Content uploaded by Jeremy A Frimer
Author content
All content in this area was uploaded by Jeremy A Frimer on Oct 03, 2015
Content may be subject to copyright.
412
Merrill-Palmer Quarterly, July 2015, Vol. 61, No. 3, pp. 412–439. Copyright © 2015 by
Wayne State University Press, Detroit, MI 48201.
MERRILL-PALMER QUARTERLY, VOL. 61, NO. 3
Developmental Trajectories of Agency and
Communion in Moral Motivation
Lawrence J. Walker and Jeremy A. Frimer University of British Columbia
How does moral motivation develop across the life span? Previous research has
indicated that moral exemplars have integrated the typically oppositional motives
of agency and communion. The present research maps developmental trajecto-
ries in these motives that may lead to this end-point integration. Participants were
140 Canadians comprising four age groups (childhood, adolescence, emerg-
ing adulthood, and mid-adulthood). Agentic and communal motivation was
assessed in an interview that asked participants about aspects of their lives and
prompted for the instrumental–terminal framing of their motives. Results indicated
that agency was the dominant instrumental motive for all ages. In terms of ter-
minal values, agency was the dominant motive early in development; however,
the effect progressively weakened and, by mid-adulthood, had dissipated. The
pattern of instrumental agency for communal goals increased across the age
groups, implying that replacing agency with communion as the characteristic
terminal motive represents an important goal for moral development.
Wherein is the motivation to act morally? How do motives of agency
and communion interrelate in moral functioning? How do they develop?
Answers to such questions have important implications for how we con-
ceptualize morality, understand moral functioning, and attempt to foster its
development. The present research attempts to advance our understanding
of how moral motivation develops.
Lawrence J. Walker and Jeremy A. Frimer, Department of Psychology.
Jeremy A. Frimer is now at the Department of Psychology, University of Winnipeg.
This research was supported by a grant from the Social Sciences and Humanities Research
Council of Canada to Lawrence J. Walker. We acknowledge the competent research assistance
of Kim McWilliam, Brenda Lee, Amanda Riches, Thomas Wiens, and Evguenia Hart, and thank
William L. Dunlop for thoughtful comments.
Address correspondence to Lawrence J. Walker, Department of Psychology, University of
British Columbia, Vancouver, BC, Canada V6T 1Z4. Phone: (604) 822-3006. E-mail: lawrence.
walker@ubc.ca.
MPQ 61.3_05.indd 412 28/05/15 2:43 AM
Trajectories of Agency and Communion 413
The program of research, of which the present study is a part,
represents an investigative process of reverse engineering. Previous
research (reviewed in this article) sought to identify the “finished
product”— developmentally mature moral exemplars—and to assess the
characteristic psychological functioning that undergirds their action.
The present research maps typical developmental trajectories in moral
motivation. Future research will then determine the psychological mech-
anisms that influence these developmental trajectories and end states
(e.g., progressions, stagnations) and their relationship to behavior, as
well as implement and assess intervention efforts in this regard. Insights
from previous research obviously set up assumptions and hypotheses for
subsequent research.
Agency and Communion
Agency and communion comprise the fundamental motivational duality of
human existence (Bakan, 1966; McAdams, 1988; Wiggins, 1991). Agency,
broadly defined, is the motive to individuate and advance the self; com-
munion is the motive to relate to others and contribute to social cohesion.
As will be explicated in this article, these metaconstructs emerge repeat-
edly in disparate literatures (e.g., social judgment, prejudice, the self; for
a review, see Paulhus & Trapnell, 2008). Therein, agency and communion
can have somewhat different meanings. However, a common theme is that
these motives are conceptualized as being in tension. Hogan (1982) framed
them antithetically as getting ahead versus getting along; and Schwartz
(1992) contended that the more people endorse the agentic values of power
and achievement, the less they will typically endorse the communal values
of benevolence and universalism (and vice versa).
Agency and communion have been defined and measured variously,
but the personality-development literature does suggest that these motives,
writ large, evidence age-related patterns (for a review, see McAdams &
Olson, 2010). For example, Ely, Melzi, Hadge, and McCabe (1998) found
that themes of agency predominated over those of communion in the
personal narratives of 4- to 9-year-old children and speculated that, with
development, communion more explicitly becomes part of the essential
self. Consistent with that suggestion, Diehl, Owen, and Youngblade (2004)
examined the implicit expression of agency- and communion-related attri-
butes in adults’ (20–88 years) spontaneous self-representations and found
that older adults expressed more communion attributes than did younger
adults and that younger and middle-aged adults expressed more agency
attributes than did the older adults.
