Eé WWW“WINWWW“llWlHMlWlml

   

540

THE 8'8

IlllflmlhllllHHHH‘HHIHUIUIIIHWIUWHIHIW

3 1293 01025 102

This is to certify that the
thesis entitled
AN EXAMINATION OF THE RELATIONSHIP BETWEEN TEACHERS'
AFFECTIVE GOALS AND STUDENT OUTCOMES IN
SECONDARY SCHOOLS

presented by

Elaine Marie Allensworth

has been accepted towards fulfillment
of the requirements for

M.A. degree jnSociology/Urban Studies

 

 

'1

Major profess!

Datei/A/l97L

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

 

LIBRARY
Mlchlgan State
University

 

 

 

PLACE N RETURN BOX to romovo this checkout from your record.

TO AVOID FINES Mum on or Moro date duo.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU to An Afflnnoflvo Action/Equal Opportunlly lm
gamma

 

 

AN EXAMINATION OF THE RELATIONSHIP BETWEEN TEACHERS'
AF FECTIVE GOALS AND STUDENT OUTCOMES IN SECONDARY
SCHCDLS

By
Elaine Marie Allensworth

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTEROF ARTS

Department of Sociology
and
Urban Studies Programs

1994

ABSTRACT

AN EXAMINATION OF THE RELATIONSHIP BETWEEN TEACHERS’
AFFECI'IVE GOALS AND STUDENT OUTCOMES IN SECONDARY
SCHOOLS

By

Elaine Marie Allensworth

A great deal of controversy exists regarding the benefits and costs
associated with promoting students’ self-esteem as a primary educational
goal. Additionally, there is little evidence that secondary schools actually can
have a substantial effect on students’ self-esteem. Therefore, this study uses a
national sample of 290 public high schools to compare changes in student
self-esteem according to their schools’ emphases on affective goals. Other
student outcomes (achievement test scores, drop out rates and college
enrollment rates) are also compared. Hierarchical linear modeling is used to
account for multi-level analyses. Results show little variance between
schools in mean student self-esteem, and no significant relationship between
affective goal emphasis and student self-esteem. However, schools that have
especially low affective goal emphasis have significantly lower mean
achievement test scores and college entrance rates than schools with average

or high affective goal emphasis.

TABLE OF CONTENTS

LIST OF TABLES

LIST OF FIGURES

INTRODUCTION

REVIEW OF THE LITERATURE
Self-Esteem as an Educational Goal

Background and Demographic Factors
Associated with Students’ Self-Esteem

METHODS
The Sample
The Variables
The Models
RESULTS
Schools and Affective Goals
Student Self-Esteem
Other Student Outcomes
DISCUSSION
Future Research

LIST OF REFERENCES

iii

Page

iv

10
10
10
12
16
16
16

21

31

LIST OF TABLES

Table 1. Predictors of Affective Goals

2. Self-Esteem Explained
by Student—Level Characteristics

3. Test Scores Predicted by
Student— and School-Level Variables

4. Predictors of Drop-out ‘
and College Enrollment Rates

iv

Page

17

19

25

LIST OF FIGURES

Figure 1. Mean Standardized Test
Scores by Affective Goal Emphasis

2. Percent of Graduating Class Going to
College by Affective Goal Emphasis

Page

INTRODUCTION

Teachers and administrators often claim that one of their educational
priorities is to increase student self-esteem. However, there is little evidence that
most schools which claim the improvement of students’ self-esteem as a priority
actually do have a substantial effect on students’ self-esteem. Especially at the
high school level, where there is less agreement on goals and methods and more
emphasis on particular curricular subjects, teachers’ impact on students’ self-
esteem may be negligible if not reinforced by other teachers in the school.
Teachers’ efforts might also be limited to affecting only students’ academic or
course-specific self—concepts, rather than enhancing students’ general well-being.
Additionally, attempts to improve self-esteem might differentially benefit those
students whose self-concepts are at greater risk within the school setting.

This study uses a sample of 290 public high sdrools from the High School
and Beyond Teacher and Administrator Survey of 1984 and Student Surveys of
1980 and 1982 to compare changes in student outcomes (i.e., reported self-esteem
and achievement scores) according to teachers’ prioritization of affective goals
within their schools. Schools are classified according to teachers’ ranking of the
goal of improving students’ personal growth and fulfillment (self-esteem,
personal efficacy, self-knowledge). The schools are then compared in terms of
changes in their students’ self-esteem scores and test scores over a two year
period. Differences in school effects on students’ self-esteem are also examined
among students of varying demographic and academic backgrounds, as previous
research has found differential school effects based on students’ race, gender,
economic background and academic records. Hierarchical linear modeling is

used to account for multi—level analyses.

REVIEWOFTHELITERATURE

glf-Esteem as an Educational Goal
A great deal of controversy exists regarding the benefits and costs

associated with promoting students’ self-esteem as a primary educational goal.
One debate questions whether or not schools and teachers can improve other
aspects of students’ lives by working to enhance students’ self-perceptions.
Student self-esteem has been positively associated with academic achievement,
behavior, prevention of delinquency, satidaction in school, less stress and
burnout, vocational identity and success in later life (Battle 1981; Bradford 1981 ;
Brynner, O’Malley and Bachman 1981 ; Dawson 1980; Finian 1988; Gruber 1980;
Kaplan 1980; Muller, Chambliss and Wood 1977; Munson 1992; Richman, Brown
and Clark 1984; Wells and Rankin 1983; Wiggins 1987). Therefore, many
educational scholars have suggested that schools should concentrate on
improving students’ self—esteem as a means of improving all of these other areas.
However, not all studies have found significant correlations between global self-
esteem and these other student outcomes, suggesting that self—esteem
enhancement might not be an appropriate means of obtaining other student
outcomes (e.g.,Brookover, et a1. 1979; Mboya 1986; McCarthy and Hoge 1984;
Mintz and Muller 1977; Wells and Rankin 1983). Additionally, most of these
studies are correlational, so that the directionality of the relationships remains
unclear.

