一 0.16 YOUTH INDEPENDENCE

- 3.21 21.23 4.99 YOUTH IDEAL SOCIAL DIMENSION

3.92 9.60 11.88 CONTEMPORARY YOUTH LIFE PATTERN

TRADITIONAL EXPECTA TIONS - 5.77 -' 8.42 - 4.11

OF YOUTH

一 0.24 1.33 6.93 YOUTH INDEPENDENCE

YOUTH IDEAL SOCIAL - 2.79 - 5.09 - 0.93

DIMENSION

CONTEMPORARY YOUTH 1.72 - 5.61 -18.40

LIFE PATTERN

TRADITIONAL EXPECTA TIONS - 6.86 - 5.03 一 7.16

OF YOUTH

1.21 - 3.13 - 2.50^, YOUTH INDEPENDENCE

YOUTH IDEAL SOCIAL - 2.61 - 6.86 -17.38

DIMENSION

一 184

-AttitudinarDifferences Between High School Students and 1heir Parents

ìñ

U.S.A.:A Case Study ofGeneration Gap which contains the loadings of three-mode factors. Entries can be defined as the hypothetical judgments of three idealized individuals on the four concept factors against the five semantic criterion dimensions. It is interesting to note that three subjects appe釘 to be more simi1ar than different in the pattern of their judg間

ments' of concept facíors against semantic criteria. This is expected to be the case since the first dimension 一 accounting for over 35 percent of the total sum of squares in the unrotated subject coefficent matrix - is a representation of the average 叫阿ects (group mean) with' 趾拉 loadings'fro)J1 all individuals. In any case

,

some consistent patterns of ''judgments'' in the rotated solution appe缸

interesting. For example

,

except for the Uniqueness dimension

,

concept factors

“

CONTEMPORARY YOUTH'S LIFE PATTERN" and

“

TRADITIONAL EXPECTATIONS OF YOUTH" always have different size. That is, DRESSING SLOPPIL Y

,

FREE TO LEA VE HOME and DRINKING ALCOHOL (in the

“CONτEMPORARY YOUTH'S LIFE PATTERN" dimension) are always better E+), stronger (P+) , more predictable (Predictabi1íty+) but less active (A一) as compared with PARENTS IMPOSING CURFEW, PARENTS CONSULTATION ON SPENDING MONEY, SHOWING RESPECT TO AUTHORITY and NEAT-NESS OF A BEDROOM (in the

“

TRADITIONAL EXPECTATIONS OF YOUTH" dimension) which are considered as bad (E一)，weak (P一)，unpredictable

(Predictabi1íty一) but very active (A+). Other comparisons can also be made from this Table.

Notes

77. Tzeng, Application of Semantic Differential technique.

78. Tzeng, Differentiation of Affective and Denotative Meaning Systems 的

Personality ratings.

79. Hannan

,

Modern Factor Analysis.

80. Tucker

,

Three-mode Factoy Analysis.

一 185 一

Attitudinal mgbmnces BEtween mgh School Students and 刃1eirparentsin U.S.A.:A CaSE Study ofGEneradon Cap

CHAPTERIV

DYNAMICS OF GENERATIONAL GAPS

The purpose of this chapter is to investigate the possible dynamics of generation gaps between high school students and their parents. Sixteen unidi-mensional variables were constructed to cover reported sources of psychological and socio呵cultural correlates which contribute the so~alled generation gaps.

Ra tings on these measures from high school students of both sexes and their parents were subjected to ana1ytic treatments of various uni- and multivariate statistic techniques. The potential dynamics of generation gaps wi11 then be theorized from cross-sex and cross-generation comparisons of their statistical solutions.

Method

Subjects

Subjects reported in Chapter III were a1so used in this part of the study.

They are 47 high school male students and their parents and 41 female students and their parents.

Procedures

A sixteen-item questionnaire was constructed to cover representative sources or correlates of generation gaps, "based on the literature review, as reported in Chapter 1, and subjective observatíons (the author was a high school teacher and has a daughter who was a member of the fema1e student subject pöpulation).

