How do they see themselves: Assessing the self-concepts of Hong Kong young people in a cosmopolitan context Christopher H K Cheng, PhD City University.

How do they see themselves: Assessing the self-concepts of Hong Kong

young people in a cosmopolitan context

Christopher H K Cheng, PhD

City University of Hong Kong

“Understanding Hong Kong Youth Symposium”

The Eighth International Conference on Language and Social Psychology, July 10-14, 2002, Hong Kong.

Outline

Part 1: How do Hong Kong young people see themselves?

• Content structure (configuration) of self-concept• Age and gender characteristics

Part 2: Comparing the self-concept of young people from three cosmopolitan cities: Hong Kong, Singapore, and Australia

General

Self-Concept

Social

Self-Concept

Emotional

Self-Concept

Physical

Self-Concept

Academic

Self-Concept

English History Math Science

General:

Academic and Non-Academic Self-Concept:

Subareas of Self-Concept:

Peers Significant

Others

Particular

Emotional

States

Physical

Ability

Physical

Appearance

Evaluation of

Behavior in

Specific

Situations:

Shavelson et al.’s (1976) model of self-concept

Configuration of Hong Kong Adolescents’ Self-Concept

• Emic and qualitative approachConsiderations:• Potential “imposed etic” problem of impo

rted scales• Cultural sensitivity and validity (use of la

nguage?)• Psychmetric properties (Berry, 1980; Bochner, 1994; Hui & Triandis, 1985; Mark

us & Kitayama, 1991; Yang & Bond, 1990)

Assessing the Self

• From the talks and written responses to the classical “Who Am I” procedure (Kuhn & McPartland, 1954), a number of self-descriptors categories were generated.

Appearance & AttractivenessAppearance & Attractiveness

肥瘦高矮、四眼（帶眼鏡）、身材好、樣子甜美、醜樣

Physical & Non-acaPhysical & Non-academic abilitiesdemic abilities

身體弱 / 強、健康、喜歡游泳、反應快、有音樂天份

Family relationsFamily relations

關心家人、喜歡和家人一起、不想講我屋企人

Academic abilitiesAcademic abilities

成績普通、老師成日針對我、沒有進步、不能升預科

General intellectual abiGeneral intellectual abilitieslities

記性好、醒目、富創作力、蠢、聰明、有分析力 Academic attitude Academic attitude

& orientation& orientation

有恆心、上堂留心、心散，不專心上課、有上進心

Rule-abiding, selRule-abiding, self disciplinef discipline

情緒化、脾氣差、有耐性、做事有原則、無厘頭

Virtues & ConductVirtues & Conduct

不貪心、為人誠實、有責任心、為人著想、對人坦白、心地善良、老實 Social skillsSocial skills

口材好、有幽默感、害羞、廣結朋友、不曉得與人溝通

Social relationsSocial relations

有義氣、齊齊玩、「八卦」、不合群

Social mannerSocial manner

文靜、斯文、大方得體、儀態得體、有禮貌

PersonalityPersonality

開朗、活躍、文靜、被動、內向 / 外向

Filial pietyFilial piety

尊敬老人家、聽父母話、激父母、亂「洗」父母錢、孝順

GeneralGeneral

失敗、有自信、獨特、信自己

General

Self-Esteem

Physical

Self

Social Self

Family Self

Intellectual Self

Moral Self

Physical Attractive-

ness

Sport &

Physical abilities

Discipline & Self-Control

Social relations

Social Skills

Family Relations

Filial piety

Academic abilities

Intellectual abilities

Virtue &

Conduct

Altruism

A conceptual structure of Hong Kong adolescents’ self-concepts.

The CASES

• Based on the content analysis of the emic study, through vigorous instrumentation procedures, a measuring instrument called the Chinese Adolescent Self Esteem Scales (CASES) was developed.

Scales No. of Items Cronbach’s Alpha

General 8 .82

Social 10 .83

Intellectual 10 .81

Appearance 8 .81

Moral 8 .80

Family 8 .83

Physical/sport 8 .89

Total 60 .93

Table 1: Means, SD, and Internal Consistency Reliability of CASES

N=558

The CASES

• Testing the measurement models (i.e. the conceptual structure that the CASES measures) – SEM using LISREL

• One general factor or multiple factors?

• How many? One level or hierarchical?

General

Physical Appearance

Social

Moral

Family

Sport and

Physical AbilitiesIntellectual

Family

Sport & Physical Abilities

Moral

Intellectual

General

Social

Physical Appearance

Model * 2 d.f. 2/d.f. CFI NNFI RMSEA

M1: One general factor model 

3029.39 349 n/a .597 .564 .117

M2: First order six correlated factors model  

1369.70 335 118.55 .845 .825 .0745

M3: First order seven correlated factors model 

767.65 329 100.34 .934 .924 .0489

M4: Hierarchical seven correlated factors model

802.70 338 3.89 .930 .922 .0497

* M2: Six-factors model formed by combining Physical Appearance and Physical/Sports Abilities to one factor.

* M4: Hierarchical model formed by the General Self scale at the apex with the six domain-specific scales at the base, all seven factors were correlated.

2/d.f. (Ratio of the differential chi-square to differential degrees of freedom) was obtained by comparing the differential statistics between the said model and the preceding model. A statistic of 6.63 and 3.84 is deemed to be significant at .01 and .05 significance levels respectively.

