THE SOCIALIZATION OF MANAGERS:
THE EFFECTS OF EXPECTATIONS ON PERFORMANCE*
â– ff- i
David E. Berlew and Douglas T.' Hall
May, 1965 #125-65
*This paper is intended to super-
sede working paper #81-64.
NOT TO BE REPRODUCED IN WHOLE OR IN PART WITHOUT THE AUTHORS' PERMISSION
H^
ACKNOWLEDGEMENTS
The materials which form the basis of this research were
collected by the American Telephone and Telegraph Company as part
of the Bell System's Management Progress Study. In this regard,
we are indebted to Dr. Douglas Bray, Director of the Management
Progress Study, and to W. D. Bachelis, H. W. Clarke, Jr., K. Con-
nors, A. Derks, W. S. Felton, W. H. James, J. P, McKinney, D. B.
Muirhead, W. Katkovsky, and J. F. Rychlak. The planning and exe-
cution of the research are the sole responsibility of the authors,
and the conclusions reached do not necessarily reflect the views
of the Management Progress Study.
This research was supported in part by grants from the
Russell Sage Foundation, the Foundation for Research on Human Be-
havior, and the National Aeronautics and Space Administration,
This paper is concerned with the socialization of new managers by the com-
panies they join. The focus will be on the effect of the company's initial expecta-
tions on the subsequent performance and success of the young manager.
To place our topic in perspective, a simple conceptual model of sociali-
zation focusing on the effects of company expectations on performance will be proposed.
Following that, the results of an empirical investigation designed to test the central
thesis of the model will be reported.
Expectations and Performance
One of the strongest determinants of our behavior is the expectations other
people have of us. In the language of role theory, the behavior of a focal person
is strongly influenced by the expectations of his significant others, sometimes re-
ferrred to as role senders. One of the primary role senders of the new manager is
the company he has just joined; the expectations of the company constitute an impor-
tant class of role forces impinging on him. To an important extent, the behavior of
the new manager will be shaped by the expectations of his associates in the company
(Schein, 1961; 1964).
Of particular interest to us are expectations regarding his contributions
or performance. Role theory would lead us to expect that new managers assigned to
relatively demanding jobs will perform better than those assigned to less demanding
tasks. Direct evidence for this relationship between expectations and performance
is provided by Stedry and Kay (1962) who found improvements in the performance of
managers assigned difficult goals which they perceived as "challenging." Improved
performance was not associated with goals perceived as "easy" or routine , or with
goals perceived as "impossible."* Further support for the expectations-performance
* There is considerable evidence that expectations set so high they are impossible
or even perceived as impossible to reach can be as debilitating to both short-run
and long-run performance as expectations set so low they can be reached without
effort. Overly high company expectations has not been included in the proposed
model as a separate situation; however, the description of what happens when an
individual fails to meet high initial expectations is directly relevant.
2-
relatlonshlp is provided by Berlew and Hall's (1964b) finding that a manager tends
to improve his performance when his contribution falls short of company expectations.
Success versus Failure in Meeting Performance Expectations
When an individual successfully meets performance expectations set for him,
he is usually rewarded - e.g., with approval, a good grade, an increase in salary,
or a promotion. If the performance expectations are reasonably high (i.e., close to
the person's own level of aspiration) he will also feel personal satisfaction at having
achieved his goal. These positive outcomes will generally lead to a higher level of
aspiration , or the Internalization of higher personal standards of performance (Lewin,
1936) and a more positive attitude toward the job or task (Gebhard, 1948). Moreover,
these higher performance standards and positive Job attitudes will generalize to task
activities that have some psychological relation to the initial activity (Lewin, 1936).
Failure to meet performance expectations has quite opposite effects from
success. Failure is not rewarded and is frequently punished. If the performance ex-
pectations are high (i.e., close to level of aspiration), the Individual will gener-
ally lower his personal performance goals or standards (Lewin, 1936) , his level of
performance will tend to drop off (Rao and Russell, 1960), and he will develop nega-
tive attitudes toward the task activity or job.
