[This text is machine generated and may contain errors.]

Journal of Personality and Social Psychology
1970, Vol. 14, No. 3, 263-270

INTERNAL-EXTERNAL CONTROL AND COMPETENT
AND INNOVATIVE BEHAVIOR AMONG
NEGRO COLLEGE STUDENTS*

ROSINA C. LAO ?

East Carolina University

A distinction was made between beliefs concerning internal-external control at
the personal and ideological levels. Rationale was given to show that these
should operate differently from each other and relate to different variables.
The following hypotheses were derived: (1) An ointernal? belief in personal
control is positively related to general competence; (2) An oexternal? belief in
ideology which blames the system for Negro disadvantages is positively related
to innovative behavior; (2a) A strong belief that discrimination may be
modified further enhances innovativeness among system blamers. Data on
1,493 male Negro college students in the Deep South supported the two main
hypotheses and also showed that the personal and ideological contro] vari-
ables are independent of each other. The subhypothesis (2a) was rejected.
Implications of the findings and suggestions for future research are discussed.

In recent years numerous studies have
demonstrated the importance of the concept
of internal-external control. Originally grow-
ing out of RotterTs social learning theory
(Rotter, 1954), this concept has now proved
useful in a great variety of problem areas.
The two review articles on internal-external
control (Lefcourt, 1966; Rotter, 1966) have
shown that when a person believes that rein-
forcements are controlled by internal rather
than external forces, he is likely to make
greater attempts at mastering the environ-
ment; to be more resistant to influence
attempts by others, yet more effective in at-
tempts to influence others; to prefer high-
probability choices in risk-taking behavior; to
be lower in anxiety and higher in achieve-
ment orientation; to act more responsively to

probability changes in the situation; to placeT

higher value on skill-determined rewards; and
to be more involved in social action.

Despite this wide range of relationships
with internal-external control, some studies

1 This research is based on a dissertation submitted
to the University of Michigan in partial fulfillment
of the requirements for the PhD degree. The author
wishes to thank her co-chairmen, Patricia Gurin and
Daniel Katz, for their guidance and support through-
out. the research period, the preparation of the thesis,
and this article. Special thanks are due Gerald Gurin
for his sound advice and continuing involvement.

2 Requests for reprints should be sent to Rosina
C. Lao, Department of Psychology, East Carolina
University, Greenville, North Carolina 27834.

have also stressed distinctions in this concept
that would enhance its predictive capacities.
For instance, in applying this concept to chil-
drenTs beliefs about academic reinforcements,
Crandall, Katkovsky, and Crandall (1965)
noted the importance of distinguishing dif-
ferent types of external environmental forces.
In their view, control by other people should
be separated from control by impersonal
forces, since academic success and failure may
have little to do with chance or luck but
still be subject to external control through
teachersT behaviors. They also distinguished
responsibility for causing positive events from
negative outcomes, since the dynamics in
assuming credit for causing good things to
happen may be very different from those
operating in accepting blame for unpleasant
consequences. These kinds of distinctions have
been shown to be helpful in understanding the
way childrenTs beliefs about internal-external
control affect their schoolroom attitudes and
performance,

Two other distinctions have been made by
Gurin, Gurin, Lao, and Beattie (1969) in
studies of Negro youth. One is the difference
between how much control one believes most
people in society possess (Control Ideology)
and how much control one personally pos-
sesses (Personal Control). Although Rotter
(1966) defined internal control as an indi-
vidualTs beliefs that rewards follow from, or

263

264

are contingent upon, his own behavior, the
Internal-External Control of Reinforcement
scale (I-E scale) developed by Rotter and
others contains only a few items that relate
to the personal belief. Most of the items deal
with the individualTs adherence to ideological
beliefs about what determines success for most
people in society. This self-other distinction
is important in the way Negro youth think
about control. Gurin et al. (1969) reported
that two separate factors are generated when
the responses of Negro students to internal-
external control questions are factor analyzed.
One, which consists of items phrased in the
first person, is very close to the conceptual
definition given by Rotter and measures the
belief that one can control what happens to
his own life. The second factor, which consists
of items phrased in the third person, measures
general or ideological beliefs about internal
and external control. It is not merely that
the students make this separation in their
own thinking about control or reinforcements;
it is also that the personal and ideological
belief measures operate very differently in
explaining the studentsT occupational aspira-
tions. Students who have a high sense of
personal control over their own lives hold
higher as well as more realistic aspirations.
In contrast, the studentsT ideological beliefs
about what generally determines success and
failure have nothing to do with their own
aspirations.

