrenUniversity of Minnesota

A review of the psychological, sociological andeducational literature indicated that the various

conceptualizations of "alienation" could be fittedinto five tentative categories appearing to have con-siderable overlap. An item pool developed to repre-sent these categories of alienation was screened byexpert review and pilot testing in the 9th grade andthen administered to 500 "normal" adolescents in

9th-grade classes in four diverse communities inMinnesota: a rural area, a suburban area, and

working class and inner city areas of a large city.Factor analysis identified three coherent dimensionsin student responses, which were labeled "Personal

Incapacity," "Cultural Estrangement," and"Guidelessness." Simple cluster scores constructedto represent these dimensions had internal-consis-tency reliabilities of .80, .70, and .67 respectively.Patterns of significant differences shown by anal-yses of variance among groups defined by commun-ity type, socio-economic status, ability, and sex,compared well with hypothesized patterns; the fewexceptions were tenable. The scales provide con-crete measures of alienation that may enable more

meaningful investigation of its incidence, correlates,and causes.

&dquo;Alienation&dquo; is one of the most frequently en-countered concepts in social science. Indeed, theamorphous, global concept of alienation hasbeen used as a catchword to explain nearly every

kind of aberrant behavior from drug abuse topolitical demonstrations. Although many schol-ars have discussed constructs that can be looselygrouped as alienation, few have attempted tomeasure it. Clark (1959), Gamson (1961), andHorton and Thompson (1962) defined alienationas a feeling of powerlessness. Nettler (1957) andKeniston (1965) conceptualized alienation as afeeling of estrangement from American societyand culture. In an attempt at synthesis, Olsen(1969) subsumed all the descriptions of aliena-tion under three dimensions: social isolation,normlessness, and powerlessness.The work of these investigators, proceeding

for the most part independently, has focused onadults and given little explicit consideration toadolescents. Three major methodological prob-lems limit the utility of the extant scales for usewith adolescents in contemporary American so-ciety. First, the available scales often were vali-dated with limited and atypical populations. Forexample, Clark’s (1959) scale was developed forand used only on members of an agricultural co-operative. Dean’s (1961) widely used scale wasvalidated with an adult sample in Ohio.A second weakness with available alienation

scales is their lack of control for a response set to

agree (or disagree) with almost all statements.

Downloaded from the Digital Conservancy at the University of Minnesota, http://purl.umn.edu/93227. May be reproduced with no cost by students and faculty for academic use. Non-academic reproduction

requires payment of royalties through the Copyright Clearance Center, http://www.copyright.com/

220

Generally all the items are stated in the samenegative direction with no reversed items. This isa weakness that is especially important for ado-lescents who take many tests of all kinds andwho may be unconscientious in marking a testthat &dquo;doesn’t count.&dquo; Frequent reversal of po-larity of items may keep attention high and dis-courage stereotyped item marking as well asneutralize a sincere response set. Even if it doesnot prevent careless marking, a mixture of re-versed polarity items will cause long runs ofidentically marked items to put a student’s scorein the middle of the distribution, where it willhave minimal effects on the analysis.

Finally, as demonstrated in a recent survey ofalienation measures (Mackey, 1973), nearly allthe available scales were developed in anotherera. In a period of emerging counter-culturesthat use alienation as a fundamental precept,this is unacceptable; new eras require new in-struments. In one of the few instances in whichalienation scales have been used with adoles-cents, Clarke and Levine (1971) used a four-itemmeasure to assess alienation in high schoolseniors. Their short scale was based upon a con-

ceptualization that defined alienation as politi-cal estrangement. Besides the limitation of thisnarrow range of meaning, seniors, at the outerrange of adolescence, cannot be considered rep-resentative of adolescents in general.

In a study of institutionalized adolescent de-linquents (Allen & Sandhu, 1967), the investiga-tors report the development of a fourteen-itemscale to measure alienation. The conceptualiza-tion of alienation in this study was narrow, thescale was brief, and the population-adolescentsin prison-was atypical.The major purpose of the inquiry reported

here was to structure empirically the severalloosely defined domains commonly conceived asalienation, first through correlational study ofthe interrelationships of components, and thenthrough dimensions of discrimination among so-cial groups.