High versus Low Performance Expectations
Routine performance, even when no more than that is expected, is not gener-
ally rewarded. More important still, none of the results of successfully meeting per-
formance standards described in the preceding section occursunless the task lies close
to the upper limit of achievement. In other words, no feeling of success arises if the
task is below a certain level of difficulty; the feeling of success occurs only if
there is a chance for failure (Lewin, 1936) . Without the feeling of success that re-
sults from meeting high expectations there is no internalization of high performance
standards nor Increase in perceived job attractiveness. Whatever positive effects res-
ult from meeting low performance expectations will be in response to external rewards
which will probably be modest if theymateriallze at all.
When an individual fails to meet low performance expectations, the worst of
3-
all possible situations, there will probably be a strong tendency to project blame
outward in a desperate attempt to preserve some measure of self-esteem (Lewin, 1936) .
Primacy and Company Expectations
Personality psychologists generally have assumed that attitudes or expecta-
tions formed early in life have important implications for later behavior and are more
resistant to change then those which develop later in life. It may be instructive
to draw an analogy between the socialization of a child and a young adult's entry into
a business or industrial organization. The analogy suggests that very early in his
organizational career an individual will develop enduring attitudes and aspirations
which will have important effects on his future behavior. Of particular interest to
us is the early development of performance standards and job attitudes. From the mo-
ment he enters the organization, a new manager is given cues with regard to the
quality of performance that is expected and rewarded. The probability that these ex-
pectations or standards will be internalized is probably higher when the individual
has just joined the organization and is searching for some definition of the reality
of his new environment. If we can carry the analogy of children and new managers one
step further, we would expect that one result of internalizing standards or expecta-
tions at an early stage is that subsequent cues, perhaps different from the initial
ones, will make less of an impression. Stated another way, once an individual has
defined for himself the expectations of his organizational environment and has adjusted
himself to cope with them, he will be insensitive to cues that would require changes
in his attitudes and behavior.
In field theory terms, when the new manager first enters an organization
that portion of his life-space corresponding to the organization is blank. He will
feel a strong need to define this area and develop constructs relating himself to it.
As a new member, he is standing at the boundary of the organization, a very stressful
location (Kahn, 1964) , and he is motivated to reduce this conflict by becoming incor-
porated into the "interior" of the company. Being thus motivated to be accepted by
this new social system and to make sense of the wealth of ambiguity surrounding him.
4-
he is more receptive to cues from his environment than he ever will be again, and
what he learns now will become the core of his organizational Identity. In terms of
Lewln's model of attitude change, the new manager is unfrozen and is searching for
Information and identification models on the basis of which he can change in the di-
rection he feels the organization expects.
The hypothesis submitted to empirical test is derived directly from the prop-
osed model: new managers given initial jobs that are demanding (and therefore challeng-
ing) will in the next several years perform better and be more successful than new
managers given less demanding initial assignments.
METHOD
Subjects
All 62 subjects are college graduates hired originally as management
trainees by two operating companies of the American Telephone and Telegraph Company.
Forty-four subjects, hired in 1956, are presently management level employees of the
Chesapeake and Potomac Telephone Company (C6cF) . An additional 18 subjects are young
managers hired in 1957 by Bell of Pennsylvania (Pa.) .* Most of the C6tP and Pa. sub-
jects joined their company directly from college; a few, however, were employed af-
ter completing a tour in the armed forces or a short period of employment in another
company.
The Independent Variable
The independent variable is company expectations , defined as the type and
quality of contribution expected of the employee by the company. This variable is
intended to reflect the demands upon an employee which result from the specific job
that he holds and from salient physical and social characteristics of his work envi-
ronment. Thus, for example, the company may expect certain things from an employee
because he is functioning as an installation foremen, but the company expects more
than that if it assigns him to work under a difficult supervisor, or as senior tele-
ptone company manager in a small town, or makes him foreman of a crew that has proven
difficult to handle.