Another distinction made in the Gurin study
has to do with the motivational implications
of believing in external forces which are
reality based instead of predictable external
forces such as fate. Most of the studies using
RotterTs I-E scale have assumed it is more
desirable to hold internal beliefs. Since the
external forces alluded to in this scale pri-
marily relate to chance, luck, or fate, it is
understandable that the results generally sup-
port the negative consequences that are
assumed to follow from externality. This does
not mean that these same negative conse-
quences would follow if questions were asked
about systematic and reality-based obstacles.
Instead of being damaging, it may be motiva-
tionally positive for a Negro youth to focus
on discrimination and the way the social

Rosina C. Lao

system structures the outcomes of Negroes in
the society. Results from the Gurin and Katz
(1966) study of Negro youth support this
point of view. Negro college students who fo-
cus on discrimination in explaining the disad-
vantaged position of Negro Americans not
only hold somewhat higher aspirations than
students who rely on internal explanations,
they are also more likely to aspire for pioneer-
ing jobs which were not traditionally held by
Negroes.

This study is concerned with these last
two distinctions and explores in greater depth
the meaning for Negro college males of hold-
ing a strong sense of personal control but
focusing on external forces in explaining
success and failure for Negroes in this society.

Following the literature on the role of
internal control in motivation and behavior,
it is expected that a heightened sense of per-
sonal control among college students will
relate positively to indicators of general com-
petency in the traditional achievement area.
What differs in the specific hypotheses pre-
sented in this paper is the fact that these
relationships are predicted only for the per-
sonal but mot the ideological measures of
internal control.

A different set of behaviors is expected to
follow from the studentsT beliefs about the
role of internal and external forces in explain-
ing the status of Negroes in the society.
Indeed, how students think about the causes
of Negro disadvantage may be quite irrele-
vant for the way they perform and how they
judge their own potential in traditional
achievement situations. Instead, southern
Negro students who actually focus on dis-
crimination as a structural determinant of
NegroesT success and failure show an un-
conventional stance about the race situation.
One would expect this kind of unconven-
tionality to characterize the system-blaming
students in other arenas of life as well,
especially in their efforts to bring about
changes in the social system. They are stu-
dents who should prefer collective instead of
individual betterment strategies for solving
racial disadvantage and who should be more
engaged in social action themselves. In con-
trast, students who focus on a traditional
Protestant Ethic type of explanation in ex-

PERCEIVED Locus oF CONTROL AND COMPETENCE

plaining Negro disadvantage are much more
likely to approach what ought to be done
about the race situation in a_ traditional
manner as well. They are likely to favor
self-improvement approaches and keep them-
selves uninvolved in collective protest and
social action.

It is also possible, however, that the mean-
ing of individual versus system blame may
depend on how much the students feel that
discrimination can be modified. The behavior
of any person who faces a major barrier will
be determined to a great extent by whether
he believes there is a good chance of altering
the obstacles. If he believes the obstacle is
modifiable, he is likely to try to overcome it;
if not, he may work around it. Students who
believe that discrimination is highly modifi-
able are likely to translate their unconven-
tional ideas, as represented by belief in system
blame, into actual innovative *® action to con-
front discrimination directly. Students who
feel that discrimination stands little chance of
being modified, and yet agree that discrimina-
tion is at the root of Negro disadvantage, are
more likely to feel frustrated, desperate, and
give up. Therefore, beliefs about modifiability
of discrimination will greatly influence the
behavior of students who blame the system.
In contrast, these beliefs about modifiability
of discrimination should make little, if any,
difference in the behavior of students who
believe that the major obstacles lie within
themselves and that discrimination has little
to do with NegroesT success and failure.