Whether working to enhance students’ self-esteem is an appropriate
method for improving other student outcomes, there is also debate as to how
important it is as a goal in itself. Adolescent self-esteem has been identified as a
critical component in the development of adult self-concept and adjustment to
adult life (Richman and Brown 1985; Purkey and Novak 1984). ”People with

2

3
high general self-esteem typically funcu'on effectively in a variety of situations
and perceive themselves as fulfilled and happy (Richman and Brown 1985)."
Because school-age children, especially adolescents, spend such a great amount
of their time at school and base their self-images on teacher and peer feedback
(Battle 1981 ; Eskilson and Wiley 1987) , many researchers and practitioners
believe that schools need to take an active role in nurturing students’ self-esteem.
In fact, some researchers have suggested that neglech'ng children’s affective
development has contributed to major educational and societal problems (Battle
1981 ; Copley» and Karban 1977), and that ”helping students learn how to be
happy and successful” is the real task of educators (Canfield 1993). However,
while some educators are claiming that teachers tend to ignore children’s
affective needs, and concentrate too much on cognitive skill development (e.g.,
Battle 1981), others feel there is too much emphasis on affective goals in
education. Rita Kramer (1991) has even declared that ”self-esteem has replaced
understanding as the goal of education."

In fact, there seems to be some variability in the emphasis schools and
teachers place on affective goals. Generally, teachers and administrators seem to
believe strongly in enhancing students’ self-concepts. Silverail (1980) found that
most teachers and administrators want students to develop positive self-images
within their schools, with the goal ”have developed a positive self-image"
ranking first out of 66 alternatives among administrators, and fourth among
teachers. However, while this goal rates highly among teachers and
administrators as a group, it is not a priority for all teachers and administrators.
The Schools and Staffing Survey of 1990-91 (NCES 1993) found that in public
schools less than 55% of all secondary teachers and administrators rated
”students’ personal growth” as one of their top three (out of eight) goals.

4

It also remains unclear how much impact individual teachers can have on
student self-esteem. Drury (1980) found that about one-tenth of the total
variation in high school students’ self-esteem lies between schools. However, he
based his analyses on school demographics, rather than using school/ teacher
goals or behaviors. Corbett (1992) hypothesized that junior high and middle
schools would differentially effect students’ self-esteem due to the different
organizational patterns of the two types of schools. However, she found no
differences in students’ self-esteem between these school types. An experiment
by Dusa cited in Canfield (1990) showed that a group of high school students
whose teachers concentrated on enhancing students’ self-esteem were absent less
frequently, completed more homework, participated more in extracurricular
activities, held more offices, and had a higher graduation rate than a control
group of students. However, this study involved a very intensive effort to
improve students’ self-esteem, and did not directly measure self-esteem. Under
normal classroom conditions, without special programs or experiments, teachers
might place less emphasis on achieving affective goals within the classroom,
even though they consider them important. For example, Strein and Murphy
(1982) found that teachers emphasize affective goals much less in actual practice
than they perceive when questioned about their practices by researchers. In fact,
Bonaguro, Rhonehouse and Bonaguro (1988) reported no changes in students’
self-esteem after a six-week course designed specifically to improve students’
self-esteem.

At the high school level, it might be especially difficult for schools to
promote the enhancement of student self-esteem. Generally, students are in
contact with many teachers for only small units of time. It is therefore likely that
significant development of students’ self-esteem will occur only if there is a
consensus among teachers that such efforts should be made. However, high

5
schools have been characterized by researchers as being ”loosely coupled"

organizations, with low goal consensus, more lower-level autonomy and fluid
participation (Herriott 8r Firestone 1984; Weick 1976). This means it is often .
difficult to initiate change, or to control the performances of different members of
the organization, due to the loose connections between the organization’s parts
and the lack of authoritative power (Fuller 1986; Herroitt and Firestone 1984).
Therefore, even if self-estem is one of the school’s priorities, one cannot be sure
that all members of a school’s faculty are in fact trying to enhance their students’
self-esteem or even know how to do so.

Changes in students’ self-concepts are rarely measured by schools or
teachers. It is also not known which types of schools stress these goals.
Therefore, this study will attempt to determine which types of schools stress
affective goals, and if:

H1: Students report higher levels of self-esteem in those schools that
stress affective goals, controlling for previous reports of self-esteem; and

H2: Other student outcomes (achievement test scores, drop—out rates and
college enrollment) are significantly better in schools that emphasize
affective goals.

Background and Demographic Factors Associated with Students’ Self-Esteem
Whether schools can affect students’ self-esteem in general, it seems likely

that schools might best be able to help students who have lower self-esteem due
to their academic or demographic situations within the school. That is, that

aspect of student self—esteem which is lowered due to comparisons with peers or
differential treatment within the school may be most easily reached by in-school
programs. (e. g., Negative effects on students’ self-esteem which result from low-

achievement might be countered by efforts to reduce competition or maximize

6
possibilities for success.) Many studies have shown differing levels of self-

esteem among students due to their academic and behavioral backgrounds, as
well as their racial classifications, SES and gender. Therefore, this study will
attempt to identify those students who might differentially benefit from school
efforts to improve their self-concepts.

Academic and behavior records

Many researchers have identified students’ academic backgrounds and
behavior as causes for differences in students’ self concepts (Alexander and
McDill 1976; Bachman and O’Mally 1986; Hagburg, Masella, Palladino and
Shepardson 1991; Heyns 1974; Kelly 1976; Loney 1974; Oakes and Lipton 1990;
Pottebaum, Keith and Ehly 1986). For example, it has been shown that impulsive
and delinquent children seem to develop low self-esteem due to difficulties
controlling themselves and their environment, and the predominantly negative
reinforcement that they receive for this behavior from parents, teachers and
peers. However, this relationship seems to be weak at the high school level
(Loney 1974; McCarthy and Hoge 1984). Low-achieving children seem to
develop lower self-esteem due to constant academic comparisons made by
teachers and other students within classrooms, low expectations and repeated
failure. However, reference group identity seems to be an important moderator
between the achievement and behavior paformance and students’ self-esteem.
N achmias (1977), for example, showed that lower-track students who use
academic track students as a reference group show a significantly lower level of
self-esteem than those lower-track students who do not compare themselves to

academic-track students. This study will attempt to discover if:

H3: The depressing effects of low-achievement and behavioral
criticism/ punishment on self-esteem are smaller in schools that express
more concern for students’ affective needs.