Method

The study consisted of five parts. First, a set of

distinct descriptions of &dquo;alienation&dquo; was dis-tilled from the literature. Second, a series ofpropositions regarding sub-groups of alienatedadolescents was deduced from the intensionaldefinition. Third, a pool of Likert-type items tomeasure the dimensions was developed. Fourth,the item pool was administered to a large and di-verse sample of early adolescents and the dimen-sions of response analyzed to provide an empiri-cal redefinition of adolescent alienation. Fifth,the construct validity of the dimension measureswas investigated by comparing the pattern ofgroup differences to the pattern expected fromtheory.

Conceptualization of Alienation

The first task of the study was to posit a pre-liminary description of adolescent alienation. Asurvey of the educational, psychological, and so-ciological literature suggested that nearly all theattitudes scholars talked about when they de-scribed alienation could be subsumed under five

categories. The well-known definitions de-

veloped by Seeman (1959) initially served as or-ganizers, but we modified them considerably forapplication to adolescents. The five tentativeand obviously somewhat overlapping categoriesin the adolescent alienation construct were:

1. Powerlessness. Powerlesness is the feel-

ing of being unable to influence the forces thataffect the adolescent’s chances for success in

life. It refers to a lack of influence over social in-

stitutions and forces. At its root, powerlessnessis a feeling of drift or helplessness in an unre-sponsive universe where decisions are made bycaprice and where luck is a primary determiningagent. This kind of alienation relates to whatColeman called sense of control of environment

(Coleman, Campbell, Hobson, McPartland,Mood, Weinfeld & York, 1966).

2. Role estrangement. Role estrangementis the adolescent’s sense of being related to solelythrough social roles (e.g., as a clerk or a stu-dent). In this dissociation from his role, he per-ceives that his &dquo;authentic self’ is not valued but

only his contrived performances. Role estrange-ment includes the belief that rewards of all kinds



221

come from extrinsic activities, that intrinsic re-wards are not important. Etzioni’s The ActiveSociety (1968) provides the clearest contem-

porary discussion of this kind of alienation.3. Meaninglessness. Meaninglessness is the

perception by the adolescent that he cannot pre-dict what will happen because he lacks adequateknowledge, and, furthermore, that it is impos-sible for him to acquire that knowledge becausethe world is so complex. To resolve meaningless-ness, adolescents often adopt simplistic solu-tions to life problems (Seeman, 1959).

4. Guidelessness. Guidelessness is a dispro-portion between the adolescent’s desires and hismeans for achieving these desires. It is the feel-ing that social rules are either ineffective or un-realistic. The guideless adolescent feels a resent-ment toward the society that fails to provide himwith sufficient opportunity to satisfy his desires.The feelings inherent in guidelessness are repre-sented by a long tradition in social theory ex-tending from Emil Durkheim to Robert Merton(Olsen, 1969).

5. Cultural estrangement. Cultural es-

trangement is an attitudinal state where the ado-lescent voluntarily separates himself from thepredominant value system. A culturally es-

tranged adolescent may have full knowledge ofthese values and the means to achieve them, buthe explicitly rejects them. These adolescentsoften actively withdraw from any involvement inthe pursuit of success, money, possessions, andother &dquo;middle class&dquo; goals. In a series of bril-lant studies, Keniston (1965) provided the foun-dation for this domain.

Hypothesized Group Differences

According to the theoretical contexts in whichthese concepts were formulated, and consistentwith previous research, different groups of

people would have markedly different likeli-hoods of having these forms of alienation. Oneset of expectations can be stated fairly simply:Because he would be more likely actually to haveadequate knowledge, skills, role models and en-couragement, the more advantaged adolescent

in terms of ability, socio-economic status andcommunity type would be expected to feel less ofa sense of powerlessness, role estrangement,meaninglessness and guidelessness than the lessadvantaged student; but as studies conductedduring the past decade (Westby & Braungart, _ _1966; Flacks, 1970) reveal, because their pri-vileged status allows them the luxury of &dquo;de-viance,&dquo; students of higher socioeconomicstatus are more inclined to develop a criticalstance toward unconventional beliefs than theirless advantaged peers. Thus, the same condi-tions both encourage alienation and discouragecultural estrangement.

It is important to emphasize that our expecta-tions were not that &dquo;disadvantaged&dquo; groupswould be globally more alienated, but that dif-ferent groups would show distinctive patterns ofalienation. The specific hypothesized deviationfrom a global alienation was high cultural es-trangement for advantaged youth, who wouldshow low alienation in other domains. Other dis-tinctive patterns were to be determined empiric-ally.