Thus we arrive at the following specific
hypotheses to be tested. (1) Personal con-
trol is positively related to general compe-
tence of a traditional achievement nature;
(2) Individual-system blame as an ideology
in explaining Negro disadvantages is related
to innovative behavior, with system blamers
likely to be more innovative than individual
blamers; (2a) A strong belief in the modifi-
ability of discrimination further enhances in-
novativeness among system blamers but has
no effect among students who focus on indi-
vidual explanations for Negro disadvantages.

3The word oinnovative? is used throughout this
study to characterize behaviors that introduce some-
thing new and that attempt to make changes.

265

METHOD
Subjects

The data in this article came from a study con-
ducted by the Survey Research Center at the Uni-
versity of Michigan under the direction of Patricia
Gurin and Daniel Katz (1966). The study was car-
ried out in cooperation with 10 Negro colleges in
the Deep South. All the students present on the
day when the questionnaires were administered were
used as subjects. From this total subject pool, data
on 50 males and 50 females from each class level at
each school were randomly selected. Since prior
analyses of the female studentsT expectancy patterns
showed numerous differences from the males, this
article presents data on the male students only.
Among these males, some subjects were further dis-
carded,* leaving an analysis sample of 1,493 male
students.

Procedure

The design of the study was both cross-sectional
and longitudinal. In the cross-sectional study an
extensive set of questionnaires, lasting approximately
3 hours, was given to all students. In the longitudinal
study, freshman students were tested when they first
entered college and again at the end of their fresh-
man year. Data on the studentsT entrance test scores
and grade point averages in the colleges were ob-
tained from the college administrations. The ques-
tionnaire data included a broad range of information
on the studentsT background, college experience, gen-
eral attitudes, future life plans, motivational patterns,
and an anagrams test. Also included was an extended
I-E scale which consisted of all but two items in
the original Rotter I-E scale (excluded because they
had the lowest biserial item correlations as reported
by Rotter in 1966), three items from the Personal
Efficacy scale developed by the Survey Research
Center at the University of Michigan, and 14
Rotter-type items written specifically about the race
situation.

Independent Variables

The major independent variables were taken from
a factor analysis of this extended I-E scale.5 These
are all expectancy variables which measure the indi-
vidualsT beliefs in internal-external control. The first
factor, Personal Control, is composed of five items
all phrased in terms of the first person. It measures
the degree to which an individual feels he has control
over what happens to him. An example is: (a2) When

*Some male subjects were found to have extremely
high scores on the Response Set, Lie, or Defensive-
ness scales included in the questionnaire. Since the
distribution of their scores on the three independent
variables did not differ much from that for the total
male sample, these subjects were excluded from the
final analysis.

5 Results of the factor analysis, including factor
loadings, are presented in Gurin et al. (1969).

266

TABLE 1

COMPOSITION AND NUMBER OF STUDENTS IN
Eacu CELL OF THE FACTORIAL DESIGN

Composition of each cell

No. stu-

Cell dents in

Personal | Individual] Discrimination | each cell

control blame modifiability

i High High High 84
2 High High Low aa
3 High Low* High 45
es High Low Low 39
5 Low High High 58
6 Low High Low 38
7 Low Low High 46
8 Low Low Low 77

Note."Low individual blame = high system blame.

I make plans, I am almost certain that I can make
them work. (0b) It is not always wise to plan too
far ahead because many things turn out to be a
matter of good or bad fortune anyhow.

The second factor, Individual-System Blame, is
composed of four items dealing with ideological
beliefs about how discrimination works. The internal
end attributes failure among Negroes to some internal
lack on the part of Negroes as a group; the external
end attributes it to some systematic obstacles result-
ing from discrimination and segregation. An example
is: (a) Many Negroes who donTt do well in life
do have good training but the opportunities just
always go to whites. (b) Negroes may not have the
same opportunities as whites but many Negroes
havenTt prepared themselves enough to make use of
the opportunities that come their way.

The third factor, Discrimination Modifiability, is
composed of three items measuring the degree to
which the individual believes that racial discrimination
can be modified. The internal end represents a belief
that discrimination can be wiped out; the external
end represents a belief that discrimination cannot be
eliminated. An example is: (a) Racial discrimination
is here to stay. (b) People may be prejudiced but
itTs possible for American society to completely rid
itself of open discrimination.