Gender comparisons

Studies regarding gender differences in self-esteem have been somewhat
inconsistent. Traditional beliefs that women and girls have lower self—esteem
than males have been supported by studies which show 1) that both women and
men perceive to value masculine traits more than feminine traits (Broverman et
al. 1970), 2) that women are less confident in their performances than men
(Deaux and Emswiller 1974; Feather 1969), and 3) and that a greater ”people-
orientation” emerges for girls 12 to 14 years old simultaneously with lower self-
esteem (Rosenberg and Simmons 1975). Studies that actually compare self-
esteem differences by gender have shown varied results, with some producing
significant self-esteem differences by gender, with girls showing lower self-
esteem (e.g., Eskilson and Wiley 1987; Foon 1988;Riclm1an and Brown 1985), and
others showing no differences, or higher female self-esteem (Bledsoe and Dixon
1980; Kohr, et al. 1988). Some of these studies seem to show differences only for
hi gh-SES white girls, or for students in single-sex rather than coeducational
schools. However, results we not consistent.

Several older studies have suggested that girls’ self-esteem is especially
affected by poor academic performance. Skipper (1976), for example, found that
in higher-ability schools, average-ability girls had lower self—esteem than higher-
ability girls, while boys’ self—esteem was not affected by ability. He hypothesized
that girls were more intrapunitive than boys, and so more negatively affected by
poor performance in school. Therefore, one might expect that school attempts to
nurture students’ self-esteem might have greater impact upon female students.

8

However, this effect might be weakened due to girls’ greater dependence on
familial (rather than school-based) approval and support for self-esteem
(Eskilson and Wiley 1987). Therefore, those analyses which test school effects on

self-esteem due to achievement differences will include controls for gender.

5 E S and Race

Higher self-esteem has been consistently identified with higher social class
(Bledsoe and DiXon 1980; Fu, Hinkle and Korslund 1983; Hare 1977; Kohr,
Coldiron, Skiffington, Masters and Blust 1988; Richman and Brown 1985).
However, conditions have been identified in which this relationship does not
hold. For example, Bledsoe and Dixon (1980) found that disadvantaged students
who remain in school and reach higher grade levels might not differ in self-
concept from those students in middle-income families. Additionally, high-SE5
white girls have actually been found to have lower self-esteem than boys or low-
SES girls (Richman and Brown 1985; Simmons, Carlton-Ford and Blyth 1987).
And high levels of pressure to succeed which exist in upper-class suburbs have
been shown to adversely affect the self-esteem of adolescents (Esklison, Wiley,
Muehlbauer and Dodder 1986).

When economic status and achievement are controlled, recent studies
(post—1975) have shown that African American high school students seem to have
the same or higher self-esteem than white students, although the opposite
appears true when no controls are used, or when students at younger grades are
studied (Drury 1980; Hunt and Hunt 1977; Kohr, Coldiron, Skiffington, Masters
and Blust 1988; Richman and Brown 1985). Additionally, African American
students show higher self-esteem in segregated (mostly black or mostly white)
schools, schools in which there are either high or low differences between

African American and white student achievement, and schools with less

9

ability-grouping (Drury 1980; Hunt and Hunt 1977). Similarly, Hunt and Hunt
(1977) reported that African American boys’ self—attitudes were more positive
when they were less attached to school, most likely because their self-images
were distanced from the implications of their school performance. However, if
African American students’ self-concepts are less affected by school factors than
white students’ self-concepts, school efforts to improve students’ self-esteem
might be less effective for African Americans. Therefore, these analyses will
attempt to discern whether:

H4: Affective goal emphasis is associated with increases in self-esteem for

non-minority and low-SE5 students to a greater degree than it is with
minority and high-SE5 students.

METHODS

Ih_eS_am_pl_e
The sample used for this study is taken from the High School and Beyond

Teacher and Administrator Survey of 1984 and the corresponding High School
and Beyond student surveys of 1980 and 1982. Only general education, public
high schools were used in these analyses,l with the final sample consisting of 290
schools, with 8077 student participants and 6380 teacher respondents. The
student-level analyses are based on pooled data from both the sophomore and
senior cohorts in each of the surveyed schools, in order to increase the sample
size of students per school. School-level data are based on the High School and
Beyond school files of 1980 and 1982, aggregation of responses from the
Administrator and Teacher Survey of 1986, and aggregation of student data.

The Variables

School-level demographic variables used in these analyses include the
percentage of students in the school classified as disadvantaged, the mean school
SE52, the percentage of students identified as minorities (African American or
Hispanic), the urban/ rural/ suburban location of the school, the grade levels
taught in the school (coded as elementary +, junior/ senior high and senior high),
the size of the school (measured by the number of students in the sophomore
cohort), the percentage of the class of 1982 who dropped out, and the percentage
of the graduating class of 1982 going to college.

 

1Private schools were not included in these analyses toavoid any selection bias.

2Mean 3133 was entered as a predictor in all or these analyses, but it was determined tint the percentage of
disadvantaged students predicted more of the variance in student outcomes than did mean-SES, so
percentage disadvantaged students was used in place of mean-SES in the final analyses.

10

11

A measure of the schools’ emphases on affective goals was developed
from a question on the Teacher and Administrator survey which asked teachers
to rank eight goals in terms of their importance in their teaching.3 It is an
average of the percentage of teachers ranking affective goals as one of their top
two goals, the percentage ranking affective goals as one of their top three goals,
and the percentage ranking affective goals as one of their top four goals.4 It
therefore represents (in general) the percentage of teachers in the school who rate
students’ affective development as one of their top goals. The mean for this
measure is 42.7, with a standard deviation of 10 and minimal skew (.17).
Interestingly, the average percentage of teachers ranking affective goals as one of
their top three, out of eight goals is 41.6 percent. This mean value is substantially
lower than the 55 percent reported in the Schools and Staffing Survey of 1990-91, 1
suggesting that teachers might be placing a higher priority on these goals now,
than they did in 1984. Variables measuring the percentages of disadvantaged
and minority students, the size of the schools and the emphasis on affective goals
(as an interval variable) were mean-corrected and used to form interaction
variables to allow for interpretation of contextual effects with reduced problems
of multi-collinearity.

Student-level variables include SES (the High School and Beyond
standardized composite), race (an effect—coded dichotomous variable, 1 = African
American or Hispanic, -1 = white), sex (1 = male, -1 = female), tenth-grade and
12th-grade achievement test scores (mean-corrected composite variables

 

3m question asked: “If you had to choose from among the eight gods for students listed below, how
would you rank them according to their importance in your teaching?“ The goal used for construction of
the afTective goals variable was worded “personal growrh and fulfillment (self-esteem, personal efficacy,
self -knowledge).” The other goals listed were: basic literacy skills (reading, math, writing, speaking),
academic excellence or mastery of the subject'matter of the course. Citizenship (understanding institutions
and public views), specific occupational skills, good work habits and selfrdiscipline, human relations skills
(cultural understanding, getting along with others), and moral or religious values.