In order to compare measured alienation

patterns to these suppositions, several measuresof advantage were chosen as independent vari-ables. Perceived academic ability was chosen be-cause the relation between ability and alienationwas one of the salient findings of Equality ofEducational Opportunity (Coleman, et al.,1966). Other research has also documented therelationship between ability and alienation (See-man, 1966; Seeman & Evans, 1962). Measuresof socioeconomic status were used because as

Templeton (1966) reported:In the case of every piece of research bear-ing on this relationship, social status wasfound to be negatively related to alienation(p. 253).

Community was included because severalstudies document a relationship between com-munity type and adolescent social attitudes. Ad-olescents in more advantaged communities areencouraged to be active and are socialized to be-lieve that they have a right to be involved in thedecisions that affect their lives; in less affluent



222

communities, the development is disproportion-ately smaller (Patrick, 1969). Finally, sex identi-ty was chosen as an independent variable be-cause of the assumption that adolescents of bothsexes are similar with regard to their attitudinaldevelopment in these domains, and that therewould be no significant sex differences in validmeasures of alienation.

Construcbon of a Representative Item Pool

The item pool was created from a variety ofsources that included other attitude scales, theunderground press, rock music lyrics, and inter-views with adolescents. The conceptualization ofadolescent alienation, however, guided the selec-tion of all items. Each of 94 Likert-type itemswas identified with one of the five domains ofadolescent alienation, and there were approxi-mately 18 items for each of the five domains. A

sample item from each of the five domains is

given in Table 1.To establish content validity, the item pool

was reviewed by five individuals with experiencein attitude scale construction. The experts wereasked, subsequent to reading the conceptualiza-tion scheme, to indicate whether the items tenta-tively identified with each of the domains wereaccurate operational exemplars of the domains.If more than one of the experts failed to endorsethe placement of an item, the item was elimi-nated from the set. Twenty-one items were elimi-nated by this screening, including six from pow-erlessness, five from role estrangement, threefrom meaninglessness, two from guidelessness,and five from cultural estrangement. Theseitems were satisfactorily categorizable and rep-resentative of domains commonly characterizedas alienation, although this in itself does notconstitute evidence for the psychological struc-

.

Table 1 .

Representative Item Statements from94-Item Alienation Pool

* Reversed item.



223

ture of alienation.

Simultaneously, the readability of the items inthe pool was determined by the administrationof the item set to a group of 145 ninth-grade stu-dents. Any item was eliminated that five per centor more of the students indicated they could notunderstand. Eleven items were identified as un-

satisfactory by this criterion; five of these wereitems already eliminated by the expert screen-ing. Thus, twenty-seven items were eliminatedby the expert and student screenings, leavingsixty-seven items for statistical analysis. Each ofthe five domains was represented by approxi-mately 13 retained items.

SubjectsThe student sample consisted of 500 adoles-

cents enrolled in grades nine and ten in publicschools in four Minnesota communities selectedto represent the spectrum of schools in Minneso-ta. The schools included two from different

areas of a large city, one from a suburban areaand one from a rural community. (For want of amore convenient term these different areas will

be referred to as communities, although only therural group clearly constitutes a community inthe full sense of the word.) Of the students in the

sample, 242 were males and 258 were females.The student sample included 100, 60, 50, and 30percent of the rural, suburban, working-class,and inner-city schools, respectively; selectionwas made from among the classes of socialstudies teachers who were interested in co-

operating with the research project. All the

classes were untracked and heterogeneous intheir particular schools. The students, by virtueof being at least marginally successful in publicschools, were thus not representative of the fullrange of alienation in adolescents.

Measures

The measures employed in the study includedthose derived from the adolescent alienationitem pool, which are described in the analysissection, and five independent variables, corres-ponding to the sociological determinants hy-

pothesized under the conceptualization. Per-ceived Ability is the student’s self-rating of hisability as a student, as &dquo;above average,&dquo; &dquo;aver-

age,&dquo; or &dquo;below average.&dquo; Sex is the student’sself-identification as male or female. Parent Oc-

cupation is the student’s report of the occupa-tion of the head of household, classified accord- -ing to the U.S. Bureau of the Census system(1970), and trichotomized into levels, subse-

quently categorized &dquo;professional,&dquo; &dquo;skilledlabor,&dquo; or &dquo;unskilled labor.&dquo; Parent Educationis the student’s report of the highest level for thehead of household, categorized as &dquo;attended orgraduated from college,&dquo; &dquo;graduated from highschool,&dquo; or &dquo;attended grade school.&dquo; The Com-munity categories are &dquo;suburban,&dquo; &dquo;city work-ing class,&dquo; &dquo;inner city&dquo; or &dquo;rural.&dquo; U.S. Bureauof Census income data (1970) was used to rankthe communities on affluence (in the above or-der). Clearly, these variables are somewhat re-dundant indicators of the general domain of so-cioeconomic status (SES), but they were sep-arately retained to determine which aspects ofSES were most relevant to expression of aliena-tion.