A list of 18 categories was empirically formulated to reflect the variety
of expectations that the companies have with regard to managerial employees. These
categories, with short definitions, are presented in Table 1.**
Data regarding company expectations were derived from In-company and Follow-
* The sample size was determined by the number of AT&T Management Progress Study
subjects hired as college graduates by C&P in 1956 and Pa. in 1957 still with
their respective companies in 1962, for whom complete assessment and annual inter-
view data were available.
** A coding manual with more complete definitions, scoring conventions, and scoring
examples is available from the junior author.
Table 1
Categories o£ Company Expectations, and Individual Contributions
1. Technical competence ; the ability to perform non-social, job related tasks
requiring some degree of technical knowledge and skill.
2. Learning capacity ; the ability to learn the various aspects of a position
while on the job.
3. Imagination ; the ability to discover new methods of performing tasks; the
ability to solve novel problems.
4. Persuasiveness ; the ability to present effectively and convincingly a point
of view (may be the point of view of the company j, a department, a
supervisor, or one's self). Persuasiveness should be differenti-
ated from leadership or popularity.
5. Group membership skills ; the ability to work productively with groups of
people. A group may consist of individuals at different levels
in' the company or of individuals from outside the company.
6. Communication skills ; the ability to make well-organized, clear presenta-
tions orally and in writing.
7. Supervisory skill ; The ability to supervise and direct effectively the work
of others.
8. Decision-making ; the ability to make responsible decisions well without
assistance from others.
9. Organizing ability ; the ability to plan and organize the work efforts of
one's self or others.
10. Time-energy commitment ; the expenditure of time and energy for the benefit
of the company.
11. Sacrifice of autonomy : the degree to which the employee accepts company de-
mands that conflict with personal prerogative.
12. Sociability ; the establishment and maintenance of pleasant social relation-
ships with other members of the company off the job.
13. Acceptance of company norms ; conformity to the folkways of the organization
or work group on the job in areas not directly related to job per-
formance.
14. Self -development ; formal or informal education pursued outside of company
time for purposes of increasing person's value to the company.
15. Maintenance of public image ; employee, acting as formal or informal repre-
sentative of the company, conveys a positive image of the company
to outsiders.
16. Loyalty ; accepting company values and goals as one's own; identification
with the company.
17. Productivity ; results; job output; the extent to which employee gets the
job done.
18. Initiative ; drive; self-motivation; the ability to see what should or must
be done and to initiate appropriate activity.
7-
up interview protocols. The In-company interviews were conducted annually with some
member of the company's management (usually a middle-management personnel man) able
to describe the job held by the subject during the preceding year, the personalities
and management styles o£ his superiors, and the organizational and physical context
in which he works.
The Follow-up interview is an annual two-to-three hour interview with the
subject, conducted by a consulting psychologist, exploring such topics as job respon-
sibilities, major sources of satisfaction or dissatisfaction, relationships with peers,
subordinates, and superiors, career aspirations, salary, major occurrences in the past
year, significant aspects of personal life, and health. Frequently, the subject's
description of his job expanded upon that provided by the personnel manager and the
coders' knowledge of positions in the Bell system; however, these data were used for
scoring company expectations only when the coders were reasonably sure that the sub-
ject's comments were objective descriptions of the job environment.
The company's expectations of a subject in a given year were rated from
1 (low) to 3 (high) on each of the 18 categories and a Company Expectations Score
was computed for each year though 1960 for all subjects (yielding four scores for
each Pa. subject and five for each C6eP manager) . Using data for a single year drawn
from the records of 15 subjects, two coders working independently obtained a reliabi-
lity index (Spearmen rank order correlation) of .97 for Company Expectations Scores,
Dependent Variables
The dependent variables can be divided into Performance Criteria and Success
Criteria. The various dependent variables and their intercorrelations are presented
in Table 2.
Success Criteria . (1) 1962 Adjusted Salary: Each Subject was assigned a
score equal to his monthly salary in 1962, corrected (within each company) for differ-
ences in starting salary.