Each item shown in Table 19 was "rated on a 7-step bipolar sca1e (scores from +3 to ^• 3 with +3 to the left and -3 to the right poles). Since these items were inc1uded as pélrt of the entire questíonnaire booklet as described in Chapter III

,

all subjects âid th巴 ratingsas a ta1ce home task.

Results and Discussion

Means of the 16 items for aU four groups are presented in columns 2-5 of Table

20.

A summary of the ana1yses ofvariance performed on these variables is presented in columns 6-8. Results from inter-group comparison on the group

一 187 一

Bulletin of National Taiwan Normal University No. 28

mean ratings ofnine significant variables are also presented in Table 20.

Among the nine variables with an F ratio significantly beyond the .05 level, two patterns of generational differences emerge: Students of both sexes considered rock music being inore enjoyable (item 6), and also they have higher level of involvement in sports and physical activities (item 12). On the other hand

,

parents of both studëilt .groups considered that the general social standards of their children were higher than their chi1dren's opinions of parents standards (item 1), .and when there were conflicting thoughts for youth, parents opinions should be more influential (item 11). Parents also perceived that the general social standards of their peers are better than those of their children's peers (item 2) and that their peers' opinion of them as a person were high (item 4).

As to the issues on generalso.cia1 po1itical system

,

the religious belief

,

and saving money for the future usage

,

parents gave significant1y more favorable responses.

In terms of pairwise comþarisons of groups mean ratings on the nine significant variables

,

two types of differences were observed: (1) Within generationaZ differences - comparisons between two students groups (m泌的

vs females). While no significant difference is found between .two parent grou阱，

there are some- differences between two student groups. That 誨， male students are more favorable to the political system in this country (item 7) and have

hi前er involvement in sports and physical activities (item 12). (2) Between genera.tionaZ'comparisons - comparisons of each student group with both parent groups. It is interes討ng to note that for both male and female students, the areas (or items) of generational differences are a1most identical for within and outside the families. That is

,

the idifferences between male students and their own parents and those between niale students and parents of female s切dents

are identical - ma1e students are more enjoyable. in rock music (item 6) and inore active in sporls and physicalacti世ties (item 12),' parents of both student groups have relatively h.igher va1ues on item 1 (the generalsocial standards of their chi1dren), item 2 (the general social standards of parents' own peers), item 10 (saving money for the fu削reas opposed to spending it now)

,

item 11 (parents' opinions .shou1d be more influential when there are conflicting thoughts for youth

一 188 一

Attitudinal mgkrenCES BEtween mgh School StudEnfs and D1eirparents in U.SAJA CGSE Study ofGeneration Gap

Table 19

Unidimensiona1 Variables

1. The genera1 social standards of my parents (or my child if answered by parents) are: (good vs bad)

2. The general social standards of my peers are: (good vs bad) 3. My own opinion ofmyself as a person is: (high vs low) 4. My peers opinion of myself as a person is: (high vs low)

5. Parents' (or students') satisfaction with me as a person is: (favorable vs unfavora ble)

6. Rock music is: (enjoyable vs unenjoyable)

7. The politica1 system in this country is: (satisfactory vs unsatisfactory) 8. The ^so自called generation gap between parents and children in my family

does: (exist vs not exist)

9. My distance from my parents (or my children if answered by parents) in most of their opinions is: (far vs close)

10. Saving money for the future as opposed to spending it now is: (good vs bad)

11. In genera1, when there are conf1icting thoughts for youth, whose opinions should be more influential: (parents vs peer group)

12. My own level of involvement with sports and physica1 acti討ties is: 他站h

vs low).

