Journal of Occupational Psychology (1990), 63, 1-18 Printed in Great Britain 1

0 1990 The British Psychological Society

The measurement and antecedents of

affective, continuance and normative
commitment to the organization

Natalie J. Allen*
Department of Psychology, The University of Western Ontario, London,
Ontario, Canada N6A SC2

John P. Meyer
Centre for Administrative and Information Studies,
The University of Western Ontario

Organizational commitment has been conceptualized and measured in various ways. The
two studies reported here were conducted to test aspects of a three-component model of
commitment which integrates these various conceptualizations. The affective component
of organizational commitment, proposed by the model, refers to employees' emotional
attachment to, identification with, and involvement in, the organization. The conti-
nuance component refers to commitment based on the costs that employees associate with
leaving the organization. Finally, the normative component refers to employees' feelings
of obligation to remain with the organization. In Study 1, scales were developed to
measure these components. Relationships among the components of commitment and
with variables considered their antecedents were examined in Study 2. Results of a
canonical correlation analysis suggested that, as predicted by the model, the affective and
continuance components of organizational commitment are empirically distinguishable
constructs with different correlates. The affective and normative components, although
distinguishable, appear to be somewhat related. The importance of differentiating the
components of commitment, both in research and practice, is discussed.

A great deal of attention has been given recently to the study of commitment to the
organization (Mowday, Porter & Steers, 1982). Like many constructs in organizational
psychology, however, commitment has been conceptualized and measured in various
ways. Common to all the conceptualizations of commitment found in the literature is a
link with turnover; employees who are strongly committed are those who are least likely to
leave the organization. Perhaps more important than this similarity, however, are the
differences between the various conceptualizations of commitment. These differences
involve the psychological state reflected in commitment, the antecedent conditions
leading to its development, and the behaviours (other than remaining) that are expected to
result from commitment.
+ Requests for reprints.
2 Natalie J . Allen and John P . Meyer
Not surprisingly, confusion surrounding the conceptual distinctions is reflected in
attempts to measure the construct. Indeed, relatively little attention has been given to the
development of measures of commitment that conform closely to the researcher’s
particular conceptualization of the commitment construct. Our intention here, therefore,
is threefold: (1) to delineate the distinctions between three of the more common
conceptualizations of ‘attitudinal’ commitment,” (2) to develop measures of each, and
(3) to demonstrate that these measures are differentially linked to variables identified in
the literature as antecedents of commitment. The third aim serves the dual purpose of
providing evidence for the convergent and discriminant validity of the new measures and
of providing a preliminary test of hypotheses concerning the development of
commitment.

The conceptualization a n d measurement of attitudinal commitment

Although several conceptualizations of attitudinal commitment have appeared in the
literature, each reflects one of three general themes: affective attachment, perceived costs
and obligation (Meyer & Allen, 1987 a ) .

Aflective attachment
The most prevalent approach to organizational commitment in the literature is one in
which commitment is considered an affective or emotional attachment to the organization
such that the strongly committed individual identifies with, is involved in, and enjoys
membership in, the organization. This view was taken by Kanter (1968) who described
‘cohesion commitment’ as ‘the attachment of an individual’s fund of affectivity and
emotion to the group’ (p. 507) and by Buchanan (1974) who conceptualized commitment
as a ‘partisan, affective attachment to the goals and values of the organization, to one’s role
in relation to the goals and values, and to the organization for its own sake, apart from its
purely instrumental worth’ (p. 533). The affective attachment approach is perhaps best
represented, however, by the work of Porter and his colleagues (Mowday, Steers & Porter,
1979; Porter, Crampon & Smith, 1976; Porter, Steers, Mowday & Boulian, 1974) who
defined organizational commitment as ‘the relative strength of an individual’s identifica-
tion with and involvement in a particular organization’ (Mowday et a / ., 1979, p. 226).
Porter and his colleagues developed the Organizational Commitment Questionnaire
(OCQ) to measure the commitment construct (Mowday et a / ., 1979). This 15-item scale
has been used extensively in research and has acceptable psychometric properties. A
parallel measure developed in Great Britain for use with blue-collar workers has also been
shown to be ‘psychometrically adequate and stable’ (Cook & Wall, 1980, p. 39).
Although other measures of affective attachment have been developed for use in specific
studies, they typically have not been subjected to rigorous psychometric evaluation.

Perceived costs
For other authors, affect plays a minimal role in the conceptualization of commitment.
* A distinction is made in the commitment literature between attitudinal commitment and behavioural commitment
(Mowday eta/. , 1982; Staw, 1977). The focus of the present research is on attitudinal commitment, conceptualized as a
psychological state that reflects employees‘ relationship to the organization.
 Organizational commitment 3
Instead, commitment is viewed as a tendency to ‘engage in consistent lines of activity’
(Becker, 1960, p.33) based on the individual’s recognition of the ‘costs’(or lost side-bets)
associated with discontinuing the activity (Becker, 1960; Farrell & Rusbult, 198 1;
Rusbult & Farrell, 1983). Kanter (1968), for example, defined ‘cognitive-continuance
commitment’ as that which occurs when there is a ‘profit associated with continued
participation and a “cost” associated with leaving’ (p. 504). For Stebbins (1970),
continuance commitment is the ‘awareness of the impossibility of choosing a different
social identity . . . because of the immense penalties in making the switch’ (p. 527).
Cost-induced commitment has typically been assessed using a measure developed by
Ritzer & Trice (1969), and modified by Hrebiniak & Alutto (1972), that requires
respondents to indicate the likelihood that they will leave the organization given various
inducements to do so (e.g. increases in pay, status, freedom, promotional opportunity). It
is doubtful, however, that this measure actually reflects cost-based commitment (Meyer &
Allen, 1984; Stebbins, 1970). Indeed, the fact that high scores on the scale reflect an
unwillingness to leave the organization, in spite of attractive inducements to do so,
suggests that it may measure affective attachment rather than, or in addition to,
cost-induced commitment (Meyer & Allen, 1984).