4Although use of just one of these measures would be more specific, it would also be arbitrary. Combining
the percentages creates a more reliable measure of each school’s emphasis (It affective goals

12

consisting of the average of the reading, vocabulary and math standardized
scores), tenth-grade behavior (a scale focusing on student behavior problems
composed of responses to the questions: 1) I have had disciplinary problems in
school during the last year, 2) I have been suspended or put on probation for
disciplinary reasons, and 3) Every once in a while I cut class), and self-esteem in
the tenth and twelfth grades (composite variables consisting of the items: 1) I
take a positive attitude toward myself, 2) I feel I am a person of worth; on an
equal plane with others, 3) I am able to do things as well as most other people,
and 4) on the whole, I am satisfied with myself). The variable measuring
behavior, however, proved to be unreliable (alpha = .5), and so was eliminated
from the analyses. Hypothesis 3 was therefore tested only for the possibility of
schools moderating the effect of achievement on self-esteem, rather than
moderating the effects of both achievement and behavior on self-esteem.

The Models

Relationships among school characteristics, including affective goal
prioritization and student-level outcomes are examined using hierarchical linear
modeling. This allows for examination of individual outcomes within grouped
units (i.e., schools), without losing the effects of either individual variability or
shared social membership, as one might by using a strictly student- or
school-level analysis. The principle student outcome considered in these analyses
is global self-esteem, however achievement test scores are also examined. Other
student outcomes for which data is available only in aggregated units (school
drop-out rates and the percentage of graduating students going on to college) are
examined through regular regression models.

l3

Hierarchical linear models5 consist of at least two equations, a within-unit
(level-1) model and a between-unit (level-2) model. Here, the units are schools.
The parameters of the within-schools equation (the intercept and slopes) can be
used as the outcome variables of the between-schools models. For example, in
these first analyses, the outcome of interest is student self-esteem. The within

model is as follows:

91.8.1.1.

(12th Grade Self-esteem)ij= 805 + Btj(10th Grade Self-Esteemhj + 82j(SES)ij
,+ B3j(Race)ij + 84j(SES * Race)ij + BSj(Achievement test score)ij +
Boj(Gender)ij + B7j(Gender * Test score)ij + eij ;where

The twelfth-grade self-esteem of student i in school j is the outcome measure;
BOj = the mean student self-esteem score in school j, controlling for Xti - X7i; 1
Bl j through B7j = student-level coefficients for school j; and

eij = the residual error associated with student i in school j.

The between-schools models allow one to predict differences in the
parameters BO through 37 (school mean self-esteem—Bo, and the relationships
between student background characteristics and self-esteem—Bl through 87)
with school-level predictors (e.g., percent disadvantaged students or goal
emphasis). Thus, in the between-schools model one is trying to predict the
coefficients of the within-school models. In the first series of analyses (testing
Hypothesis 1, that students report higher levels of self-esteem in schools which
stress affective goals), the following between schools models are posed:

Level 2
BOj zyoo +701le + +W7W7j + lle
Bij zyio +711W1j + +717W‘rj + Lllj

B7j = “W0 + Y71le + +an7; + u7j ; where

 

5For a complete description of hierarchical linear modeling see Bryk and Raudenbush (1992).

14

W1 j through W7j = the school-level predictors: affective goal ranking(Wl), %
disadvantaged(W2), and % minority(W 3), plus the following interaction
terms: Goals x Disadvantaged(W 4), Goals x Minority(W5), Disadvantaged
x Minority(Wo), Goals x Disadvantaged x Minority(W7);

you = the average6 school mean self-esteem (adjusted for BI through B7),
controlling for W1 through W7;

701 = the average effect of W1 on school mean self-esteem (Bo), controlling for
W2 through W7;

7 10 through y‘70 = the average slopes of predictors X1 through X7 and self-
esteem controlling for W1 through W7 ; and

um through u7j = the residual error associated with school j (i.e., the difference
between the coefficient for the school and the average coefficient for all

schools.

Hypotheses one, three and four predict that schools can effect students’
general self-esteem and reduce discrepancies in self-esteem which are based on
comparisons with others of different academic backgrounds, SES and race. They
are tested by first predicting tenth-grade self-esteem with 3135, race, gender,
achievement and behavior scores at level-1, and affective goals, percentage
disadvantaged students and percentage minority students at level-2. Then,
twelfth-grade self-esteem is predicted, with the addition of tenth-grade self-
esteem as a level-1 predictor.7 Each level-1 slope that is found to be a significant
predictor of twelfth-grade self-esteem, and that is found to vary significantly
between schools, is also predicted with school-level variables, to determine if
emphasis on affective goals might affect the relationship between self-esteem and
other student characteristics (i.e., testing Hypotheses three and four, that
achievement, behavioral, SES and racial differences between students might

 

6The gammas are precision-weighted averages. Their values are weighted according to the amount of
variance between and within schools and the sample size of each school, giving estimates that are between
the unweighted mm, and a mean weighted by sample size only.

7Tenth-grade self esteem is predicted as a means of accwnting for differences in self-esteem which
developed prior to the beginning of the study, and which may or may not be related to school variables.
Tenth-grade scores were entered as a predictor of twelfth-grade scores as a way of controlling for
differences that might have existed prior to any school effects.

15
show less of a relationship with self-esteem in schools which stress affective
goals).

Hypothesis 2, that emphasis on affective goals can impact student
outcomes other than global self-esteem, is tested with the same predictors as
Hypothesis 1, with scores for achievement replacing self-esteem scores.
Additionally, multiple regression analyses are used to examine student outcomes
for which there is only school-level data (i.e., drop-out and college enrollment

rates).

RESULTS

Schools and Affective Goals

Before examining any school effects, a series of analyses were conducted
to determine if affective goal emphasis was more prevalent in particular types of
schools. First, a multiple regression was conducted in which the percentage of
teachers prioritizing affective goals was predicted with the following school
variables: size, percentage minority students, percentage disadvantaged students
and mean SES. No significant relationship was found (see Table 1). Two
AN OVAs were then used to discern relationships between school location
(urban, rural and suburban), grade levels served in the school, and affective goal _
emphasis. There was no significant relationship between school location and
goal emphasis. There was a significant relationship between grade levels served
by the school and school emphasis on affective goals, with an average of 5.6%
more teachers emphasizing affective goals in junior / senior high schools than in
schools serving only students in ninth through twelfth grades (see Table 1).
Perhaps teachers in these schools are more concerned with students’ transitions
into adolescence, and so more concerned with students’ affective needs.

