![](https://csdnimg.cn/release/download_crawler_static/19896806/bg1.jpg)
THE PYGMALION EFFECT:
WHAT TEACHER BEHAVIORS MEDIATE
IT’
G.
THOMAS BELLAMY
Uniiiersity
of
Oregon
Current literature suggests that a teacher’s expectancies for
a
pupil may in
part determine the pupil’s subsequent behavior. This review
asks
how such
expectancies affect the teacher’s own behavior, with the goal of determining
what
teacher behaviors mediate the observed effect on the pupil.
A
model of
this
process
is
proposed, using Skinner’s analysis
of
rule-dependent behavior
as
an organizational framework for recent research from several fields. The
model identifies these sets
of
independent variables which warrant further
investigation: (a) variables affecting the process through which
a
teacher
deduces contingency rules from information about a pupil;
(b)
variables
affecting the content of these rules; and (c) variables affecting the teacher’s
performance
of
the rule-specified actions.
The notion that one person’s beliefs
or
expectancies about another may in
part determine the other’s behavior
is
an increasingly popular one. Perhaps the
most widely acknowledged report of this effect was that of Rosenthal and Jacob-
son (1968), who gave teachers false information that they could expect “late bloom-
ing” from 20% of their students; i.e., that selected students were likely to
show
“spurts” in their academic progress. Primarily on the basis of changes in
I&
scores
of the selected students, Rosenthal and Jacobson concluded that their expectancy
instructions were causally related to intellectual gains made by these students.
The teacher’s expectancy for
a
pupil thus seemed to serve as
a
“self-fulfilling proph-
ecy.”
Methodological aspects of the Rosenthal and Jacobson study have been ably
criticized (Jensen, 1969; Thorndike, 1968). However, in spite of some unsuccessful
replications (e.g., Gozali
&
Meyer, 1970; Schwarz
&
Cook, 1972), the concept of
an interpersonal expectancy effect continues to receive empirical support. False
ability information given to teachers has been related to the amount of symbol
learning by Head Start children
(Beea,
1968), improvements
in
academic per-
formance by juvenile delinquents (Meichenbaum, Bowers
&
ROSS, 1969), develop-
ment of self-care skills in retarded children (Anderson
&
Rosenthal, 1968), par-
ticipation in group discussions by high school students (Rothbart, Dalfen
&
Barrett,
1971), and the development of verbal reasoning skills in elementary children (Conn,
Edwards, Rosenthal
&
Crowne, 1968).
The social impact of these studies has been significant, and the use
of
disa-
bility and diagnostic labels in educational settings
is
increasingly under fire (Gozali,
1972; Whelan,1972). Lest this zeal serve only to create
a
new set of labels which
function like those discarded, research into factors which might account for the
effects of interpersonal expectancy seems needed.
Through what process does
a
teacher’s expectancy cause behavior change
in the pupils? Barring extrasensory effects,
a
change in the teacher’s own behavior
seems necessary to mediate this relationship. The purpose of this review
is
(a)
to evaluate existing evidence of effects on the teacher’s own behavior
of
expec-
’This study is supported in part by the Center on Human Development, University of Oregon,
Requests
for
reprints should
be
sent to
G.
Thomas Bellamy, Center of Human Development,
Eugene, Oregon.
University
of
Oregon, 901
E.
18th
St,.,
Eugene, Oregon,
97403.