Dependent Variables

There are two types of dependent variables: one
concerns general competent behavior, the other
concerns innovative behavior.

Measures of competence. Since the sample was
made up of college students, the measures of compe-
tence are drawn from the academic area where
students traditionally show their competence. The
competence measures are divided into the following
three groups: (a) performance measures"entrance
test scores; grade point averages (all transformed
into the 4-point system), and actual scores on an
anagrams test; (b) academic confidence"self-

confidence in the highest grade the student believed
he could get, and relative confidence in how well he

Rosina C. Lao

could do when compared with other students in his
college class; (c) educational expectations and aspi-
rations"how certain the student was of finishing
college, and how certain he was in going on to
graduate or professional school.

Measures of innovativeness. These measures are
drawn from the social areas where innovativeness
is most likely to be manifested. These measures are
also divided into two groups: (a) actual participa-
tion in civil rights activities"frequency of participa-
tion in civil rights activities in the past 2 or 3 years,
and the various degrees of involvement in these
activities; (b) preference for social action strategies"
preference for individual versus collective type of
action in overcoming discrimination, and preference
for negotiation versus protest type of action in
solving racial problems.

RESULTS

Since the aim of this study was to find out
how these expectancy variables work in con-
junction with each other and whether they
interact in explaining competence and inno-
vativeness, a multiple analysis of variance
technique, allowing for unequal cell frequen-
cies, was employed. The interrelationships of
the three independent variables are small
(Personal Control and Individual-System
Blame = .042; Personal Control and Dis-
crimination Modifiability = .124; Individual-
System Blame and Discrimination Modifiabil-
ity = .100); therefore this technique can be
safely used. In order to deal with a reasonable
number of cells, the distributions of all the
independent variables were cut in thirds and
the high and low groups on each variable were
used. Thus the final analysis strategy was a
2 X 2 X 2 factorial design. Both the cell com-
position of this design and the number of
students in each of the eight cells are pre-
sented in Table 1.

Relationship between Personal Control and
Competence

It was hypothesized that students with high
personal control would perform better, have
higher confidence, and hold higher educational
expectations and aspirations. In general, these
hypotheses were supported. The cell means
for measures of competence are presented in
Table 2.

Academic performance results. On all three
measures of academic performance, Personal
Control stands out as the single most sig-
nificant predictor. There was no interaction

PEerceIvep Locus or ContTROL AND COMPETENCE

267


TABLE 2
(~ELL MEANS OF MEASURES OF COMPETENCE
Cell means
High personal control | Low personal control
Measures of competence
Individual System Individual System
High DM | Low DM High DM | Low DM | High DM | Low DM | High"DM'| Low DM
Performance measures
Entrance test scores §:35 5.75 5.38 5.78 3.61 4.63 3.36 4.55
Grade point averages 1.93 2.02 1.80 1.97 1.67 1.77 1.72 1.87
Anagram performance 37.57 40.27 35.54 44.64 31.74 35.91 33.17 35.73
Academic confidence
Self-confidence in own grades 7.05 6.98 6.69 6.85 6.06 6.80 6.26 6.33
Relative confidence in self
vs. others 3.67 3.39 3.62 3.72 3.40 3.47 3.61 3.51
Educational expectations and
aspirations
Certainty of finishing college 4.40 4.30 4.49 4.51 4.24 4.11 4.26 4.13
Certainty of pursuing further
education 4.68 4.64 4.73 4.31 4,12 4.34 4.20 4.12
Note."On all measures of competence, a higher cell mean represents a higher level of competence.
® Individual = high individual blame; system = high system blame; DM = discrimination modifiahility.
effect on any measure. For test scores, stu- indicated a significant first order inter-

dents who had higher Personal Control ob-
tained higher entrance test scores when com-
pared to those who had lower Personal
Control (F = 18.87, df= 1/214, p 01).
The relationship between Personal Control
and grade point averages was similar;
students with higher Personal Control ob-
tained higher grades (F = 4.67, df = 1/309,
p .05). Although the results on anagrams
test performance were slightly different from
the preceding sets of results, the dominant
effect of Personal Control was still apparent,
with the direction still the same (F = 12.42,
df = 1/347, p .001). In addition to Per-
sonal Control, Discrimination Modifiability
was also related to anagrams performance,
although the effect was not as strong as that
of Personal Control (F = 5.74, df = 1/347,
p .05). Those who saw little chance of
modifying the discrimination situation per-
formed better on anagrams.