13‘ My level of boredom is: (high vs low)

14. My (i此， the student) 趾拉 school education is: (satisfactory vs unsatis-factory)

15. My satisfaction level of childhood in general was: (high vs low) 16. 1 consider that the religious beHef is: (important vs unimportant~

-189 一

Table 20

Summary Statistics of 16 Unidimensional Variables

Significant t testsa

Pairwise comparisons inc1ude A: Male students/Parents ofMa1e'studets,iB: Male ~tqdents/Female studc;mts, C: 、 Mal~，

students/Parents ofFemale shldents,D:Parents ofMale studentdlFemale students,E:Parents ofMale st114mts/PMeats of Female students, F: Female students/Parents of Female students. Non:-significant enlries are omitted: 三

* P ~ .05 牢牢 P ~ .01 ***p ~ .001

只.這λ-HNSAEEg正是§主自~惡意遺忘泛流通老

Attitudinal Differences Benveen High School Students and Their Parents in U.SA.:A Case Study ofGeneration Gap

(item 7), parenta1 opinions sho^u1d be more inf1uential when there are conf1icting thoughts for youth (item 11), parents consider the religious belief is more im-portant (item 16). However, thcre are two items which are different between the two generations but are not commonly different to both parent groups.

While item 2 is on1y significant between female students and their cwn parents (i.e., parents consider the genen

,

l social standards of parents' own peers 訂e

better), item 4 is significant between female students 血 d the p叮e口ts of ma1e students (i.e., parents of ma1e students have higher mean value on

“

my peer's opinion of myslef 品 a person刊).

It was designed that item 8 woul丘 measure the perceived levels of fhe so-called generation gaps in each fami1y from both ^studenτs and thcir pare^1).ts.

Therefore, it seems feasible to detem1Ïne the relationships betv:舟 en this perception and the other fifteen social and psychological variables for each of the fou1" groups. A scparate mu1tiple regression analysis was performed by treating item 8 as the clitelion variable and tb.e othcr 15 items as preιictor

variables. The standar吐 regression coefficients (B's) and the product-moment correlation coefficient (r's) of thεse variables with the criterion are summarized in Table 21.

As indicated by squared multiple regressio肘， at least 42 percent of the variance in perception of so-ca11ed generation gaps were accounted for among the four groups. The parents of fe^r.:1ale ^stude前s group i.n particular have the best predìction value with 74 percent of valiance accounted for. It should be noted that since all scores are bi-polar, negative signs o:f regression weights wil1 have meanings di:fferent from the usual prediction of unipolar scores (Tzeng

& Osgood^81). Variab1es attr也 uting significantly to the multip1e regressio了I

equations 0:'" the so心cal1edgeneration gaps are quite similar between male students and their parents. Two most important items are ,:11e ini:ergenerational distances in most of opinions at home (item 的 and the generaJ. social standards of

(chi1已ren's as well as parents') peers (item 2) - botl立來咐:可 εstand訂益 regression

weight significant beyond the .û 1 level. Among the remaini了19 varlables, the zero-order correlations wεrc signifjcant fcr item (the ge:nera1 social standar這s

of parents, when answerεd by the students, and of chUdren, whε<， answered by the parents), item 5 (parent<ch

一 191 一

Table 21

Prediction of Perceived Generation Gaps Within the Family

Parents of'

認-ARbBKEH忍之也妄kch§-dR~sahN2.片。shub3向

(.74)

Attitudinal Differences Between High School Students and

T有 eirParents in U.SA.:A Case Study ofGeneration Gap

12 are also significant - that is, the existence of the generation gap is highly correlated with parents' own low self esteem and lower level of involvement with sports and physical activities. Due to different intervariable relationships, some of these variables do not automatica11y become significant predictors.

For t的 [emale student group, item 9 (inter generational distances in most of their opinions) is c1early a significant predictor for the existence of the so-<;a11ed generation gap at home. Two other iterns are a1so rnoderately good in prediction, they are ite 3 (persona1 esteern of selves) and 10 (favorable attitude toward saving money for the future). However, in terms of product-moment correlations

,

both of these two i1:ems are not significant. Other variables with high r's va1ues include item 1 (genera1 social standards of their parents)

,

item 2 (genera1 social sta^Jldards of peers), item 4 (peer's opinion of themselves), item 11 (emphasis of peer group's influence on conflicting thoughts)

,

and. item IS (1ower ^‘satisfaction level of chi1d.hood in genera1). Compared with the solutions from the male students group, female students tend to rely heavi1y on self esteems and chi1dhood development in the course of inter-generation understanding.