Student Self-Esteem

Hypothesis 1 questions whether schools that emphasize affective goals
have a positive effect on students’ global self-esteem. The first step in analyzing
this effect is to determine how much of the variance in students’ self-esteem is
between schools, and how much is within them. This is accomplished through
unconditional within— and between-schools hierarchical models (i.e., models
without predictors), and is equivalent to a one-way AN OVA with random

16

l7

 

 

 

 

 

Table 1
Prelictors of Affective Goals
Standardized Coeff. Proportion of
Predictor or Group Mean Variance Explained

Multiple Size 0.01 3 .007
Regression 96 Minority -0.018
Mean SES 0.007
96 Disadvantaged -0.065

ANOVAs School location .009
-Urban 41.6%
-Suburban 42.5%
-Rural 44.0%

Grade levels served .026*
-Senior High Only 42.4%
-Jr./ Sr. High 48.1%
-Elementary + 40.5%

 

 

 

 

*p<.05

 

18
effects. This yields the following models:

Level 1 (Within-Schools)
Yij = BOj + 11 j
12th Grade Self-Esteem of student i =
Mean school self esteem + residual for individual i.

Level 2 (ktween-Schools)

M=m+m
Mean school self-esteem of school j =
Grand mean self-esteem + residual for school j.

These analyses show that only 1.5 percent of the variance in tenth and
grade self-esteem scores and 1.7 percent of the variance in twelfth grade self-
esteem scores are between schools. This suggests that schools make little
difference in the average self-esteem of students. However, while schools have
little effect on average self-esteem, they could differentially effect different
groups of students. Such a possibility (as advanced in Hypotheses 3 and 4) can
be tested by examining the slopes of SES, race, gender and achievement with self-
esteem, to see if they vary significantly between schools. Therefore, the next set
of models attempted to discern which student characteristics should be
considered in the final model. Table 2 shows the results of these analyses. SES,
race and gender all significantly predicted sophomore self-esteem (see Table 2,
column 4). However, sophomore self-esteem did not vary significantly between
schools, so it was not analyzed further. Significant predictors of twelfth-grade
self-esteem included tenth-grade self—esteem, SES, race, gender and achievement
test scores. Additionally, the SES- and race-slopes were found to vary
significantly between schools (see Table 2, column 5), so they were included in
the final analysis, not only as controls in the student-level model, but also as
parameters to be predicted. The effects of gender and achievement test scores on
self-esteem were not found to vary significantly between schools. Thus

19

80.5.2... .839: .53: .899.
3 5:238 .8388 cc 85E? $853353 0: co 3:3305 can
3. c8333 .32ch .05. ace—Sum 9: co oucaoEcgm 8855 9.032

 

 

 

 

 

 

 

 

 

 

- .m... .9954 .. .25
6.: «:5 $00.. acoEm>mEo<
.m.: stONOf xwm
- .mé oummsmmm
cringe. .5330... 83.
inc 5. £38.. mwm E08323
88d .2. 313m. .m.m $56 Sop Smdm 390 55
- .m... 208 amok .cmom
.m... trimwor xom
- .m... 83. * mum
a... $.39- £352 scan. :3
.seed 15m poo. stsmmor mum «3.8 880 59
3:356 3523) wmgm .00ch emamamam £028 :Emarl- Emmemmll
85:; Begum couoEmcms -:_5_>> _m>m_-ucov3w m_ 55 oocmta> 23:83
.5523 cc 3 3553mm $894 .30... co 8

 

gofiefio .92.”.53um 5 3.335 E8338

N 03a...

 

20

Hypothesis three, which states that the harmful effects of low-achievement on
students’ self-esteem are reduced in schools which emphasize affective goals, is
disconfirmed.

The final models were specified as follows:

Level 1 (Within-@0018)
(12th grade self esteem)ij= BOj + Blj(SES)ij + sz(Race)aj +
83j(10th grade S.C.)ij + B@(Gender)ij + BSj(Achievement)ij + Cij

Level 2 (Qtween-Schools)
Bo = 700 + you (affective goals)j + 7m (%DISADVANTAGED)j +
Yx3 (%MINOR1TY )j + interactions + my

81 = 710 + y 11 (affective goals» + ya (%DISADVANTAGED)j +
yrs (%MINOR1TY )i + interactions + lllj

32 = yzo + y21 (affective goals)j + Y2: (%DISADVANTAGED)j +
723 (%MINORH‘Y)j + interactions + m;

Bo is the mean twelfth-grade self-esteem, adjusted for SES, race, tenth-grade self-
esteem, gender and achievement. BI and B: are the SES/ self-esteem and
race/ self-esteem slopes. They are predicted with the same school-level variables
as mean self-esteem, to determine if school goals can reduce the effects of these
variables on selfiesteem. 83 through 85 (the slopes on self-esteem for the
variables tenth-grade self-esteem, gender and achievement) are not predicted
with school-level variables as they were not found to vary between schools.

Results of these analyses showed no effects of affective goal emphasis on
mean student self-esteem, or on either of the slopes (SES/ Self-Esteem or
Race/ Self-esteem). Affective goals did not significantly predict either of the
slopes by itself or in any of the interaction terms, so Hypothesis 3, that affective
goal emphasis is especially beneficial for the self-esteem of non-minority and
low-SE8 students, was not supported. Thirty—three percent of the

21

between-school variance in mean twelfth-grade self-esteem was reduced when
predicted by school variables. However, only the percentage of minority
students proved to be significant. Additionally, since only 1.7 percent of the
variance in twelfth-grade self-esteem exists between schools, this accounted for
only about half of a percent of the total variance in self-esteem. Thus, affective
goals seem to have little, if any, relationship with students’ overall self-concept.

These analyses, however, might have missed a non-linear relationship
between affective goals and student self-esteem. This relationship might exist,
for example, if student outcomes could only be affected by a consensus of
teachers stressing affective goals. Therefore, a non-linear analysis was conducted
by transforming the interval-level affective goals variable into two dummy
variables. Schools were divided into three groups according to their emphasis on V
affective goals, and formed into two dummy variables: high affective goal
emphasis and low affective goal emphasis.8 The two dummy variables
representing schools with especially low or high affective goal emphases were
entered into the final models to determine if a non-linear model was better than
the linear model. However, no significant difference was found.