Academic confidence results. The dominant
influence of Personal Control was also clear
in the studentsT academic self-confidence

(F = 12.83, df = 1/423, p .01). Those who
felt they had more Personal Control were
also more confident they could get higher
grades in the coming year. The results also

action between Personal Control and Indi-
vidual-System Blame (F = 4.42, df = 1/423,
p .05). The cell means in Table 2 suggest
that Personal Control really conditions how
the Individual-System Blame factor operates.
Among students who possessed a higher de-
gree of Personal Control, those who had a
stronger belief in Individual Blame had more
confidence in their grades. On the other hand,
among students who did not have much Per-
sonal Control, there was little difference be-
tween system blamers and individual blamers.
Results on the measure of how the student
felt he would perform in comparison to other
students produced no significant main effect
or interaction effect. Still there was a slight
tendency for those who had a high sense
of Personal Control to feel that they could
do better than others. |
Educational expectations and aspirations.
On the measure dealing with educational
expectations, the results showed that Per-
sonal Control had a significant main effect
(F = 17.30, df = 1/423, p .001). Those
who felt they had greater Personal Control
were also more certain they would finish col-
lege. Discrimination Modifiability also had a
significant main effect, although it was of less

268

Rosina C. Lao

TABLE 3
Crett MrAns oF MEASURES OF INNOVATIVENESS
Cell means
High personal control Low personal control
Measures of innovativeness
Individual® System Individual System
High DM | Low DM | High DM | Low DM | High DM | Low DM High DM | Low DM
Actual participation in civil
rights activities
Frequency of participation 1.41 1.41 1.56 1,54 1.40 1.51 1,52 1,58
Degrees of involvement cae 2.74 3.00 3.00 2.81 2.77 3.36 3.41
Preference for social action
strategies
Individual vs. collective action 1.83 1.79 2.67 2.34 1.96 1.68 2.52 2.35
Negotiation vs. protest action 2.24 2.72 2.91 2.90 2.56 2.51 3.07 3.25

_ _ Note."On measures of innovativeness, a higher cell mean represents a higher degree of participation and involvement in civil
rights activities, and a greater preference for collective, protest action. Soe
* Individual = high individual blame; system = high system blame; DM = discrimination modifiability.

magnitude than that of Personal Control
(F = 4.18, df = 1/423, p .05). The cell
means in Table 2 show that there was actu-
ally not much difference between the four
pairs of cells which differed only on Discrimi-
nation Modifiability. However, it is interesting
to note that unlike previous trends where
perception of low modifiability goes with
higher competence, here it is those who per-
ceived high modifiability who were more con-
fident. Individual-System Blame had no sig-
nificant main effect, but the means show
clearly that certainty of finishing college was
not a linear function of internality, as one
might expect from the two significant main
effects. Results on educational aspiration
dealing with how certain the student felt in
pursuing further education again showed a
simple and clear significant main effect of
Personal Control (F = 9.22; df= 1/423,
p .01. High Personal Control goes with
higher certainty of wanting to go on to gradu-
ate school or some kind of professional school.

Relationship between Individual-System
Blame and Innovativeness

It was hypothesized that students who
blame the system instead of personal inade-
quacies of Negroes in accounting for Negro
disadvantage would be more innovative by
(a) taking a more active part in the civil
rights activities, and (6) taking a social action

stance that differs from the position of pre-
vious generations, especially advocating col-
lective action rather than self-betterment ap-
proaches. These hypotheses were supported.
The cell means on measures of innovativeness
are presented in Table 3.