Analysis and Results

The task was to seek structure in the amor-

phous concepts of alienation by analyzing re-sponses to items representing five rather looselydefined domains found in the literature. The

analysis had three stages: first, to identify theunderlying dimensionality of adolescent aliena-tion through factor analysis of responses to theitem pool; second, to construct simple item-clus-ter scores to approximate the factors; and third,to ascertain the discriminant validity of the clus-ter scores by comparing their relationships tothe independent variables with expected rela-tionships.

Because the results of each stage of the anal-ysis determined the details of the next stage, theanalysis and results are presented below in anintegrated format.

Identification of Empirical Dimensions

The primary function of the factor analysis



224

was to determine whether the initial five do-mains could be represented by independent di-mensions or, if not, how the domains wouldcombine into a smaller number of factors. Fur-ther, factor analysis was used to determine ten-tative clusters of intercorrelated items and toeliminate items that showed no strong relation-ship to any of the dimensions.

Principal components analysis with varimaxrotation was performed on the entire sample of500 subjects. Although six components wereidentified by the criterion of eigenvalue 1.0, ac-counting for 74% of the total variance amongthe 67 items, rotations of the first 2, 3, 4,and 5components had well separated eigenvalues. The&dquo;scree&dquo; criterion advocated by Cattell (1966)suggested that a three factor solution would beinterpretable.

After prolonged reflection on the loadings andthe conceptual meaning of the items, it was de-cided that the varimax three-factor set did in-deed provide the best characterization of re-sponse patterns; the set of high-loading itemsfor the three rotated factors was congruent withthe theoretical structure of alienation, while thepatterns of loading for the sets of two, four, fiveor six rotated factors were more difficult to in-

terpret.’ The first three factors accounted for55% of the total variance.Two of the factors were congruent with the

cultural estrangement and guidelessness con-ceptions ; the highest-loading items on factors 2and 3 were those initially identified as belongingto these concepts. The first factor seemed to bean amalgam of the hypothesized concepts ofmeaninglessness and powerlessness. The factoranalysis indicated that the conception of role es-trangement was not empirically coherent; theitems associated with it had little in commonwith each other or with other item sets, and this

concept was thus abandoned.The first factor accounted for over 25 % of the

total variance in the item responses and loaded

heavily on items that began with or includedpersonal pronouns. More particularly, all theitems in this factor tapped the adolescent’s per-ception of his competence in knowing about anddealing with his world. Since this dimension re-lated to the individual.’s feeling of personal effec-tiveness or capacity, it was named &dquo;Personal In-capacity.&dquo;The second factor accounted for 15% of the

total variance. The items were indicators of a re-

jection of middle class values. Dominant in thisdimension was a rejection of a perceived over-emphasis on materialistic values. Although intwo instances the items asked the adolescent to

pass judgment on the general state of Americansociety, the items were mainly attitudes aboutspecific elements such as school, dress, successand the importance of social status. This factorwas thus given the name of the initial concep-tion, &dquo;Cultural Estrangement.&dquo;The third factor accounted for 15% of the

total variance. The items in the factor reflectedan agreement that society fosters infinite desiresbut provides limited means for satisfying thesedesires. Items in the factor measured rejection ofnot the culturally prescribed goals but the so-cially acceptable means of achieving the goals.This factor, too, was named for the initial con-ception, &dquo;Guidelessness.&dquo;

Construction of Reliable Cluster Scores

For convenience in administering and scoringthe instrument and in explaining the meaning ofthe dimensions, an attempt was made to repre-sent the three factor dimensions by simple clus-ter scores. The adequacy of the cluster scoreswas assessed by: (1) how large their internal con-sistencies were; (2) how exclusively items relatedto only one of the three scores; and (3) how smallthe cluster intercorrelations were. Althoughthere was nothing in the theory that would pre-clude correlations among different forms of

alienation, the usefulness of the scores for em-phasizing the multidimensionality of alienationwould require a fair degree of independence.

’A list of items, the correlation matrix and factor matrices,are available from the authors for those readers inclined totheir own prolonged reflection.