Other student outcomes

Little variance in students’ self-esteem can be accounted for by school
differences. However, it is possible that teachers’ efforts to help students’
affective needs can be demonstrated through outcomes which are more school-
related. Table 3 shows the results of analyses which use twelfth grade

achievement test scores as the outcome variable. Twelfth-grade achievement

 

8m: mean percentage of teachers emphasizing affective goals in schools classified as having low affective
goal emphasis was 28% (ranging from a low of 12% to a high of 33%), while schools classified as having
high affective goal emphasis had an average of 58% (ranging from 53% to 74%) of their teachers
emphasizing affective goals.

22

Table 3

Test Scores Predicted by Student- and School-Level Variables

 

 

 

 

 

 

 

Level 1 Average Estimated 96 of Within-
Student-level Within- Parameter School Variance
Parameters School slope Variance gxplained
SES 3.06**** .864**** 1 5.096
Race -2.53**** 3.1“"
SES * Race -.3 56* 1.08*
Level 2 Significant School- Gamrna and 96 of Between-Schooll
Parameter Level Predictors S'gnificance Variance Explained
Linear 80 (Mean) Affective Goals .030* 33.496
Analysis 96 Disadvantaged -.018*
96 Minority -.014*
Non-Linear 80 (Mean) Low Affective Goa -1.04*** 35.796***
Analysis High Aff. Goals ns.
96 Disadvantaged -.01 7*
96 Minority -.015*

 

 

 

 

*<.OS, **<.01, ***<.OOS, ****<.OO1
(The significance level in the variance explained column represents a significant
increase in variance explained in the non-linear model, compared to the linear model.)

 

23

Low Ave. High
Affective Affective Affective
Goals Goals Goals
0.00 1‘ 4.

 

Mean Standardized Test Scores

 

 

Figure 1. Mean Standardized Test Scores by Affective Goal Emphasis

24

test scores were entered into hierarchical models similar to those used to predict
twelfth-grade self-esteem (i.e., predicted by $138 and race at level 1, and affective
goals, percent disadvantaged students and percent minority students at level 2).
These analyses showed a slight but significant relationship between goals and
test scores, witha ten percent increase in the number of teachers emphasizing
affective goals corresponding to a mean test score increase of about .04 of a
standard deviation (a mean raw score increase of about 0.3). When test scores
were predicted with dummy variables representing schools with low or high
affective goal emphasis, schools that had low affective goal emphasis showed
mean test scores one-tenth of a standard deviation below schools with an average
emphasis on students’ affective needs (see Table 3, row 4; and Figure 1).
Additionally, the dummy-variable model explained significantly more of the
variance in test scores than the model with the interval-level affective goals
variable, suggesting that a non-linear model is better for predicting achievement
test scores with teachers’ goals.

Finally, school-level regression analyses were used to predict outcomes for
which only aggregated (school-level) data was available. Once again the
percentage of disadvantaged students, the percentage of minority students and
the affective goals variable were used as predictors. The outcome variables were
the percentage of the graduating class going to college, and the percentage of the
sohpomore class that dropped out of school prior to graduation. Table 4 shows
the results of these school-level regression equah'ons. No significant relationship
was found between either outcome variable and affective goal emphasis using
the interval-level affective goals variable, although the affective-goals variable
did approach significance on the percentage of students going to college (2 = .07).
However, when dummy variables representing schools with high- or low- goal

emphasis were entered into the equations in place of the interval-level affech've

25

Table 4

PredictorsofDropwtandCollegeEmolnentRates

 

 

 

 

 

 

 

 

 

 

Dependent
Variable Predictors Coefficient R-square
96 of Class of Affective Goals n.s. .10
'82 Sophomores 96 Disadvantaged n.s.
who drgpped out 96 Minority .055*
96 of Class of '82 Low Aff. Goals n.s. .11
Sophomores who High Aff. Goals -Z.67*
dropped out 96 Disadvantaged n.s.
96 Minority 056""
96 of Graduating Affective Goals .20 .07
96 Disadvantaged -.30****
96 Minority .15***
Low Aff. Goals -8.63*** .09*
High Aff. Goals n.s.
96 Disadvantaged -.29****
96 Minority .15”

 

*p<.05, “p<.01 , ***p<.005, ****p<.001
(Significance levels in the R—square column represent a significant

increase in variance explained in the non-linear model,

compared to the linear model.)

 

26
goals variable, affective goals became a significant predictor for both outcomes
(see Table 4, rows 2 and 4). Schools ranking low on affective goal emphasis had
an average of 8.6 percent fewer students going on to college (see Figure 2), while
schools with a high emphasis on affective goals had about a 2.7 percent lower
drop-out rate. Additionally, the non-linear (dummy variable) model predicting
college entrance explained significantly9 more variance than the linear model. In
all, these analyses provide some evidence to support Hypothesis two: that
attempts to meet students’ affective needs improve students’ school-related self-

images or aspirations, even if not affecting global self-esteem.

 

9incremental F-tcsts were performed to determine if the non-linear models were significantly better.

27

Low Ave. High
Affective Affective Affective
Goals Goals Goals

50%
48%
46%
44%
42%
40%
38%
36% --
34% 1-
32% 4-
30%

 

 

.L

Percent of Graduating Class Going to College

Figure 2.
Percent of Graduating Class Going to College by Affective Goal Emphasis

DISCUSSION

These results show that there is almost no difference in students’ global
self-esteem between schools. Such results call into question whether high schools
can have any real effect on students' self-concepts, and whether such goals
should be school priorities. It could be that the diversity of experience inside a
high school (little contact with many different school personnel), and the variety
of school and non-school influences on students’ self-esteem do not allow for
much school influence. However, it should be noted that none of the schools
used in this sample could be described as having a real consensus among
teachers that affective goals are important in their teaching. While schools
ranged from having 12 to 74 percent of their teachers emphasizing affective
goals, there was no school in which 75 percent or more of the teachers
emphasized these goals. Although it seems to be a rare phenomenon, such a
consensus might have an impact which these analyses could not discern.

These results also suggest that measurement of global self-esteem might
not be the most effective means to judge teachers’ efforts to meet students’
affective needs. Such efforts might be judged ineffective because they can have
only little impact on golbal self-esteem, although they may be beneficial in other,
more school-oriented outcomes. For example, a study by Corbett (1992) that
found no differences in students’ self—esteem as a result of different school
organizational patterns might have found significant results if additional
outcome measures were evaluated.