Obligation
Finally, a less common but equally viable approach has been to view commitment as a
belief about one’s responsibility to the organization. Wiener (1982, p. 47 1) defined
commitment as the ‘totality of internalized normative pressures to act in a way which
meets organizational goals and interests’, and suggests that individuals exhibit behaviours
solely because ‘they believe it is the “right” and moral thing to do’ (p. 42 1). Although
they do not refer to it as commitment, other authors (e.g. Prestholdt, Lane & Mathews,
1987; Schwartz, 1973; Schwartz & Tessler, 1972) have identified personal norms (defined
as internalized moral obligation) as important contributors to behaviour, including
terminating employment with an organization (Prestholdt et al. , 1987).
The only measure of this obligation-based commitment in‘ the literature is the
three-item scale used by Wiener & Vardi (1980). Respondents are asked the extent to
which they feel ‘a person should be loyal to his organization, should make sacrifices on its
behalf, and should not criticize it’ (Wiener & Vardi, 1980, p. 86, italics added). Other
than internal consistency, the psychometric properties of the scale are not reported.

A three-component conceptualization of organizational commitment

In a model of commitment developed recently by Meyer & Allen (1987 a), the three
approaches outlined above were labelled ‘affective’, ‘continuance’ and ‘normative’
commitment, respectively. Although common to these approaches is a link between the
employee and organization that decreases the likelihood of turnover, it is clear that the
nature of that link differs. Employees with strong affective commitment remain because
they want to, those with strong continuance commitment because they need to, and those
with strong normative commitment because they feel they ought to do so.
Affective, continuance and normative commitment are best viewed as distinguishable
4 Natalie J . Allen and John P . Meyer
components, rather than types, of attitudinal commitment; that is, employees can experience
each of these psychological states to varying degrees. Some employees, for example, might
feel both a strong need and a strong obligation to remain, but no desire to do so; others
might feel neither a need nor obligation but a strong desire, and so on. The ‘net sum’ of a
person’s commitment to the organization, therefore, reflects each of these separable
psychological states.
Given their conceptual differences, it seems reasonable to suggest that each of the three
components of commitment develop somewhat independently of the others as a function
of different antecedents. It has been suggested (Mowday etal., 1982) that the antecedents
of affective attachment to the organization fall into four categories: personal characteris-
tics, job characteristics, work experiences and structural characteristics. As Meyer & Allen
(1987 a ) pointed out, however, the strongest evidence has been provided for work
experience antecedents, most notably those experiences that fulfil employees’ psychologi-
cal needs to feel comfortable within the organization and competent in the work-role.
It is proposed that the continzrance component of organizational commitment will also
develop on the basis of two factors: the magnitude and/or number of investments (or
side-bets) individuals make and a perceived lack of alternatives. These predictions derive
from the theoretical work of Becker (1960) and Farrell & Rusbult (1981; Rusbult &
Farrell, 1983). According to Becker, individuals make side-bets when they take an action
that increases the costs associated with discontinuing another, related, action. Consider,
for example, employees who invest considerable time and energy mastering a job skill that
cannot be transferred easily to other organizations. In essence, they are ‘betting’ that the
time and energy invested will pay off. Winning the bet, however, requires continued
employment in the organization. According to Becker, the likelihood that employees will
stay with the organization will be positively related to the magnitude and number of
side-bets they recognize. *
Like investments, the lack of employment alternatives also increases the perceived costs
associated with leaving the organization (Farrell & Rusbult, 198 1; Rusbult & Farrell,
1983). Therefore, the fewer viable alternatives employees believe are available, the
stronger will be their continuance commitment to their current employer.
Finally, it is proposed that the normative component of organizational commitment will
be influenced by the individual’s experiences both prior to (familial/cultural socialization)
and following (organizational socialization) entry into the organization (Wiener, 1982).
With respect to the former, we might expect, for example, that an employee would have
strong normative commitment to the organization if significant others (e.g. parents) have
been long-term employees of an organization and/or have stressed the importance of
organizational loyalty. With respect to organizational socialization, it is proposed that
those employees who have been led t o believe - via various organizational
practices - that the organization expects their loyalty would be most likely to have strong
normative commitment to it.

* In the past, Becker‘s (1960) side-bet theory has often been discussed in the context of behavroural commitment. Like
Kiesler (1971) and Salancik (1977), Becker defines commitment as the tendency to persist in a course of action. Unlike
Kiesler and Salancik, however, Becker emphasizes the importance of recognizing the cost associated with discontinuing an
action. Recognition of the costs associated with leaving an organization can be viewed as a psychological state reflecting the
employee’s relationship to the organization and is, therefore, included here as a component of attitudinal commitment.
 Organizational commitment 5
Research overview
If the three components of commitment reflect distinct psychological states, it should be
possible to develop independent measures of these states. This was the purpose of Study 1.
Moreover, if the three components of commitment develop as described above, these
measures should correlate with measures of those work experiences predicted to be their
antecedents, but not with those predicted to be antecedent to the other components. This
hypothesis was tested in Study 2. Because there has been less of a research tradition to
guide the identification and measurement of predictors of normative commitment
compared to affective and continuance commitment, the focus of Study 2 was on the latter
two components. Normative commitment was included because, by examining the
pattern of its relations with predicted antecedents of the affective and continuance
components, its relation to these components can be better understood.

Study 1: Development of measures

Method
Subjects and &tu collection procedures. Data were collected from full-time, non-unionized employees in three
organizations: two manufacturing firms and a university. Approximately 500 questionnaires were distributed
to employees in clerical, supervisory and managerial positions at these organizations; of these, 256 (52 per
cent) were completed and returned. Females comprised 57 per cent of this sample. Forty-two per cent were
under 30 years of age; 39 per cent were between 3 0 and 40 years of age; the remaining 19 per cent were above
40 years of age.
Questionnaires were distributed to employees by a member of the personnel department in the participat-
ing organization. Accompanying each questionnaire was a letter explaining the general purpose of the study
i n d a stamped envelope addressed to one of the authors. ParGipation in the study was entirely voluntary.
Measures. A pool of 5 1 items was generated for purposes of scale construction. Some of these items were
modified versions of those used in other scales; others were written by the authors. Each item was worded in
accordance with one of the conceptualizations of commitment described above. Included along with these 5 1
items was the 15-item O C Q (Mowday et al., 1979). W i t h the exception of the O C Q items, which were
presented first, the order of items on the questionnaire was random. Responses to all 66 items were made on
seven-point scales (‘strongly disagree’ to ‘strongly agree’).