Participation in civil rights activities. Re-
sults on the frequency of participation showed
a single significant Individual-System Blame
effect (F = 6.66, df = 1/422, p .05). Those
who attributed the problems of Negroes to
discriminatory practices of the system tend
to participate more in civil rights activities.
Results using the various degrees of involve-
ment in civil rights showed the same relation-
ship. Individual-System Blame was the only
significant factor (F=5.18, df = 1/380,
p .05): system blamers had a higher degree
of involvement and commitment in civil
rights activities.

Preference for social action strategies. Indi-
vidual-System Blame was the only decisive
factor in explaining who favored individual
versus collective approaches in the civil rights
movement (F = 56.30, df = 1/418, p .001).
System blamers favored collective action
and individual blamers favored individual
action. On the measure of preference for
protest action or negotiation, the results were
complicated. Still, Individual-System Blame
stood out as the most significant predictor
(F = 41.18, df = 1/423, p .001), system

PERCEIVED Locus OF CONTROL AND COMPETENCE

blamers always preferred protest action,
while individual blamers always preferred
negotiation. Discrimination Modifiability had
the next most significant effect (F = 12.60,
df = 1/423, p .001). Personal Control also
had a small but significant effect (F = 4.41,
df = 1/423, p .05). The significant second-
order interaction (F=4.01, df= 1/423,
p .05) contributed to an understanding of
how each individual main effect works. Table 3
suggests one way we can interpret the find-
ings. When system blamers had high Personal
Control, Discrimination Modifiability bore
no relationship with preference for type of
action; but when system blamers had low
Personal Control, a low degree of Discrimina-
tion Modifiability tended to make them favor
protest type action. When individual blamers
had high Personal Control, believing that
discrimination was not modifiable encouraged
positive attitudes about protest action; in con-
trast, for individual blamers who had low
Personal Control there was no relationship
between Discrimination Modifiability and
type of action preferred.

SUMMARY AND DISCUSSION

Not only did we find clear support
for the expected competent behavior from
students with high Personal Control, we
also found that the ideology measure"
Individual-System Blame"was not related to
competence. Discrimination Modifiability was
related to some competent behaviors, but the
direction was inconsistent. In contrast, Indi-
vidual-System Blame was the only predictor
of innovative behavior in the social action
arena, The other two expectancy variables"
Personal Control and Discrimination Modi-
fiability"bore little or no relationship to how
innovative a student is.

It was also suggested that the relationship
between System Blame and innovativeness
would depend on how much students feel
discimination is modifiable. However, the re-
sults in Table 3 do not support this kind of
interaction. One possible explanation for this
may be due to the nature of the population.
It seems logical that belief in some degree
of discrimination modifiability was necessary
to motivate these students to go to college at
all. This point becomes even more clear when

269

we take into account that these students are
from the Deep South, from very poor families,
and that their parents have had little educa-
tion. In other words, there is a floor effect for
this group of college students with respect to
their perception of the degree of Discrimi-
nation Modifiability. Therefore, within this
rather high expectancy group, the students
who see more difficulty may put more effort
in the fight against discrimination. The inter-
action expected may still work among the
majority of the Negro population where
expectancies may not be as high as in this
college group.

Patterns of Internal-External Control

A belief in internal or external control at
the personal level operates very much as the
generalized expectancy of control has been
presumed to operate in the literature. To have
a sense of internality regarding oneTs own life
situation has been shown to be desirable, and
results reported here indicate that it is also
good for Negro students. It should be empha-
sized again that the ointernals? in past studies
using RotterTs I-E scale are those who believe
in a sense of greater control without distin-
guishing the personal and ideological differ-
ences (indeed since there are more items on
ideological belief than on personal belief in
the I-E scale, they are more likely to be
internal at the ideological level rather than
at the personal level). Yet the results are
compatible to our findings, using only the
Personal Control factor among Negroes. The
reason for this is probably because whites
show a much higher relationship between per-
sonal and ideological beliefs than do Negroes.
For without the same experiences of discrimi-
nation and racial prejudice, whites are less
likely to perceive an inconsistency between
cultural beliefs and what works for them-
selves. Nevertheless the author feels that
sharpening the internal-external control con- .
cept by making this personal-ideological dis-
tinction may enhance its predictive capacities
for all populations.