225

t/)a0

4-)to

zLioU

N

(oUtn

cli I

Zm m- P-0 -.’0h- -0

c<0

NE

Ha)

0-Eeo00

EQJ4*

-0NnmuN

$3

<u0)&-<u>dex:

.0

TB.0Q1r-

ImZN<uoo

OJ

-d2:-

m

rUN

N(n

UcNNILLr-

TI’ 4)

j 3o

0..

) Ej mi4-

’v00oiX-o.0

E0

..,m



226

Beginning with simple sums of items that hadfactor loadings of .35 or higher, items wereadded or deleted to optimize conditions (1), (2),and (3) above. At each step, Cronbach’s (1951)alpha coefficient of internal consistency, cluster-item correlations, and cluster intercorrelationswere recomputed. Twenty-nine items comprisedthe initial clusters, and 13 more items were suc-cessfully incorporated; one item of the initial 29was omitted. The final clusters comprised 41items: 14 in Personal Incapacity, 15 in CulturalEstrangement, and 12 in Guidelessness. Repre-sentative items from the final clusters are dis-

played in Table 2.Internal-consistency reliabilities of the final

cluster scores were .80 for Personal Incapacity,.70 for Cultural Estrangement, and .67 forGuidelessness. The correlation between Cultural

Estrangement and the other two scores was low:.03 with Personal Incapacity and -.11 with

Guidelessness. The correlation of Personal In-

capacity and Guidelessness was a moderate .43.The .43 correlation of Personal Incapacity

and Guidelessness clusters was somewhat higherthan expected but still low enough to argue thatthe two factors measure different dimensions ofadolescent alienation. Since both dimensions re-late to alienation that is essentially involuntaryand to the adolescent’s response to his per-sonalized experience, it is reasonable that someof the feelings inherent in each of the two dimen-sions are common to the other.

(Incidentally, in spite of the persuasive thrustto the factor analytic approach, the authors sat-isfied their curiosity by calculating internal con-sistencies for the five clusters corresponding tothe initial domain identification of the items.

They were reassuringly low.)

Group Differences

Because of correlations among the five inde-

pendent variables, it was impractical to run five-way factorial analyses of variance. Instead, one-way analyses of variance were run on each of thefive independent variables; these were run sepa-rately for each of the three dimensions of aliena-

tion as criteria, giving 15 one-way analyses ofvariance in all. The results are displayed inTable 3. Within-group variances are substan-tially the same across groups in all 15 analyses,as is evident in Table 3; there was only one in-stance of significant heterogeneity of variance(Bartlett-Box F test, p < .05), which resultedfrom unusually consistent and low Cultural Es-trangement responses of students in the ruralcommunity.

Although a multivariate test comprising allthree criteria was not explicitly made, the uni-variate results indicate a clearly significant mul-tivariate relation: of the fifteen one-way analysesof variance that were performed, seven showedsignificant differences between groups at the p <.001 level, and four more at the p ~ .05 level; andtwo of the four insignificant differences hadbeen hypothesized not to be significant.

Table 4 summarizes the significant differ-

ences among groups and indicates where there

were discrepancies from the hypothesized grouprankings. In each case the discrepancy can becharacterized as a single shift in order of onegroup to an unexpectedly extreme position.Perceived ability. The differences among the

ability groups in Personal Incapacity and Guide-lessness were significant at the p < .001 level inthe hypothesized directions. Adolescents who re-ported high ability as students also had feelingsof efficacy in the world outside the school; theseadolescents indicated that they can know and in-fluence their social worlds. They felt, further-more, that the rules that guide them are under-standable and that others are not getting morethan their share by breaking the rules. The dif-ferences in Cultural Estrangement among thethree ability groups were significant at the p <.01 level but not in precisely the direction thathad been predicted: the average-ability studentswere less culturally estranged from middle classvalues than were above average students, but

low-ability students were even more estranged.Parent occupation. The differences in Per-

sonal Incapacity among the three occupationalgroups were significant at the p < .01 level, in thehypothesized direction. The differences in



NUco

io

>

4-0

NNt/)

b

t0

Q

b<a##<uc:O

&-04-

t/)m

-

Cl)u-

U z

r 10m c:o oH

N

N-

fl5Ut~)

c:0

5ocm

Ia

c:0

U%u

4-)t/)

4.Jtis4 )V)

CL:30

C3

o mc*m

tI)Mor-

z

0)nm3

M

EN+’

mr

oUM

oC

-C:u

·rS

S-O4-

mM

o

..c:+’

iS-’0

<n

a> -o 0S- 0C7) .

r V-’- Ct.