Results and discussion

Scale development. Selection of items for inclusion on the final scales was based on a series of
decision rules concerning item endorsement proportions, item-total correlations (with
both keyed and non-keyed scales), direction of keying and content redundancy. Specifi-
cally, items were eliminated if ( a ) the endorsement proportion was greater than .75,*
(6) the item correlated less with its keyed scale than with one or both of the other scales,
and (c) the content of the item was redundant with respect to other items on the scale. An
attempt was made to select both positively and negatively keyed items. Finally, the
number of items selected for each scale was set equal to that for the scale with the
minimum number of items surviving the aforementioned exclusion criteria. Although it

* To calculateendorsement proportions for ratings made on seven-pointscales, responses were divided into three categories:
1, 2 and 3 (strongly, moderately and slightly disagree);4 (neither agree nor disagree);and 5 , 6 and 7 (slightly, moderately,
or strongly agree).
6 Natalie J . Allen and John P. Meyer
was not of primary concern, equality of scale length was considered desirable when it
became apparent that few items would be lost as a result. Following the application of
these rules, eight items were selected for inclusion in each of the Affective Commitment
Scale (ACS), Continuance Commitment Scale (CCS) and Normative Commitment Scale
(NCS). The reliability for each scale (i.e. coefficient alpha) was as follows: ACS, .87; CCS,
.75; NCS, .79. The 24 items comprising these scales were subjected to a factor analysis
(principal factor method). Three factors, accounting for 58.8, 25.8 and 15.4 per cent of
the total variance, respectively, were extracted and rotated to a varimax criterion. The
items and their factor loadings are reported in Table 1. In all cases, the items loaded
highest on the factor representing the appropriate construct. Shown in Table 2 are the
means and standard deviations for the three new commitment scales and the OCQ.
Relationships among the commitment scales. Also shown in Table 2 are the correlations among
the new scales and the OCQ. As can be seen, the CCS is relatively independent of both the
ACS and NCS. The OCQ correlated significantly with the ACS but not with the CCS,

Table 1. Varimax rotated factor matrix based on correlations among the items of the
affective, continuance and normative commitment scales

Factor 1 Factor 2 Factor 3

A f f e i v e Commitment Scale items

1. I would be very happy to spend the rest of my career
with this organization 47
- -07
2. I enjoy discussing my organization with people outside it 10 -07
3. I really feel as if this organization’s problems are my own 39 -06
4. I think that I could easily become as attached to another
organization as I am to this one (R) 21 18
5 . I do not feel like ‘part of the family’ at my organization (R) 15 -04
6. I do not feel ‘emotionally attached’ to this organization (R) 23 03
7. This organization has a great deal of
personal meaning for me 19 02
8. I do not feel a strong sense of belonging to my
organization (R) 82
- 18 -05

Continuance Commitment Scale items

1. I am not afraid ofwhat might happen if I quit my job with-
out having another one lined up (R) - 10 02 39
2. It would be very hard for me to leave my organization right
now, even if I wanted to 22 14 F
3. Too much in my life would be disrupted if I decided I
wanted to leave my organization now 33 27 44
-
 Organizational commitment 7
Table 1. Continued
Factor 1 Factor 2 Factor 3
-

4 . It wouldn’t be too costly for me to leave my

organization now (R) 18 12 46
-

5. Right now, staying with my organization is a matter of

necessity as much as desire - 24 -0 1 53
6 . I feel that I have too few options to consider
leaving this organization - 14 00 67
-

7 . One of the few serious consequences of leaving this

organization would be the scarcity of available alternatives - 17 -07 60
-

8. One of the major reasons I continue to work for this

organization is that leaving would require considerable
personal sacrifice - another organization may not match the
overall benefits I have here 15 -01 so
Nonative Commitment Scale item
1. I think that people these days move from company to
company too often. 14 67
__ -06

2. I do not believe that a person must always be loyal to his or

her organization (R) 29 43
- 00

3. Jumping from organization to organization does not seem at

all unethical to me (R) 07 63 01

4. One of the major reasons I continue to work for this

organization is that I believe that loyalty is important and
therefore feel a sense of moral obligation to remain 17 53 07

5 . If I got another offer for a better job elsewhere I would not

feel it was right to leave my organization 17 9 09
6 . I was taught to believe in the value of remaining loyal to
one organization 15 9 10

7 . Things were better in the days when people stayed with

one organization for most of their careers 05 56 11

8. I do not think that wanting to be a ‘company man’ or

‘company woman’ is sensible anymore (R) 17 47
- -03
% of variance accounted for 58.8 25.8 15.4

Note. The following items were adapted from items used in previous research: ACS items 1 and 3 from Buchanan (1974) and
CCS items 1 and 2 from @inn & Staines (1979). Factor loadings greater tham 0.40 are underlined; decimal points have
been omitted. R = reverse keyed items.
8 Natalie J , Allen and John P . M v e r
Table 2. Means, standard deviations and intercorrelations of commitment measures

Scale x SD ACS ccs NCS OCQ

ACS 4.63 1.33 -
ccs 4.51 1.16 .06 -
NCS 3.77 1.13 .5 1* .14 -
OCQ 5.32 1.07 .83* -.02 . 5 1* -

* P < 0.001. ACS = Affective Commitment Scale; CCS = Continuance Commitment Scale; NCS = Normative
Key.
Commitment Scale; OCQ = Organization Commitment Questionnaire.

thus providing evidence for the convergent validity of the former and for the discriminant
validity of the latter. The correlation between the OCQ and the NCS is consistent with the
correlation between the ACS and NCS noted above.
Taken together, the results of the present study suggest that each of the psychological
states identified in the literature as ‘commitment’ to the organization can be reliably
measured. The independence of CCS scores from scores on the other two measures was
expected. Not expected, however, was the significant relationship between the ASC and
the NCS. This was observed despite efforts to include items that correlated only with
others on the same scale, suggesting that feelings of obligation to maintain membership in
the organization, although not identical to feelings of desire, may be meaningfully linked.
The purpose of Study 2 was to examine the generalizability of these findings and to test the
hypothesis that the three components of commitment would be related to variables
assumed to be their antecedents.