(2) Global Appraisal: This variable reflects the company's estimation of
a subject's overall performance and potential at the end of his fourth year (Pa.) or
Table 2
8-
Intercorrelations of Performance and Success
Measures for C&P Co. (N ~ 44)
Success
Criteria
Success Criteria
1. 1962 Adj. Salary
2. Global Appraisal
3. Success Index
Performance Criteria
4. Average Performance
5. Over Contribution
6. Cumulative Contribution
7. Performance Index
Performance
Criteria
1962
Adj. Sal. GA SI ^
, OC CC PI
1.00 o68 .92 .56 .50 .48 .53
1.00 .92 .53 .60 .60 .65
1.00 .60 .60 .59 .65
1.00 .72 .53 ,68
1,00 .69 .92
1.00 .92
1,00
Note. - All correlations included in Table 2 exceed the 17, level of significance.
Table 3
Intercorrelations of Performance and Success
Measures for Bell of Pa. (N = 18)
Success
Criteria
Success Criteria
1. 1962 Adj. Salary
2. Global Appraisal
3. Success Index
Performance Criteria
4. Average Performance
5. Over Contribution
6. Cumulative Contribution
7. Performance Index
Performance
Criteria
1962
Adj. Sal. GA SI
AP OC CC PI
1.00 ,54 .88 .67 .47 .63 .63
1.00 .88 .76 .61 .58 .70
I. 00 .81 .62 .69 .76
1.00 ,73 ,81 .91
1.00 .85 .93
1.00 .95
1.00
Note. - All correlations included in Table 3 exceed the 17, level of significance.
9-
fifth year (C&P) with the company. On the basis of data included in the fourth and
fifth year In-company interviews, each subject was rated on a scale ranging from 1
(completely unsatisfactory performance) to 10 (outstanding performance) . On a sample
of 23 randomly selected subjects, two coders working independently obtained a reliabi-
lity index (Spearman rank correlation) of .91.
(3) Success Index: This variable is the sum of standardized 1962 Adjusted
Salary and Global Appraisal Scores, with each given equal weight.
Performance Criteria . (4) Average Performance: These scores reflect the
company's overall appraisal of a man's performance for each year averaged across five
years for C6tP subjects and four years for Pa. subjects. Annual In-company interview
data were used to rate each subject on a scale ranging from 1 (completely unsatisfac-
tory performance) to 10 (outstanding performance) . These yearly ratings were then
averaged to obtain an Average Performance Score.
(5) Over Contributions: Each subject's contributions in a given year were
rated using the same 18 categroies that were used to rate Company Expectations (see
Table 1) and the same 1 to 3 rating scale. 'These 18 category scores were then summed
to obtain a single Individual Contributions Score.* Each subject's Individual Contri-
butions Score was then subtracted from his Company Expectations Score for the same
year to obtain an index of whether he exceeded, met or fell short of his company's
expectations of him in that year. Finally, these yearly indices were summed algebra-
ically across four years for Pa. subjects and five years for C6eP subjects to obtain
an Over Contribution Score for each subject.
Annual In-company interview data were used as the primary data source for
these ratings; however, if the Follow-up interviews provided relevant data that were
not contradictory to In-company interview data and coders felts reasonably sure the
data were objective, such data were also considered in arriving at a final contribu-
* Using data for a single year drawn from the records of 15 subjects, two coders work-
ing independently obtained a reliability index (Spearman rank correlation) of .97
for Individual Contribution Scores. (A detailed coding manual is available from
the junior author.)
lo-
tion rating.
(6) Cumulative Contribution: A Cumulative Contribution Score was computed
by summing the Individual Contributions Scores described above across the first four
years for Pa. subjects and across the first five years for C6iP subjects. Each sub-
ject's score reflects his overall contribution to his company in his first several
years of employment.
(7) Performance Index: These scores were computed by summing each subject's
standaaJLzed Average Performance, Over Contributions and Cumulative Contributions Scores,
with each weighted equally.
Dependent Variable Intercorrelations
From the description of the various performance and success criteria it
should be apparent that each is different in terms of either what it measures or the
methoc of measurement.
The relatively high intercorrelations among these variables suggests that,
while each may reflect something unique, they all share a sizable common variance.
Results obtained using each criterion variable separately are reported in the follow-
ing section.