Table 2: Goodness-of-fit statistics of CASES competing models

Confirmatory Factor Analysis (Structural Equation Modeling)

Confirmed:

• Six domain specific factors and one general factor

• Hierarchical model supported

General Self

Intellectual Self

Social Self

Moral Self

Family Self

Physical Appearance Self

Physical Abilities & Sport Self

Structural Model of the CASES

  General Social Intellectual Appearance Moral Family Physical

General 1.0            

Social .64 1.0          

Intellectual .81 .55 1.0        

Appearance .77 .52 .62 1.0      

Moral .49 .45 .39 .40 1.0    

Family .39 .30 .29 .31 .34 1.0  

Physical/Sport .48 .37 .44 .49 .20 .21 1.0

*All coefficients significant at p<.01, and were obtained by LISREL.

Table 3: Correlation between latent factors of the CASES (Hong Kong sample)

  Age Groups  

Scales 11-13 14-15 16-17 18-20 F-tests

General 4.43 4.43 4.48 4.44 .15 ns

Social 6.01 6.05 5.97 5.82 1.38 ns

Intellectual 4.29 3.98 4.14 4.19 3.36 *

Appearance 3.90 3.97 3.98 4.04 .52 ns

Moral 4.57 4.61 4.88 5.04 11.84 *

Family 4.80 4.86 4.93 4.93 .66 ns

Physical/Sport 4.19 4.12 4.23 4.18 .65 ns

* p < .01, d.f.=1, 829ns = non-significant

Table 4: Age Differences on Multiple Self-Concepts

Age Differences

Age Trends in Moral Self and Intellectual Self

3.5

4

4.5

5

5.5

11-13 14-15 16-17 18-20

Age

Mea

n

Intellectual

Moral

Age Differences

• Moral self increased with age (linear contrast significant)

• Intellectual self changed in a quadratic fashion – rapid drop from age 12 to age 14 (i.e. F.1 to F.3), then some recovery in higher forms

Scales Boys Girls F-tests

General 4.56 4.36 7.63 *

Social 5.91 6.05 3.25 ns

Intellectual 4.13 4.12 .01 ns

Appearance 4.12 3.85 14.96 *

Moral 4.64 4.85 11.37 *

Family 4.77 4.98 8.24 *

Physical/Sport 4.66 3.85 77.99 *

* p < .01, d.f.=1, 829ns = non-significant

Table 5: Gender Differences on Multiple Self-Concepts

Gender Differences

Self-Concept Means of Boys and Girls

3

3.5

4

4.5

5

5.5

6

6.5

Self-Concept Facets

Mea

ns

BoysGirls

Boys high:Physical self-concepts

(incl. appearance & physical abilities)

General self-esteem

Girls high: Moral self-concept

Family self-concept

~ similar levels in Social Self and Intellectual Self ~

Comparing the self-concepts of young people from three

cosmopolitan cities: Hong Kong, Singapore, and Sydney

ScalesAustraliaM(s.d.)

SingaporeM(s.d.)

Hong KongM(s.d.)

F-tests

General 4.32(.93) 4.09(.93) 3.49(.93) 49.60*

Social 4.02(.77) 3.89(.77) 3.31(.78) 58.08*

Intellectual 3.34(.93) 3.47(.93) 3.03(.93) 14.21*

Appearance 3.37(.87) 2.98(.87) 2.88(.87) 19.54*

Moral 4.34(.71) 4.14(.71) 3.50(.70) 92.72*

Family 3.18(.87) 4.04(.87) 3.45(.87) 59.12*

Physical/Sport 3.80(1.19) 3.47(1.18) 3.30(1.12) 10.38*

Notes:

All multivariate tests (Pillai’s, Wilks, Hotelling’s) significant at p<.001.

* p< .01, N(Singapore)=249, N(Australia)=211, N(Hong Kong)=251, d.f.=2,708

# Post-hoc tests of the underlined pairs were non-signficant, all other pairwise comparisons were significant at p<.01.

Table 6: Comparing the multiple self-concepts across Australia, Singapore and Hong Kong

Comparing young people in Hong Kong, Singapore, and Australia

General self-esteem Physical self-concepts

(appearance and abilities)

Social & Intellectual self-concepts

Moral Self Family Self

Reflection 1:Developmental Differences

Intellectual Self shows a clear drop from F.1 to F.3, then some recovery from F.4 onward, similar to other research (e.g. Cole et al., 2001; Lau, 1990; Marsh, 1989)

• Physical, cognitive, social transitions: Educational transition can be detrimental (Cole et al., 2001; Harter, 1998)

Moral Self shows a linear rise: consistent with cognitive development (from concrete to formal operations), i.e. adolescents are becoming more confident of their moral/ethical selves during their maturation

But no significant changes in other facets (incl. general self-esteem) – quite stable

Reflections 2: Gender Differences

Consistent with gender-role stereotypes, but boys tend to have higher general self-esteem

• Social desirability of masculine stereotypes?

• Impacts of the new secondary school placement allocation (SSPA) system?

Reflection 3: Challenges

Self-concepts of H.K. youth rather low but not entirely grim, e.g. similar to Singapore youth in family and physical self-concepts (incl. appearance & abilities)

• Need to note the weak facets – in particular intellectual & social self-concepts (academic pressure? peer’s acceptance? self-expectation and parent’s aspiration?)

• More work on enhancing self-esteem and self-efficacy in these facets is necessary (Note: enhancing specific facets will automatically enhance general self-esteem)