Study 2: Antecedents of commitment

Method
Subjects and d7ta collection procedures. Following the same procedure used in Study 1, data were collected from
full-time, non-unionized employees in three organizations: a retail department store, a hospital and a
university library. In total, 634 questionnaires were distributed to employees in clerical, supervisory,
management and technical positions within these organizations. Of these, 337 (53.2 per cent) were completed
and returned. Females comprised 80.2 per cent of this sample. The mean age of participants was 38 years.
Meawes. The commitment and antecedent measures used were as follows.
Organizational commitment measures. The Affective, Continuance and Normarive Commitment Scales,
developed in Study 1, were included on the questionnaire.
Proposedantecedents of the affective component. Also included on the questionnaire were 11 two-item measures
assessing various work experiences. The work experience variables selected for inclusion were those found in
previous research to correlate with affective commitment. Specifically, we assessed employees’ perceptions of
the extent to which their jobs were challenging (job challenge), roles (role clarity) and goals (goal clarity) were
clearly defined, goals were difficult (goal difficulty), management was receptive to employee suggestions
(management receptiveness), employees were cohesive (peer cohesion), the organization was dependable
(organizational dependability), employees were treated equitably (equity), employees were made to feel that
they were important to the organization (personal importance), feedback concerning their work performance
was provided (feedback), and they were allowed to participate in decisions regarding their own work
(participation). As noted earlier, work experience variables contributing to affective commitment can be
grouped into those that satisfy employees’ needs to feel comfortable in their relationship with the organization
 Organizational commitment 9
and to feel competent in the work-role. Although experiences may contribute to the satisfaction of both needs,
it might be argued that the comfort need would be best served by organizational dependability, management
receptiveness, equity, peer cohesion, role clarity and goal clarity, whereas feelings of competence would be
enhanced most by iob challenge, goal difficulty, personal importance, feedback and participation.
Some of the work experience measures used were modifications of those used in Buchanan’s (1974) study of
managers; others were developed by the authors. Each scale item had a seven-point response format (‘strongly
disagree’ to ‘strongly agree’). Sample items are presented in the Appendix.
Proposed antecedents of the continuance component. Several single-item measures were used to assess the
magnitude of the organization-relevant investments respondents had made. These included questions about
the transferability of both organization-based skills (skills) and formal education (education) to other
organizations, the likelihood that employees would have to move to another geographical area if they were to
leave the organization (relocate), the extent to which employees felt they ‘themselves’ had invested (i.e. time
and energy ‘learning the ropes’) in the organization (self-investment), and the extent to which their pension
fund would be reduced if they left the organization (pension). Responses to these questions were made on
seven-point scales. Also included - as an index of ’investments in the community’ - was the proportion of
employees’ life during which they had resided locally (community). It was predicted that each of these
variables, with the exceptions of skills and education, would be positively correlated with continuance
commitment; the relationships involving skills and education were expected to be negative.
The perceived availability of alternatives was assessed by asking employees to indicate, on a seven-point
response scale, how easy they felt it would be to obtain comparable or better employment in another
organization (alternatives). Scores on this measure were expected to correlate negatively with CCS scores.
Proposed antecedent ofthe normarive component. Although normative commitment was included in this study
primarily for exploratory purposes, one variable, the two-item Organizational Commitment Norm Scale
(Buchanan, 1974), was included as a potential predictor of this component of commitment. Scores on this
scale, which reflect the extent to which employees feel that the Organization expects their loyalty, were expected
to correlate positively with NCS scores.
Data analysis. The relationships among the three commitment measures and those variables hypothesized to
be their antecedents were examined using canonical correlation analysis. The aim of canonical correlation
analysis is to derive a linear combination from each of two sets of variables (commitment measures and
antecedent measures, in this case) in such a way that the correlation between them is maximized. As such, this
analysis is particularly appropriate for the present data. If the affective, continuance and normative
commitment measures represent separable components, we would expect them to define distinct canonical
variates. In addition, the linear combination of antecedent variables that corresponds to a particular
commitment measure should weight most heavily those variables predicted to be antecedent to that
commitment component.

Results and discussion

Relationships among the commitment and antecedent measures. To handle missing data, a
list-wise deletion procedure was used; this reduced the sample to 250 individuals who
responded to every item on the questionnaire. Means, standard deviations, reliabilities
and intercorrelations ofall measures appear in Table 3. * As in Study 1, the reliabilities for
each of the newly developed commitment measures were acceptably high; the same is true
of the two-item antecedent measures. Once again, the correlation between the ACS and
the CCS was negligible (Y = .O l),and that between the ACS and the NCS was significant
(Y = .48, P < .001). Although the relationship between the CCS and the NCS was also
significant (r = .16, P < .Ol), the magnitude of the correlation suggests that the two
scales share little variance. In general, the pattern of correlations between the antecedent