In predicting the perceptions the parents o[ [emale students, the existence of so呵called generation gaps is best predicted, in order, by item S (the student's satisfaction with me as a person), item 9 (distances with chi1dren in most of their opinions), item 2 (general social standards of parent's own peers), item 15 (satisfaction level of parents' own chi1dhood)

,

item 3 (parents own opinions of selves), itern 4 (opinions of selves as a person from parents' own peers) and itern 14 (the satisfaction level of chi1dren's high school education). lt is interesting to note that like the solution of female students

,

a11 these items a1so cover rnostly the parents' own self esteem and development.

Since itern 9 (inter-generational distances in rnost of chi1dren's opinions) is high1y prec1ictive for the existence of generation gaps across a11 four groups, it is therefore to treat it as the criterion variable to be predicted from the remaining fourteen variables. As given in Table 22, multiple R's are higher than .59 for a11 four groups. The predictions of the two parents groups are special1y significant beyond the .01 level. For the male students group, four variables have si

-193 一

Table22

Prediction of Perceived Generational Distances on Opinions

Parents of

Attitudinal Differences Be趴leenHigh School Students and Y于leirParents

iñ

U.SA.:A Case Study ofGeneration Gap me as a person), item 3 (1ow personal opinion of self), item 6 (favorable attitude toward rock music) , and item 13 (parents own high leve1 of boredom). On the other hand, items, which do not contribute significant1y to the prediction but are high1y correlated with the criterion variable by r, inc1ude item 1 (general social standard of children), item 2 (social standards of parents own peers), and item 4 (parent's opinion of self as a person). Results from both male students and their pa.rents seem to suggest a possible relationship between paretns own socia1 and/or economic adjustments (not necessarily status) and their relation-ships with children.

For the lemale student group, three items are moderately predictive -item 1 (parents general social standards), item 4 (peer's opinion of slef) and item 14 (satisfaction leve1 of students' own high schoo1 education). Items 2 and 15 (i ι ， genera1 social standards of students' peers, and students' own satisfaction level of children in general) have significant correlations with the criterion variable. For the parents ollemale studetns, the general pattems of prediction and correl<:ttions found for the parents of male students group a1so hold, but with minor deviations in the order ofmagnitude among coefficients.

Note

81. Tzeng, O.C.S. & Osgood, C.E. Va1idity tests for Componential Analysis of Conceptua1 Domains: A Cross-Cu1tura1 Study in Methodology. Behavioral Sciences, 1976,21(2).

-195 一

Atdfltdinaf DifYMnces BEtween High School Sfudents and

'ih-;;;-P~ren

ts i;ì

U.Sλ :A Cúse Study ofGeneration Gap CHAPTER V

DISCUSSION AND CONCLUSIONS

The present research is a case study of the so-cal1ed generaticn gaps between

趾gh school students of both sexes and their ^pruτents. It is an explcratory study in both the theoretic31 and methodologica1 aspects. In theory, the nature and extent of generationa1 discrepancies are examined through emp出cal rLeasure-ments of inter-generation31 adjustrLeasure-ments and opinions; and in method, a more powerfu1 research strategy and technique is employed to gaurantee the content and construct v31idities of the soIutions. Intra- and inter-generation31 comparisons on issues of opinion differences and on soci31 and psychologic31 correlates of the discrepancies provide some promising new information on the na心re and dynamics of generation gaps. Since 311 of these aspects - theory，昂的ho泣， and resu1ts - and their imp1ications are equa11y important for fu切re studies in generations an吐 contexts of inter-person31 communications, the present chapter will discuss each of them separately.