On the other hand, the rather common as-
sumption in the literature that it is also good
to believe that internal forces are generally
important determinants of success in this
culture is not supported by the present data.

270

Results presented in this paper support and
add meaning to the exception of this common
assumption which was first discussed in the
Gurin study of Negro youth. It seems very
clear that it is mot always desirable for
Negro youth to believe in internal control,
particularly when the sense of control deals
with success and failure for Negroes them-
selves. Instead, Negro students who can focus
on system obstacles seem to have a more real-
istic assessment of the situation, to have a
higher level of sophistication in distinguish-
ing between cultural and personal limitations,
and thus are more likely to choose innovative
roles in the areas of occupation as well as
social action.

The present results are additive, instead of
interactive. Internality in a personal sense
relates to competent behavior in the academic
domain; externality in an ideological sense
relates to innovative behavior in the social
arena. The personal and the ideological vari-
ables are not only independent in a correla-
tional sense (r= .124), but they are also
independent in the sense that neither affects
how the other operates. These two expectancy
variables should be used together if we hope
to increase both competent and innovative
behavior.

Admitting that this is a desirable goal, how
can these expectancies be developed? We
expect the socialization of these two expec-
tancies to differ in many ways. The line of
research by Rotter and his colleagues (James
& Rotter, 1958; Rotter, Liverant, & Crowne,
1961) suggests that more experience with
skill-based reinforcements leads to higher
expectancies. This is also true of the series
of experimental research by Feather (1966,
1968; Feather & Saville, 1967). G. GurinTs
(1968) survey study of a large sample of
high school dropouts undergoing job retrain-
ing also supports this finding. Still, there is no
reason to believe these kinds of success experi-
ences will affect the ideological sense of con-
trol. Instead, educational and training pro-

Rosina C. Lao

grams probably need to deal directly with
the distinction between cultural and personal
limitations to indicate that things are begin-
ning to change and that chances are im-
proving. This is important if positive action
instead of further frustration is to follow.
Future research along this line is certainly
needed.

REFERENCES

CRANDALL, V. C., KatKovsky, W., & CRANDALL, V. J.
ChildrenTs beliefs in their own control of rein-
forcement in intellectual-academic achievement
situations. Child Development, 1965, 36, 91-109.

FEATHER, N. Effects of prior success and failure of
expectations of success and subsequent perform-
ance. Journal of Personality and Social Psychol-
ogy, 1966, 3, 287-298.

FEATHER, N. Changes in confidence following success
and failure and its effect on subsequent perform-
ance. Journal of Personality and Social Psychology,
1968, 9, 38-46.

Gurin, G. Inner-city Negro youth in a job training
project. (Final Rep., Department of Labor and
Department of Health, Education and Welfare)
Washington, D. C.: United States Government
Printing Office, 1968.

Gurin, P., & Katz, D. Motivation and aspiration in
the Negro college. (Final Rep., Office of Educa-
tion, Department of Health, Education and Wel-
fare) Washington, D. C.: United States Govern-
ment Printing Office, 1966.

Gurin, P., Gurin, G., Lao, R., & Beatriz, M. In-
ternal-external control in the motivational dy-
namics of Negro youth. Journal of Social Issues,
1969, 25, 29-53.

James, W. H., & Rotter, J. B. Partial and 100 per-
cent reinforcement under chance and skill condi-
tions. Journal of Experimental Psychology, 1958,
55, 397-403.

Lercourt, H. M. Internal versus external control of
reinforcement: A review. Psychological Bulletin,
1966, 65, 206-220.

Rotter, J. B. Social learning and clinical psychology.
Englewood Cliffs, N. J.: Prentice-Hall, 1954.

Rotter, J. B. Generalized expectancies for internal
versus external control of reinforcement. Psycho-
logical Monographs, 1966, 80(1, Whole No. 609).

Rotter, J. B., Liverant, S., & Crowne, D. P. The
growth and extinction of expectancies in chance
controlled and skilled tasks. Journal of Psychology,
1961, 52, 161-177.

(Received April 15, 1969)