* The factor structureof the 24 commitment items administered in Study 2 closely resembled that ofStudy 1. Although not
included here, it is available from the authors.
able 3. Means, standard deviations, reliabilities and intercorrelations of commitment and antecedent measures

riable 8 SD 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

ACS 4.36 1.38 (86) 01 48 63 53 35 56 48 5 1 61 5 5 68 36 5 1 25 -15 06 13 20 -12 -1

ccs 4.49 1.35 (82) 16 -14 - 1 1 -04 -17 -16 -05 -04 -02 -18 -18 -12 -20 -16 06 06 17 14 -4
NCS 3.80 1.08 (73) 29 39 3 1 25 20 24 38 26 34 21 27 19 - 1 5 -10 00 22 14 -0
Job challenge 4.75 1.86 (89) 42 3 1 78 44 40 48 40 66 37 48 37 - 1 1 06 25 2 1 -20 -0
Role clarity 4.55 1.59 (74) 65 36 58 31 66 48 63 44 48 25 -04 00 -12 10 -12 -0
Goal clarity 4.91 1.56 (76) 25 41 18 52 37 36 32 40 21 02 -02 -06 07 -06 -0
Goal diffy 4.74 1.76 (73) 36 39 32 36 57 48 36 35 -08 06 30 23 -18
Mgmt recept. 3.98 1.75 (78) 44 5 5 53 63 42 59 22 -08 12 00 02 -20 -0
Peer cohesion 4.32 1.52 (44) 35 40 48 31 40 22 -09 06 -07 05 -03 -0
Org. depend’y 4.61 1.75 (73) 5 1 54 37 5 1 18 - 1 1 -05 -12 14 -10 - 1
Equity 3.62 1.49 (61) 52 30 47 18 -06 10 05 05 -16 -1
Pers. import. 4.61 1.79 (78) 46 57 35 -12 07 09 14 -18 -0
Feedback 3.97 1.86 (81) 48 20 01 04 06 06 -08 -0
Participation 4.46 1.60 (61) 25 -10 13 02 10 -17 -0
Skills 5.77 1.59 - -09 14 04 04 -18
Education 5.04 2.07 - - 1 1 -02 -08 -04
Relocate 3.10 2.29 - 14 02 -27 -1
Self-invest. 4.51 1.90 - 10 -07 -0
Pension 3.02 2.32 - 03 -1
Community 0.47 0.32 -
Alternatives 3.78 1.88 -
Commit. norm 4.64 1.43

e. Decimals have been omitted from correlation coefficients.

correlations > .12, p < .05. Reliabilities appear in the diagonal.
 Organizational commitment 11
and commitment measures provides support for the hypothesis outlined above. This was
examined further using canonical correlation analysis.
Antecedents of commitment: Canonical correlation analyis. The results of the canonical corre-
lation analysis are summarized in Table 4. As can be seen, three significant canonical roots
were produced; the correlations associated with these were .8 1; .56 and .38, respectively.
For each of the three canonical variates, both the structure coefficients and standardized
coefficients associated with each variable are shown. The standardized coefficients are the
weights used in the computation of the commitment and antecedent canonical variates
whereas the structure coefficients are the correlations of these linear combinations with
each variable. Because structure coefficients tend to be more stable (Cooley & Lohnes,
197 l), these were used for the purpose of interpretation. Following Pedhazur’s (1982)
recommendation, interpretation was based on variables correlating .30 or greater with the
canonical variate.
Within the commitment set, both affective and normative commitment correlate
strongly with the first canonical variate. A comparison of the two correlations (.98 VJ.
. 5 3) suggests, however, that the first variate represents affective commitment. The fact
that this variate also correlates with normative commitment reflects the overlapping
variance between the two scales. Within the antecedent set, the largest correlations are
associated with those variables hypothesized to be antecedents of affective commitment;
these range from .46 to .87. Only two variables in the antecedent set (skills and
commitment norm) which were not identified as affective antecedents correlated above .30
with the first canonical variate.
The second canonical variate was clearly defined, within the commitment set, by
continuance commitment (r = .99) and, within the antecedent set, by education,
pension and alternatives, all of which were hypothesized to be antecedent to continuance
commitment. The correlation with skills, also predicted as an antecedent of continuance
commitment, fell just below the cut-off (r = -.28).
Finally, the third canonical variate was defined, within the commitment set, by
normative commitment and, within the antecedent set, by goal clarity, role clarity,
relocate and community. Organizational commitment norm, the only variable within the
set which was predicted to be an antecedent of normative commitment, did not correlate
significantly with the third canonical variate.
This pattern of results provides partial support for the predictions made about the three
components of commitment and their antecedents. At this stage in the development of the
three-component view of commitment, it is particularly encouraging that the three
commitment measures defined the three canonical variates as distinctly as they did. The
fact that normative commitment correlated with both the first and third variates suggests
that, although the desire to remain with an organization is not synonymous with the
feeling of obligation to do so, there is a tendency for these feelings to co-occur. It is not
clear at this point whether there is a causal ordering in the development of these two
attitudes or, if so, what that ordering might be. One might speculate that as moral
obligations are internalized to form personal norms, they influence individuals’ feelings
about what they want to do. Alternatively, to justify behaving in accord with their desires,
individuals may come to accept that their actions ate morally right.
With respect to the antecedent variables, the evidence seems strongest for our
12 Natalie J . Allen and John P. Meyer
Table 4. Results of canonical correlation analysis of the relationships of affective,
continuance and normative commitment and their hypothesized antecedents

Canonical Wilks’
root R C R,2 lambda F Rdx

1 .81 .65 .20 8.43** .28 .29

2 .56 .32 .59 3. $9- .ll .06
3 .38 .14 .86 2.25* .03 .06

Canonical variates

1 2 3

Struc. Stand. Struc. Stand. Struc. Stand.

Variables coeff. coeff. coeff. coeff. coeff. coeff.

Commitment variables
Affective 98 93 17 14 - 10 - 65
Continuance - 15 - 19 99 98 -01 - 19
Normative 53 11 26 03 81 116
Antecedent variables
Job challenge 80 19 -08 -05 - 12 -25
Role clarity 69 12 -04 05 32 41
Goal clarity 46 -04 03 14 35 22
Goal difficulty 71 09 - 15 - 17 - 12 -07
Mgmt. receptiveness 62 -11 - 15 -27 - 14 - 24
Peer cohesion 63 20 05 19 - 13 -2 1
Org. dependability 77 31 09 12 14 -02
Equity 67 13 12 12 - 14 -09
Personal importance 87 31 - 12 05 -06 23
Feedback 49 00 -22 -20 10 11
Participation 65 07 -07 -06 01 14
Skills 36 01 -28 - 13 25 37
Education - 16 -02 -32 -23 - 13 - 16
Relocate 04 -01 13 13 -45 -35
Self-investment 14 09 13 19 -25 - 14
Pension 22 02 35 29 24 18
Community - 15 03 22 18 56 48
Alternatives -06 03 -79 -69 19 06
Commitment norm 50 01 -09 -11 11 33

* p < .01; **p < .001.