<~ -}tP itU itN

’~’7-Q=t r&horbar;

t~ 0

6 1%oCL

Mi it0) itJ~E

.«~ 0’&dquo;

x Lf51 0I .

g VP CLo~ ~



(ALdm<0n:

CD.::s0S-CD

r-

~0=3

.¡..J(J«

&dquo;’Cc’0

’0a)N

lqr --

v tn

cu (D- -C-0 4-~)<0 0I- CL

~-s=(L)a)B:4-)(U-0

0>vRCL)-0a0Q.en(DS-&-0Cm

x

T3N4-ttoUor-

cOr-

QJs-’0

MQJUcUS-U4-4--0

GL

oS-

~0

r34-11Uto

cto

T3NNor-

Us4J0r-L

sCtoUor-

4-or-

Cenor-

tnI

02:

Ii

QJ02=



229

Guidelessness among the three occupationalgroups were significant at the p < .001 level, alsoin the hypothesized direction. While the meanscores on Cultural Estrangement revealed slightdifferences in the direction of the hypothesizedoutcome-the higher the socioeconomic status,the higher the cultural estrangement the dif-ferences were meager and not statistically signi-ficant.Parent education. The differences in Per-

sonal Incapacity and Guidelessness among thethree educational groups were significant at thep < .001 level in the hypothesized direction. Theresults supported the proposition that adoles-cents from homes where educational level of thehead of the household is high were less subject tofeelings of guidelessness than were adolescentswith less educated parents. The differences inCultural Estrangement among the three educa-tional groups, although essentially in the hy-pothesized direction, were not significant at thep < .05 level.

Community. The differences in Personal In-

capacity among the community groups were sig-nificant at the p < .001 level in the hypothesizedorder: the higher the average income of a com-munity, the less likely the adolescents were tofeel personal incapacity. The differences inGuidelessness among the four communitygroups were significant at the p < .01 level. Themean differences showed that the more advan-

taged the community in terms of average in-come, the less the adolescents in the communityfelt guideless, with one exception: the adoles-cents in the inner-city neighborhoods felt the ef-fect of guidelessness somewhat more than didthe more advantaged adolescents. The differ-ences among the adolescents in the four commu-nities were significant at the p < .001 level onCultural Estrangement. The group means werein the order of average income with one excep-tion : adolescents from the working class weremore culturally estranged than any of the otherthree groups. The standard deviation for therural group on Cultural Estrangement was lessthan half of the other community groups. Al-though the rural group also reported the least

cultural estrangement, their scores were stillwell above the minimum possible score for thescale; so the small spread of responses cannot beattributed merely to a floor effect.Although communities did not fall strictly in

line in order of affluence on the three scales, theoverall alienation pattern still shows the com-munity &dquo;distances&dquo; ordered by affluence. Figure -1 shows the community group means plotted ina three-dimensional space defined by the threealienation scales. Even though the metric issomewhat arbitrary, it is obvious that communi-ties closest in affluence are also closest in posi-tion.

Sex. There was no significant difference be-tween the sexes on Personal Incapacity and Cul-tural Estrangement. This was as predicted. Thedifference between the scores on Guidelessness,however, was significant at the p < .01 level. Fe-males felt a greater disparity between their am-bitions and the means to achieve them.

Discussion

The analysis provides substantial confirma-tion for the construct validity of the three dimen-sions of adolescent alienation. The group differ-

Figure 1

Pattern of Community Group Means on Three Dimensionsof Adolescent Alienation

(The range shown for each dimension is approximatelyt 0.5 SD to either side of the pooled grand mean.)



230

ences for Personal Incapacity were as expected;the differences for Guidelessness were as ex-

pected, with two minor exceptions; differencesfor Cultural Estrangement were as expected,with some distinct exceptions.

Discrepancies in guidlessness. Both males

and inner-city adolescents were unexpectedlyhigh on Guidelessness. The sex difference mightbe explained by the differences in the process ofidentity formation in early adolescence: for boysthe key developmental task of this period is theestablishment of independence; for girls theprocess of identity is less discontinuous, con-cerned more with deepening extant relationships(Elkind, 1971). (Perhaps this difference will fadeas the new role directions fostered by thewomen’s liberation movement permeate adoles-cent consciousness.)