It should also be noted that while Mchers’ emphases on affective goals
showed only a slight beneficial relationship with test scores, drop—out rates and
college enrollment, they did not show any negative effect. This is somewhat

surprising since teachers were asked to rank affective and academic goals against

28

29

each other in the same question. Therefore, concerns that teachers’ emphases on
affective goals are having a deleterious effect on students' achievement appear to
be misplaced. This might be of special concern in poor, urban schools that are
being increasingly forced to rationalize their curriculum“), and promote learning
of the ”basics" at the expense of other (e.g., affective) goals, since these results
also imply that it is beneficial to have at least some concern with students’
affective needs. .

Finally, it is interesting to note that the lack of many teachers emphasizing
affective goals seems to be more highly related to unfavorable outcomes (low
achievement scores, less students going to college), than a consensus is to
favorable outcomes. This could be due to a lack of much consensus within this
sample, but it also implies that a consensus is not necessary to make an impact.
It seems to be more important that at least a moderate number of teachers are
concerned with students’ affective needs.

Future Research

These analyses only involve teachers’ affective goals. The relationship
between these goals and teachers’ other goals (e.g., academic excellence, basic
skills, etc.) are unknown. No distinction is made between teachers and schools
which rate both affective and academic goals highly, and those that stress one at
the expense of the other. Additionally, the reasons various teachers prioritize or
do not prioritize affective goals are unknown. One might speculate that teachers
who are interested in students’ affective needs might also be more committed to
their roles as teachers, or they might be more nurturing individuals. Future
research might analyze the relationship between goals and goal selection and
then look at the effects on student outcomes from a multi-goal perspective.

 

loFor a description of this phenomenon see Apple (1991).

30
Additionally, future research might analyze additional student outcomes
that affective-goals proponents have declared will be improved through
enhancement of student self-esteem (e.g., delinquency, stress, satisfaction in
school). Variables chosen for analysis in this paper were limited by constraints of
the data set and the HLM program used.ll

 

llThe HLM program used for these analyses did not allow for logistic regression, which would have made
possible the analysis of several dichotomous outcomes. However, a more recent version (HLMIZ) does

have this option.

LIST OF REFERENCES

LIST OF REFERENCES

Alexander, KL. and E.L. McDill. 1976. ”Selection and Allocation Within
Schools: Some Causes and Consequences of Curriculum Placement." American
Sociological Review, 41: 963-980.

Apple, Michael W. 1991. ”Conservative Agendas and Progressive Possibilities:
Understanding the Wider Politics of Curriculum and Teaching.” Education and
Urban Society, 23(3): 279-91.

Bachman, J.G. andP. O’Malley. 1986. ”Self-Concepts, Self-Esteem and
Educational Experiences." Journal of Personality and Social Psychology, 50(1): 35-
46.

Battle, James. 1981. ”Enhancing Self—Esteem: A New Challenge to Teachers.”
Academic Therapy, 16: 541-550.

Bledsoe, Joseph C. and Curtis Dixon. 1980. ”Effects of Economic Disadvantage
on Self-Concepts of Urban Black High School Students." Journal of Psychology,
106: 121-127 .

Bonaguro, John, Michael Rhonehouse and Ellen Bonaguro. 1988. ”Effectiveness
of Four School Health Education Projects Upon Substance Use, Self-Esteem and
Adolescent Stress." Health Education Quarterly, 15(1): 81—92.

Brad ford, Ronald W. 1981. ”The Importance of Need and Success in Motivating
Students to Achieve.” NASSP Bulletin, 65(441): 32-36.

Brookover, Wilber, Charles Beady, Patricia Flood, John Schweitzer and Joe
Wisenbaker. 1979. School Social Systems and Student Achievement. New York:
Praeger.

Broverman, S.K., D.M. Broverman, F.E. Clarkson, P.S. Rosenkrantz, and SR.
chel. 1970. ”Sex-role Stereotypes and Clinical Judgments of Mental Health.”
[ou mal of Consulting and Clinical Psychology, 34: 1-7.

Bryk, Anthony S. and Stephen W. Raudenbush. 1992. Hierarchical Linear Models.
Newbury Park, California: Sage.

31

32

Brynner, J. P. O’Malley and]. Bachman. 1981. ” -Esteem and Delinquency
Revisited.” Journal of Yourth and Adolescence, 10(6): 407-441.

Canfield, Jack and Frank Siccone. 1993. 101 Ways to Develop Student Self-Esteem
and Responsibility, Boston: Allyn and Bacon.

Canfield, Jack. 1990. ”Improving Students’ Self-Esteem.” Educational Leadership ,
48(1): 48-50.

Copley, Donna and Ruth Karban. 1977. ”Anonymity...Scourge of Our Time.”
NASSP Bulletin, 61(405): 89-92.

Corbett, Frances Dickens. 1992. A Comparison Study of Student Self-Esteem in 5
Schools with Diflierent Organizational Patterns. Unpublished Doctoral Dissertation, E
North Carolina State University, Raleigh.

Crisci, Pat E. ”Role of Teachers and Administrators in Providing for Positive
Mental Health.” Journal of School Health, 48(9): 551-53.

 

Dawson, Don. 1980. ”Some Mental Health Concerns of School Psychologists.”
School Guidance Worker, 35(6): 19—23.

Deaux, K. and T. Emswiller. 1974. ”Explanations of Successful Performance on
Sex-Linked Tasks: What is Skill for the Male is Luck for the Female.” Journal of
Personality and Social Psychology, 29: 80—85.

Drury, Darrel W. 1980. ”Black Self-Esteem and Desegregated Schools.”
Sociology of Education , 53: 88-103.

Eskilson, Arlene, Mary Glenn Wiley, Gene Muehlbauer and Laura Dodder. 1986.
”Parental Pressure, Self-Esteem and Adolescent Reported Deviance: Bending the
Twig Too Far.” Adolescence, 21(83): 501-515.

Eskilson, Arlene and Mary Glenn Wiley. 1987. ”Peers, Perceived Pressure, and
Adolescent Self-Concept: Is a Daughter a Daughter All of Her Life?” Sociological
Quarterly, 28(1): 135-145.

Feather, NT 1969. ”Attribution of Responsibility and Valence of Success and
Failure in Relation to Initial Confidence and Task Performance.” Journal of
Personality and Social Psychology, 13(2): 129-144.

Finian, Michael]. 1988. ”Predictions of Classroom Stress and Burnout
Experienced by Gifted and Talented Students.” Psychology in the Schools 25(4):
392-405.