Note. R, = canonical correlation; Rdr = redundancy of antecedent variables given the canonical variate for commitment
variables; Rdy = redundancy of commitment variables given the canonical variate for the antecedent variables; Struc.
coeff. = structure coefficients: Stand. coeff. = standardized coefficients. Decimals have been omitted from coefficients.
 Organizational commitment 13
predictions about affective commitment. Each of the affective antecedent variables
correlated significantly with the first canonical variate. Thus, employees who felt
comfortable in their roles and who felt competent in the job, expressed greater affective
attachment to the organization.
The relationships between the ‘continuance antecedents’ and the second canonical
variate suggest that, in general, the strength of employees’ need to remain with an
organization is related to their perceptions regarding the availability of alternatives and
the magnitude of particular investments they have made. It will be noted, however, that
these relationships are not as strong as were those for the antecedents of affective
commitment. This may be explained, at least in part, by the fact that, because
respondents were employed at different levels in different organizations, it was necessary
to restrict the assessment of investments to categories that would be relevant to employees
in general (e.g. time and energy). The costs associated with leaving an organization,
however, may be quite specific to an individual (e.g. losing a day-care benefit; moving
from the home one has built) and not easily translated into these more general categories.
It may be possible to account for more variance in continuance commitment using
measures that permit employees to generate (and perhaps weigh the importance 00
specific investments they have made. This would be analogous to the improvements in
prediction obtained in expectancy theory research when outcomes were generated by
subjects rather than the experimenters (Matsui & Ikeda, 1976).
The failure of the organizational commitment norm to correlate with the third
canonical variate (defined by normative commitment) also may be attributed, in part, to
its very general nature. Unfortunately, the diversity (e.g. organization, position, tenure)
within the sample used in this study made it difficult to obtain more specific information
about the various channels (e.g. recruitment, selection, socialization) through which the
organization’s expectations regarding commitment (or norms) are communicated.

General discussion
The purpose of this research was to provide preliminary evidence that the affective,
continuance and normative components of attitudinal commitment are conceptually and
empirically separable. It was argued that each component corresponds closely to one of
three major conceptualizations of commitment found in the literature and represents a
somewhat distinct link between employees and an organization that develops as the result
of different experiences. It was found in Study 1 that the three components of commit-
ment can be measured reliably and that, although there was some overlap between
affective and normative commitment, both were relatively independent of continuance
commitment. The results of Study 2 revealed a pattern of relationships between the
commitment measures, particularly affective and continuance commitment, and the
antecedent variables which was, for the most part, consistent with prediction.
Although the results of Study 2 are generally consistent with the hypothesis that the
components of commitment develop as a function of different work experiences, it was not
our intention at this preliminary stage in the research to address issues of causality. The
dangers of causal inference in research of this nature have been described elsewhere (e.g.
Clegg, 1983;James, Mulaik & Brett, 1982) and need not be elaborated here. The present
results, therefore, should be seen as having set the stage for the kind of longitudinal study
14 Natalie J . Allen and John P. Meyw
of commitment antecedents carried out recently by Bateman & Strasser (1984) and Meyer
& Allen (1987 6, 1988) using the OCQ.
Finally, it is clear from the results of Study 2 that further attention should be given to
the development of the normative component of commitment. The inclusion of a single
general antecedent measure in this study was not intended to suggest that this component
is any less important that the other two. Unfortunately, there is little in the literature
upon which to base predictions regarding the antecedents of normative commitment.
Wiener (1982) suggested that such commitment may develop as the result of socialization
prior to and following entry into the organization, but it is unclear how such influences,
particularly those occurring prior to entry, might be measured. One potential source of
pre-entry socialization is the example set by parents or significant others in an employee’s
life. The impact of this modelled behaviour is extremely complex, however, as illustrated
in the following comment from one of the subjects in Study 2.

I was brought up in a house where we were taught that loyalty was very important to the company you
worked for. My father gave that kind of loyalty [to his employer} for 30 years. He is now on strike pay
receiving $65 per week . . . and really doesn’t know if he will have a job tomorrow. So although I was
taught to believe in the value of loyalty, my attitude has changed.

The focus of the present research was on post-entry socialization. Even here, however,
because of diversity in the sample tested, the measurement was restricted to a very general
assessment of perceived organizational expectations of commitment. It would be useful in
future research to examine the impact of specific socialization practices such as those
identified by Van Maanen & Schein (1979).