,

The reversal of order for the two communities

highest on Guidelessness might be explained bya truly greater complexity of a metropolitan areawhere the rules may appear highly ambiguous toadolescents; the simpler rural environment is

probably easier to interpret.Discrepancies in cultural estrange-

ment. The construct validation for Cultural

Estrangement was not as clear cut as for theother two dimensions. As expected there weresignificant differences among the ability groupsand community groups, but not precisely in thepredicted order.

Contrary to the initial conjecture, low-abilitystudents were more culturally estranged thaneither of the higher-ability groups. This is un-derstandable if Cultural Estrangement is con-sidered to be essentially a measure of lack of ad-justment to middle-class values. The superiorstudent who is secure in his ability as a student isapt to develop the intellectually critical stancetoward middle-class values inherent in Cultural

Estrangement, while low-ability students are in-clined to reject cultural dictums that are cast interms of values and opportunities that they feelare beyond their reach.The communities fell in the expected order on

Cultural Estrangement except that the highestlevel of alienation was found for the working-

class community rather than for the suburbancommunity. This deviation from the simplistic&dquo;advantage&dquo; proposition is not unduly disso-nant because the working class has recently hadspecial cause for this type of alienation. Duringthe past decade the working class in Americahas been alternately ignored and caricatured.All around them, people of the working class seechange and ferment; groups above and belowthem in the social structure are seen obtainingadvantages, while the working class is ignoredand derided (Simon, Gagnon, & Carns,1969). Insum, members of the working class increasinglyseem to feel poorly served by the institutions ofAmerican society. Although the rural group hadan unusually small spread of scores on CulturalEstrangement, they showed no such differencefor the other two scales; apparently rural stu-dents shared a highly uniform perception oftheir cultural identity.The lack of significant difference among

socioeconomic status groups, defined by parenteducation and parent occupation, was surpris-ing. Perhaps it can be understood if the his-torical context of the late 1960’s and early 1970’sis considered. That short but vivid period repre-sented an aberration in American social history.Inner city and campus riots, political assassina-tions, and especially the Vietnam War left in-delible marks on the era (Yankelovich, 1974).This maelstrom of events may have combined tomold the attitudes of adolescents at all socio-economic levels in a similar fashion. With the

winding down of the war, the situation may bereverting to normal. In fact, preliminary analysisof more recent data (Mackey, 1975) indicatesthat class differences more in line with those

predicted are now appearing.

Conclusions

The inquiry has sought to define alienationmore precisely by developing and validating aset of attitude scales to measure dimensions ofalienation in adolescents. Analysis demon-strated that adolescent alienation could be char-acterized by three independent dimensions,



231

which were measurable with an efficiently ad-ministrable and easily interpretable set of threescales. The three scales, each comprising about15 items, are: Personal Incapacity, the feeling ofnot having the skills needed to succeed; CulturalEstrangement, the rejection of the predominantcriteria for success; and Guidelessness, the re-

jection of socially acceptable means of achievinggoals. Patterns of significant group differenceswere substantially the same as had been hy-pothesized, except for Cultural Estrangement.Even there, the exceptions could be accountedfor tenably. (Some deviation from the expectedpattern is entirely tolerable in the construct vali-dation of scales; if the precise pattern of aliena-tion were already predictable, there would belittle reason to measure it.)

Greater facility in measuring alienationshould lead to : (1) expansion of research to dys-functionally alienated groups (which could pro-vide further validation for the instrument); (2)identification of groups or individuals that couldbenefit from special instruction or counseling;(3) evaluation of school and community pro-grams, including differential effects on groupswith different alienation patterns; (4) correlationwith general climate of schools and institutions(apart from curriculum or service content) asmeasured by such instruments as the LearningEnvironment Inventory (Anderson, 1970) or theMinnesota School Affect Assessment (Ahlgren,Johnson, & Johnson 1973); (5) correlation withcurrent behavior and, subsequently, behavioralprediction in longitudinal studies; (6) investiga-tion of the etiology of alienation, perhapsthrough correlation with traits such as thosemeasured by the MMPI, the Study of Values(Allport, Vernon, & Lindzey, 1960), the SixteenPersonality Factor Questionnaire (Cattell, 1973),objective derivatives of Kohlberg’s Moral Judg-ment Scale (Rest, Cooper, Coder, Masanz, &

Anderson, 1974), and actual measures of ability.The authors have already begun investiga-

tions in some of these directions. In general,with the greater clarity afforded by the scales inthe definition of adolescent alienation, re-

searchers may be able to investigate more mean-ingfully its incidence, correlates and causes.