Distribution Characteristics
The means and standard deviations of the variables described above are pres°
ented in Table 4. The distributions were plotted and found to be normal.
Table 4
Distribution Characteristics of the Independent
and Dependent Variables
11-
Variable
C&P
Mean
S. D.
Pao
Mean
D.
Company Expectations
Year 1
Year 2
Year 3
Year 4
Year 5
35.8
36o7
36,6
38.0
38.3
3,22
3.25
3,12
3.65
3,40
36,8
37.6
37,1
38.4
1.92
2,33
1.63
1.38
Success Criteria
1962 Adjusted Salary*
Global Appraisal
Success Index**
237.0
6.80
,040
33,9
1.66
1.85
511,0
7.0
,040
15.0
1.70
1.75
Performance Criteria
Average Performance
Over Contribution
Cumulative Contribution
Performance Index**
3.65
-:"3,77
189,0
.000
1,34
7.93
16,7
1,84
3.7
+ 5.7
155.0
.000
1.10
6,40
8,72
2,79
* Adjusted Salary figures are in units of dollars per month. Because of the ad-
justing operation, the averages have only mathematical meaning; the unadjusted
salary averages are, of course, appreciably higher,
** Ordinanly,indices computed as the sums of N independent standardized variables
would have means of zero and standard deviations of (Tn. The means of the suc-
cess indices are greater than zero because of computational rounding. The
standard deviations are greater than J~^ because the component variables are not
independent.
RESULTS
The hypothesis to be tested is that new managers whose first jobs are
highly demanding will perform better subsequently and be more successful than new
managers initially assigned to less demanding jobs. Data bearing on the relationship
between Company Expectations and performance over several years are presented in
Tables 5 and 6; those bearing on the relationship between Company Expectations and
later success are included in Tables 7 and 8. Data from the two companies are pres-
ented separately.
The results tend to support the hypothesis. The correlations indicate
that C6eP managers whose initial jobs were more demanding performed better over a
four or five year period than those whose initial jobs were less demanding. Simil-
arly, C&P college hires given demanding jobs in their first year with the company
were more successful after several years than those given less demanding jobs.
The same relationships appear to hold for the Pa. Companyj although not
all of the key correlations between first year Company Expectations and the perfor-
mance and success criteria reach statistical signif icarce. However, the magnitude
of the Pa. correlations is of the same order at that for C&P, suggesting that with
a larger sample more of the correlation coefficients would have been statistically
significant.
The correlations between the performance and success criteria and Company
Expectations for years other than the first year are also presented in Tables 5
through 8. What is notable is not that there is a relationship between Company Ex-
pectations and the criteria in years two through five, but that the correlations for
the first year are on the average as high as those for later years.
At least two factors may be contributing to the significant correlations
between first year Company Expectations and subsequent performance and success; (1)
the company somehow managed to assign the best men to the most demanding jobs in the
13-
Table 5
Correlations of Company Expectations, Years 1-5,
with Total Performance Measures
for the C6tP Company (N â– 44)
Average Over Cumulative Performance
Performance Contribution Contributions Index
Company Expec
tations
Year 1
.29*
.29*
.69**
.54**
Year 2
.16
.16
.61**
.42**
Year 3
.26*
.29*
.78**
.58**
Year 4
.14
.26*
.78**
.56**
Year 5
.16
.11
.55**
.36**
*p is less than .05 (one-tailed test)
**p is less than .01 (one-tailed test)
Table 6
Correlations of Company Expectations, Years 1-4,
with Total Performance Measures
for Bell of Pa. (N = 18)
Average
Over
Cumulative
Performance
Performance
Contribution
Contribution
Index
Co rap any
Exp
ectations
Year 1
.24
.12
.46*
.29
Year 2
.18
.00
.39
.21
Year 3
.42*
.09
.46*
.35
Year 4
.58**
.45*
.53*
.56**
*p is less than .05 (one-tailed test)
**p is less than .01 (one-tailed test)
Table 7
14-
Correlations of Company Expectations, Years 1-5,
with Success Measures for the C&P Company (N = 44)
Company Expectations
Year 1
Year 2
Year 3
Year 4