ImplicationJ
The research reported here ties together what has, to date, been three separate streams of
commitment research. The use of the term ‘Commitment’ to describe very different
constructs has led to considerable confusion in the literature. Indeed, some may argue that
the term commitment cannot, or should not, be used to describe such distinct constructs.
We retain the term, however, in recognition of the literature on which this research was
based, and because it reflects the common denominator believed to underlie each of the
three conceptualizations. In each case, commitment refers to a psychological state that
binds the individual to the organization (i. e. makes turnover less likely). However, this
may be where the similarity ends. Nonetheless, i t should be recognized that the three
approaches provide valuable insight into the employeeorganization link, and that a more
comprehensive understanding of this link is achieved when all three are considered
simultaneously. By attaching the labels affective commitment, continuance commitment
and normative commitment to the three constructs, both the similarity (i.e. link to
turnover) and differences among them are acknowledged.
Another contribution of this research was the development of reliable measures of the
three components of commitment. With the exception of the OCQ, relatively little
attention has so far been given to the development and psychometric evaluation of
commitment measures. The fact that the Affective and Continuance Commitment Scales
generally correlate as predicted with the proposed antecedent variables provides prelimi-
 Organizational commitment 15
nary evidence that they are valid measures of commitment, as conceptualized here, and
may be useful tools in future research. More evidence is required before the Normative
Commitment Scale can be used with as much confidence. The Affective Commitment
Scale is shorter than the OCQ” and has the advantage that its items were written to assess
only affective orientation towards the organization and not employees’ behaviour or
behavioural intentions (e.g. intention to exert effort or leave the organization). Thus, it
can be used to test hypotheses concerning the consequences of affective commitment
without concern that the relationships obtained merely reflect overlap in the content of the
commitment and behaviour measures. This problem has been identified in studies that
have used the OCQ (e.g. Hom, Katerberg & H u h , 1979). To date, the Continuance
Commitment Scale has no counterpart in the research literature. Although the commit-
ment scale developed by Ritzer & Trice (1969), and revised by Hrebiniak & Alutto
(1972), was purported to measure Becker’s (1960) ‘side-bet’ commitment, this inter-
pretation has been challenged both conceptually (Stebbins, 1970) and empirically (Meyer
& Allen, 1984).
Finally, the conceptual framework provided here suggests an important consideration
for the future study of the consequences of commitment. As noted above, in all three
approaches to organizational commitment, commitment is seen as a negative indicator of
turnover. A logical conclusion to be drawn from this is that one form of commitment is as
useful as another. What is not recognized in such logic, however, is the fact that what
employees do on the job is as important, or more important, than whether they remain.
Dalton and his associates have acknowledged this by making a distinction between
functional and dysfunctional turnover (Dalton, Krackhardt & Porter, 198 1; see also Staw,
1980). It is important to consider, therefore, that the link between commitment and
on-the-job behaviour may vary as a function of the strength of the three components.
There is some evidence that this is the case. Specifically, Meyer, Paunonen, Gellatly,
Goffin &Jackson (1989) found that supervisors’ ratings of the overall job performance and
the promotability of their subordinates correlated positively with those subordinates’
affective commitment scores and negatively with their continuance commitment scores.
Allen & Smith ( 1987) reported a positive relationship between affective commitment and
a self-report measure of employee innovativeness; this measure was negatively correlated
with continuance commitment scores. In the same study, affective and normative
commitment scores, but not continuance commitment scores, were positively related to
self-report measures of employees’ consideration for co-workers and their efficient use of
time.
In future research, it may be possible to identify ‘commitment profiles’ that differen-
tiate employees who are likely to remain with the organizatim and to contribute
positively to its effectiveness from those who are likely to remain but contribute little. If
so, it should be possible for organizations to use the results of research examining
antecedents (e.g. Study 2) to better manage the experiences of their employees so as to
foster the development of the desired profile.

* Although a nine-item version of the OCQ has been used in the literature, Mowday ef a / . (1979)caution that thls version
includes only positively keyed items and may, therefore, increase the tendency towards acqulescent responding.
Furthermore, they point out that some of the negatively keyed items excluded from the nine-item version ‘were more highly
correlated with the total score than several of the positively phrased items’ (p. 244).
 16 Natalie J . Allen and John P . Meyer
Acknowledgements
This research was supported by a grant from the Social Sciences and Humanities Research Council of Canada
(NO. 410-87-1235).