On Theor泣ingof Generation Gaps

The gener31 purpose of this study is to detect the areas of generationa1 discrepancies between high school students and their parents in order to probe the influence of these discrepancies on students' soci31 adjustrnents ^ä....'1d person31ity development. Un1ike other research in generations where the ιomain

of issues was usua11y defined by 世le researchers, and where inter-generation31 differences in perceiving the issues were then regarded as the contributing factor to children's person31ity development and soci31 adjustments

,

the present research follow Tzeng's theoretica1 formu1ation and strategy with emphasis of direct and simu1taneous ev31uation of the three major variables involved in human cognition and judgments 一 individuals， objects, and underlying psychological critenon.

Tr.rough a naturalistic elicitation procedure, important issues which have significant effects on inter-generation31 communications and adjustments at home, were direct obtained from the subject population. As a resu1t, two types of discrepant opinions could be identified 一 one as being common to both generations, and the other being unique to parents or to students (of both or either sex). While the common items can be considered as mutual perceived generation gaps, the parent31 and/or chi1dren unique variables can be conceived

一 197 一

Bulletin 01 Natìonal Taiwan Normal University No. 28

的 partial perceived generation gaps. It seems reasonable to assume that in the course of inter-generational communications, mutual generation gaps are discrepancies known to both parties with obvious effects on their relationships and adjustments, but partial generation gaps are usua1ly unknown to either p訂ty

with only potential effects. In the present study, the m吋ority of reported discrepant issues belong to the common (known) generation gaps category and thus were used for subsequent intra- and inter-generational comparison. On the other hand, most of partial issues are not inter-related with very low reporting frequencies. Therefore, unless being highly correlated to mutual discrepant issues

,

they were not included for subsequent treatment.

Since underlying psychological criterion variables are the major deterrninant factors for human behaviors and intentions, each reported common variable should in theory have value with respect to any under1ying psychosemantic criterion. This imp1ies that each perceìved generational gap could be mathe-matically mapped onto various inter-generationa1 difference continuum. The differences of the perceíved values for an issue from the two generations should then be expected rangíng from zero to some maximal magnitude. For example

,

the concept FREE TO LEA VE HOME should have two judgment values on the good-bad scale from both parents and their children. The difference between the two values should reflect the generational difference in opinion of the concept.

However, íf one is able to identify an índíscrimínant ínterval around the va1ue zero to represent the acceptable simi1arity of psychologica1 characteristics of an íssue, then va1ues withín the interva1. would represent the pseudo-generationa1 gaps - reflecting that both generations actua11y perceive the same issue in a non-díscrimínant way with respect to the same psychological criterion. On the other.

hand, va1ues beyond the ínterval would índícate the true quantítatíve natures of generation gaps. In thís respect, the semantic differentia1 technique was used to measure 18 elicited opinions on a set of 29 bi-polar sca1es whích al1 have direct concept domain relevancy.

On the Method of Data Analysis

For the data of semantic differentia1 ratings from students of both sexes and their parents

,

factor analytic technique is the major treatment too1. It ís to identify the underIying features or structures dominating the inter-sca1e and inter-concept relationships for a11 four groups. However, it should be noted that

扭 usua1 app1ication of factor analysis, inter-variable correlations are usua1ly used

一 198 一

Attitudinal 間fferences Be似leenHigh School Students and 1heir Parents iñ U.SA.:A Case Study ofGeneration Gap 前 input (cf., Hannan⁸²). But according to Tzeng⁸³ a product-moment correlation coefficient is not sensiti.ve to the constant group mean differences in ratings of objects against various scales, and nor is r stable if some sca1e poles are presented in a differer..t form. Therefore, the cross-products of variables were use泣的 input

factor analysis in the present research.

Intra..generationa1 as well as inter-generationa1 comparisons are also made on both the concept and sca1e fac1:or structures across all four groups. Tl世sis to examine the generationa1 phenomina with respect to both macro (age-cohort)

Intra..generationa1 as well as inter-generationa1 comparisons are also made on both the concept and sca1e fac1:or structures across all four groups. Tl世sis to examine the generationa1 phenomina with respect to both macro (age-cohort)

在文檔中 Attitudinal Differences Between High School Students and Their Parents in U. S. A. (頁 78-99)