33

Foon, Anne E. 1988. ”The Relationship Between Sdrool Type and Adolescent
Self-Esteem, Attribution Styles and Affiliation Needs: Implications For
Educational Outcome.” British Journal of Educational Psychology, 58: 44-54.

Fu, V., D. Hinkle and M. Korslund. ”A Developmental study of ethnic self-
concept among preadolescent girls.” Journal of Genetic Psychology, 142: 67-73.

Fuller, Bruce. 1986. ”Defining School Quality.” Pp. 33—69 in Hannaway, Jane
and Marlaine E. Lockheed (Eds), The contributions of the Social Sciences to
Educational Policy and Practice: 1965-1985. Berkeley, CA: McCutchan.

Gruber, James. 1980. ”Sources of Satisfaction among Students in Postsecondary
Education.” American Journal of Education, 88(3): 320-344.

Hagburg, Winston J., Gaelano Masella, Phillip Palladino and Jerry Shepardson.
1991. ”A F ollow-up Study of High School Students with a History of Grade
Retention.” Psychology in the Schools 28(4): 310-317.

Hare, B. 1977 . ”Racial and Socioeconomic Variations in Preadolescent area-
specific and General Self-Esteem.” International Journal of Intercultural
Relations, 1(3): 31-51.

Herriott, Robert E. and William A. Firestone. 1984. ”Two Images of Schools as
Organizations: A Refinement and Elaboration.” Educational Administration
Quarterly, 20(4): 41-57.

Heyns, B. 1974. ”Social Selection and Stratification Within Schools.” American
Journal of Sociology, 79: 1434-1451.

Hunt, Janet G. and Larry L. Hunt. 1977 . ”Racial Inequality and Self-Image:
Identity Maintenance as Identity Diffusion.” Sociology and Social Research , 61 (4):
539-559.

Lee, Valerie E. and Anthony S. Bryk. 1989. ”A Multilevel Model of the Social
Distribution of High School Achievement.” Sociology of Education, 62(July): 172-
192.

Kaplan, Howard. 1980. Deviant Behavior in Defense of Self} New York: Academic
Press.

Kelly, Delosh. ”T racking and its Impact on Self-Esteem: a Neglected
Dirnmension.” Education, 96: 2-9.

Kohl, Richard L., Robert Coldiron, Eugene Skiffington, James Masters and Ross
Blust. 1988. ”The Influence of Race, Class, and Gender on Self-Esteem for Fifth,

34

Eighth, and Eleventh Grade Students in Pennsylvania Schools.” Journal of Negro
Education 57(4): 467-481.

Kramer, Rita. 1991. Ed School Follies, New York: Free Press.

Loney, Jan. 1974. ”The Relationship Between Impulse Control and Self-Esteem
in School Children.” Psychology in the Schools, 11(4): 462-466.

Mboya, Mzobanzi. 1986. ”Black Adolescents: A Descriptive Study of their Self-
Concepts and Academic Achievement.” Adolescence 21 (83): 689-696.

McCarthy, John and Dean Hoge. 1984. ”The Dynamics of Self-Esteem and
Delinquency.” American Journal of Sociology, 90(2): 396-410.

Merton, Robert K. 1968. Social Theory and Social Structure, New York: The Free
Press.‘

Mintz, Richard and Douglas Muller. 1977 . ”Academic Achievement as a
Function of Specific and Global Measures of Self—Concept.” Journal of Psychology ,
97: 53-57.

Muller, Douglas, Jack Chambliss and Martha Wood. 1977. ”Relationships
Between Area-Specific Measures of Self-Concept, Self—Esteem and Amdemic
Achievement for Junior High School Students.” Perceptual and Motor Skills , 45:
1117-1118. .

Munson, Wayne. 1992. ”Self-Esteem, Vocational Identity and Career Salience in
High School Students.” The Career Development Quarterly 40. 361-368.

Nachmias, Chava. 1977. ”The Issue of Saliency and the Effect of Tracking on
Self-Esteem.” Urban Education, 12(3): 327-344.

National Center for Education Statistics. 1993. Schools and Stafiing in the United
States: A Statistical Profile, 1990-91, Washington DC: National Center for
Education Statistics.

Oakes, Jeannie and Martin Lipton. 1990. Making the Best of Schools, New Haven:
Yale University Press.

Pottebaum, S.M., T. Keith and S. Ehly. 1986. ”Is there a causal relation between
self-concept and academic achievement?” Journal of Educational Research, 79(3):
140-144.

Purkey, W. and J. Novak. 1984. Inviting School Success, Belmont, CA:
Wadsworth Publishing.

35

Richman, Charles and Kathryn Brown. 1985. ”General and Specific Self-Esteem
in Late Adolescent Students: Race x Gender x SES Effects.” Adolescence, 20(79):
555-566.

Richman, Charles, Kathryn Brown and Maxine Clark. ”The Relationship
between Self-Esteem and Maladaptive Behaviors in High School Students.”
Social Behavior and Personality, 12(2): 177-185.

Rosenberg, RR. and RG. Simmons. 1975. ”Sex Differences in the Self-Concept in
Adolescence.” Sex Roles, 1: 147-159.

Silvernail, David L. 1980. ”How Do Administrators, Teachers Rank Educational
Goals?” NASSP Bulletin, 64(440): 101-108.

Simmons, R.G., S. Carlton-Ford and D. Blyth. 1987. ”Predicting how a child will
cope with the transition to junior high school.” Pp. 325-375 in Richard M. Lerner
and T. Foch (Eds) Biological-psychosocial interactions in early adolescence: a lifi—span
perspective, Hillsdale, New Jersey: Erlbaum. ’

Skipper, Charles E. 197 6. ”Personal Characteristics of Adolescents with Average
Intellectual Ability in a High Ability Suburban School District.” The Clearing
House , 50(4): 166-167.

Strein, William and James P. Murphy. 1982. ”Fostering Student Emotional
Growth: Teacher Self-Assessments.” Humanist Educator, 20(3): 124-33.

Weick, Karl E. 1976. ”Educational Organizations as loosely-coupled systems.”
Administrative Science Quarterly, 21(1): 1-19.

Wells, Edward and Joseph Rankin. ”Self-Concept as a Mediating Factor in
Delinquency.” Social Psychology Quarterly, 46(1): 11-22.

Wiggins, James. 1987. ”Self-esteem, earned grades and television viewing habits
of students.” School Counselor 35(2): 128-133.

   

"lllll'llllllllllllllli