References

Ahlgren, A., Johnson, D. W., & Johnson, P. Minne-sota school affect assessment. Center for Educa-tional Development, University of Minnesota,1973.

Allen, D., & Sandhu, H. Alienation, hedonism, andlife vision. Journal of Criminal Law, Criminologyand Police Science, 1967, 58, 325-329.

Allport, G., Vernon, P., & Lindzey, G. A study ofvalues. Boston: Houghton Mifflin Co., 1960.

Anderson, G. Effects of classroom social climate onindividual learning. American Educational Re-search Journal, 1970, 7, 135-152.

Cattell, R. The meaning and strategic use of factoranalysis. In Raymond Cattell (Ed.), Handbook ofmultivariate experimental psychology, Chicago:Rand McNally and Company, 1966.

Cattell, R. Personality and mood by questionnaire.London: Methuen, 1973.

Clark, J. Measuring alienation within a social system.American Sociological Review, 1959, 24, 849-852.

Clarke, J., & Levine, E. Marijuana use, social discon-tent, and political alienation: A study of highschool youth. American Political Science Review,1971, 65, 120-130.

Coleman, J., Campbell, E., Hobson, C., McPartland,J., Mood, A., Weinfeld, F. & York, R. Equality ofeducational opportunity. Washington: Govern-ment Printing Office, 1966.

Cronbach, L. Coefficient alpha and the internalstructure of tests. Psychometrika, 1951, 16, 297-334.

Dean, D. Alienation: Its meaning and measurement.American Sociological Review. 1961, 26, 753-758.

Elkind, P. A sympathetic understanding of the childsix to sixteen. Boston: Allyn, Bacon, Inc., 1971.

Etzioni, A. The active society. New York: Free Press,1968.

Flacks, R., Social and cultural meanings of studentrevolt: Some informal comparative observations.Social Problems, 1970, 17, 340-357.

Gamson, W. The fluoridation dialogue: Is it an ideo-logical conflict? Public Opinion Quarterly, 1961,35, 526-537.

Horton, J. & Thompson, W. Powerlessness and politi-cal negativism: A study of defeated local refer-endums. American Journal of Sociology, 1962, 67,485-493.

Keniston, K. The uncommitted: Alienated youth inAmerican society. New York: Harcourt, Brace andWorld, 1965.

Mackey, J. The dimensions of adolescent alienation.Unpublished doctoral dissertation, Indiana Uni-versity, 1973.

Mackey, J., The discriminant validity of an adoles-cent alienation scale. Paper read at the annual



232

meeting of the American Education Research As-sociation. Washington, D.C., 1975.

Nettler, G. Measure of alienation. American Socio-logical Review, 1957, 22, 670-677.

Olsen, M. Two categories of political alienation. So-cial Forces 1969, 47, 288-299.

Patrick,-J. Democratic political orientation of ninthgrade students in four communities. Unpublisheddoctoral dissertation, Indiana University, 1969.

Rest, J., Cooper, D., Coder, R., Masanz, J., &Anderson, D. Judging the important issues inmoral dilemmas—An objective measure of de-velopment. Developmental Psychology, 1974, 10,491-501.

Seeman, M. On the meaning of alienation. AmericanSociological Review, 1959, 24, 783-791.

Seeman, M. Alienation, membership and politicalknowledge: A comparative study. Public OpinionQuarterly, 1966, 30, 353-367.

Seeman, M., & Evans, J. Alienation and learning in ahospital setting. American Sociological Review,

1962, 27, 772-782.Simon, W., Gagnon, J., & Carns, D. Working class

youth: Alienation without an image. National As-sociation of Secondary Schools Bulletin, 1969, 53,60-70.

Templeton, F. Alienation and political participation.Public Opinion Quarterly, 1966, 30, 249-261.

U.S. Bureau of the Census. Census of population:1970. General Social and economic characteris-tics. Final Report PC (1) C-25 Minnesota. Wash-ington: Government Printing Office, 1972.

Westby, D., & Braungart, R. Class and politics in thefamily backgrounds of student political activists.American Sociological Review, 1966, 31, 690-692.

Yankelovich , D. The new morality. New York: Mc-Graw-Hill, Co., 1974.

Author’s Address

James Mackey, 152 Peik Hall, University of Minne-sota, Minneapolis MN 55455.



dimensions of adolescent alienation - university of minnesota · dimensions of adolescent...

Documents