References
Allen, N. J. & Smith, J. (1987). An investigation of ‘extra-role’ behaviours within organizations. Paper
presented at the annual meeting of the Canadian Psychological Association, Vancouver, British Columbia,
June.
Bateman, T. S. & Strasser, S. (1984). A longitudinal analysis of the antecedents of organizational
commitment. Academy of Management Journal, 27, 95-112.
Becker, H. S. (1960). Notes on the concept of commitment. Amwican Journal of Sociology, 66, 32-42.
Buchanan, B. ( 1974). Building organizational commitment: The socialization of managers in work
organizations. Administrative Science Quartwly, 19, 5 33-546.
Clegg, C. (1983). Psychology of employee lateness, absence, and turnover: A methodological critique and an
empirical study. Journal of Applied Psychology, 68, 88-10 1.
Cook, J. & Wall, T. (1980). New work attitude measures of trust, organizational commitment and personal
need non-fulfilment. Journal of Occupational Psychology, 53, 39-52.
Cooley, W. W. & Lohnes, P. R. (1971). Multivariate Data Analysis. New York: Wiley.
Dalton, D. R., Krackhardt, D. M. & Porter, L. W . (1981). Functional turnover: An empirical assessment.
Journal of Applied Psychology, 66, 7 1 6 7 2 1 .
Farrell, D. & Rusbult, C. E. (1981). Exchange variables as predictors of job satisfaction, job commitment,
and turnover: The impact of rewards, costs, alternatives, and investments. Organizational Behavior and
Human Performance, 27, 78-95.
Hom, P. W., Katerberg, R. & Hulin, C. L. (1979). Comparative examination of three approaches to the
prediction of turnover. Journal of Applied Psychology, 64,280-290.
Hrebiniak, L. G. & Alutto, J. A. (1972). Personal and role-related factors in the development of
organizational commitment. Administrative Science Quarterly, 17, 5 55-573.
James, L. R., Mulaik, S. A. & Brett,J. M. (1982). Causal Analysis. Beverly Hills, CA: Sage.
Kanter, R. M. (1968). Commitment and social organization: A study ofcommitment mechanisms in utopian
communities. Amerjcan Sociological Review, 33, 499-5 17.
Kiesler, C. A. (197 1). The Psychology of Commitment. New York: Academic Press.
Matsui, T. & Ikeda, H. (1976). Effectiveness of self-generated outcomes for improving predictions in
expectancy theory research. Organizational Behavior and Human Performunce, 17, 289-298.
Meyer, J. P. & Allen, N. J. (1984). Testing the ‘side-bet theory’ of organizational commitment: Some
methodological considerations. Journal of Applied Psychology, 69, 372-378.
Meyer, J. P. & Allen, N. J. (1987 a). Organizational commitment: Toward a three-component model,
Research Bulletin No. 660. The University of Western Ontario, Department of Psychology, London.
Meyer, J. P. & Allen, N. J. (1987 b). A longitudinal analysis of the early development and consequences of
organizational commitment. Canadian Journal of Behavioural Science, 19, 199-2 15.
Meyer, J. P. &Allen, N. J. (1988). Links between work experiences and organizational commitment during
the first year of employment: A longitudinal analysis. Journal of Occupational Psychology, 61, 195-210.
Meyer, J. P., Paunonen, S. V., Gellatly, I. R., Goffin, R. D. &Jackson, D. N . (1989). Organizational
commitment and job performance: It’s the nature of the commitment that counts. Journal of Applied
Psychology, 74, 152-156.
Mowday, R. T., Porter, L. W . & Steers, R. M. (1982). Employee-Organization Linkages: The Psychology of
Commitment, Absenteeism and Turnouw. New York: Academic Press.
Mowday, R. T., Steers, R. M. & Porter, L. W . (1979). The measurement of organizational commitment.
Journal of Vocational Behavior, 14, 224-247.
Pedhazur, E. J. (1982). Multiple Regression in Behavioral Research, 2nd ed. New York: Holt, Rinehart &
Winston.
Porter, L. W., Crampon, W. J. & Smith, F. J. (1976). Organizational commitment and managerial
turnover. Organizational Behavior and Human Performance, 15, 87-98.
 Organizational commitment 17
Porter, L. W . , Steers, R. M., Mowday, R. T . & Boulian, P. V. (1974). Organizational commitment, job
satisfaction, and turnover among psychiatric technicians. Journal of Applied Psychology, 59, 603-609.
Prestholdr, P. H . , Lane, I. M. & Mathews, R. C. (1987). Nurse turnover as reasoned action: Development of
a process model. Journal of Applied Psychology, 72, 22 1-228.
Quinn, R. P. & Staines, G . L. (1979). The 1977 Quality ofEmployment Survey. Ann Arbor, MI: Institute for
Social Research.
Ritzer, G. &Trice, H . M. (1969). An empirical study of Howard Becker’s side-bet theory. SocialForces, 47,
475-479.
Rusbult, C. E. & Farrell, D. (1983). A longitudinal test of the investment model: The impact on job
satisfaction, job commitment, and turnover of variations in rewards, costs, alternatives, and investments.
Journal of Applied Psychology, 68, 4 2 9 4 3 8 .
Salancik, G. R. (1977). Commitment and the control oforganizational behavior and belief. In B. M. Staw &
G. R. Salancik (Eds), New Directions in Organizational Behavior. Chicago: St Clair Press.
Schwartz, S. H . (1973). Normative explanations ofhelping behavior: A critique, proposal, and empirical test.
Jorrnal of Experimental Social Psychology, 9, 349-364.
Schwartz, S. H . & Tessler, R. C. (1972). A test of a model for reducing measured attitude-behavior
discrepancies. Journal of Personality and Social Psychology, 24, 225-236.
Staw, B . M. (1977). Two sides of commitment. Paper presented at the annual meeting of the Academy of
Management, Orlando, FL, August.
Staw, B. M. (1980). The consequences of turnover. Journal of Occupational Behavior, 1, 253-273.
Stebbins, R. A. (1970). O n misunderstanding the concept of commitment: A theoretical clarification. Social
Forces, 48, 526-529.
Van Maanen, J . & Schein, E. H . (1979). Towards a theory of organizational socialization. In B. M. Staw
(Ed.), Research in Organizational Behavior. Greenwich, CT: JAI Press.
Wiener, Y.(1982). Commitment in organizations: A normative view. Academy of Management Review, 7,
418-428.
Wiener, Y.& Vardi, Y.(1980). Relationships between job, organization, and career commitments and work
outcomes: An integrative approach. Organizational Behavior and Human P w f m n c e , 26, 8 1-96.

Received 12 November 1987; revised version received 12 August I988

Appendix

Sample items: Proposed antecedents of organizational commitment

Antecedents of affective commitment

Job challenge: In general, the work I am given to do at my organization is challenging and exciting
Role clarity: This organization always makes clear what is expected of me
Goal clarity: In my organization, I often find myselfworking on assignments without a clear understanding of
what it is I am supposed to be doing (R)
Goal difficulty: The requirements of my job are not particularly demanding (R)
Management receptiveness: The top management people in my organization pay attention to ideas brought to
them by other employees
Peer cohesion: Among the people in this organization there are few close relationships (R)
Organizational dependability: I feel I can trust this organization to do what it says it will do
Equity: There are people in this organization who are getting much more than they deserve and others who are
getting much less (R)
Personal importance: In this organization you are encouraged to feel that the work you do makes important
contributions to the larger aims of the organization
18 Natalie J . Allen and John P. Meyer
Feedback: I am rarely given feedback concerning my performance on the job (R)
Participation: In my organization, I am allowed to participate in decisions regarding my workload and
performance standards

Antecedents of continuance commitmeni

Skills: To what extent do you think the skills and experiences you have obtained at your current organization
would be useful at other organizations? That is, how many of these skills/experiences would ‘transfer’ from
one organization to another?
Education: My formal education would not be very useful if I was working anywhere but at this or a very
similar organization (R)
Relocate: If you were to leave your organization, do you think you would have to move to a different location?
Self-investment: I have had to invest a great deal of time and effort in this organization (‘learning the ropes’,
etc.)
Pension: If you were to leave your current organization now, would you lose any of the retirement funds you
would have received if you stayed with the organization?
Community: Approximately how long have you resided in the local area? (Response to this item was divided
by the respondent’s age.)
Alternatives: If I were to leave this organization, I would have little difficulty finding a comparable or better
job elsewhere

Antecedent of normative commitment

Organizational commitment norm: Employees in this organization are expected to have a strong sense of
personal commitment to the organization

Note. Reverse keyed items are indicated by (R). With the exception of the community variable, all items had
seven-point response scales.