MPQ 61.3_05.indd 413 28/05/15 2:43 AM
414 Merrill-Palmer Quarterly
A great deal of the empirical efforts in the area of personality has been
framed within the five-factor model and involves reliance on standard
personality inventories. Wiggins and Trapnell (1997) argued that agency
and communion map onto the factors of extraversion and agreeableness,
respectively. In that light, the meta-analysis of 92 longitudinal studies of
mean-level personality change in behavioral traits, reported by Roberts,
Walton, and Viechtbauer (2006), is illustrative. They observed that com-
munal traits of agreeableness tend to increase across the life course, as well
as traits of conscientiousness. They also found that agentic traits of social
dominance (a subcategory of extraversion) showed a consistent pattern
of increase from adolescence through mid-adulthood. However, a finer-
grained analysis than is afforded by these broad factors of personality is
necessary to adequately examine developmental trends in agentic and com-
munal motivation. Roccas, Sagiv, Schwartz, and Knafo (2002) reported,
for example, that different facets of extraversion were positively related to
both agency and communion; that communion was also tapped by facets of
agreeableness and openness; and that agency was also tapped by traits of
conscientiousness. Contrary findings were reported by Diehl et al. (2004),
who found that agency was negatively related to conscientiousness and
positively related to openness. All of this to say that the Big-5 factors do
not clearly overlap with the dual motives of agency and communion.
The conception of agency and communion that, in our view, best
addresses the issue of moral motivation focuses on promoting the interests
of self versus promoting the interests of others (Frimer, Walker, Dunlop,
Lee, & Riches, 2011). Schwartz’s (1992) values typology reflects this
distinction well with the self-enhancing values of power and achieve-
ment defining agency and the self-transcending values of benevolence
and universalism defining communion. Schwartz (2006) reported devel-
opmental trends, based on a sample of over 35,000 people (15 years and
older), drawn from the 20-nation European Social Survey. He reported that
agentic self-enhancement values decreased with age, whereas communal
self-transcending values increased with age.
What Motivates Moral Exemplars?
These developmental trends and the common perspective that agency and
communion are antagonistic motives imply that moral exemplarity would
require communion to somehow prevail over agency. But when agency and
communion are in opposition, the motivation to contribute to the greater
good lacks the impetus necessary to move from judgment to action since
there is little personal investment in the enterprise.
MPQ 61.3_05.indd 414 28/05/15 2:43 AM
Trajectories of Agency and Communion 415
To address this issue, Walker and Frimer (2007) undertook an
examination of the motivational profiles of moral exemplars. Their partici-
pants were recipients of a national award for extraordinary volunteerism, as
well as a demographically matched comparison group. Among other find-
ings, these moral exemplars evidenced both more agency and more com-
munion in their life-story interview than did the comparison participants.
Both motives were strongly operative in their psychological functioning.
Perhaps, for exemplars, agency and communion are not in opposition but
instead have, in some way, become synergistically integrated.
Colby and Damon (1992) provided the first suggestive evidence of
such integration by drawing qualitative impressions of the psychological
functioning of a small sample of moral exemplars. Their major conclusion
was that these exemplars did not subjugate their personal interests for the
sake of their prosocial causes; rather, personal and moral concerns seem-
ingly had become fused in their identity.
Frimer et al. (2011) found the first evidence of this agency– communion
integration by using the same sample of adult moral exemplars as Walker and
Frimer (2007) but implementing more precise definitions of agency and com-
munion, microanalytic coding procedures, and person-level analytic strate-
gies to more appropriately examine the issue. These researchers coded goal
motivation as expressed in personal strivings and a life-story interview for
10motivational themes, representing Schwartz’s (1992) typology of universal
values. This typology arrays values around a circular motivational continuum,
and this circumplex structure has been validated in 344 samples from
83countries in different regions of the world (Schwartz et al., 2012). These
motivational values fall into four quadrants, reflecting two bipolar dimensions:
r Agency (power, achievement) versus communion (universalism,
benevolence):1 Agency entails the motivation to advance the self in a
social hierarchy through social power, dominance, material wealth, and
achievement, whereas communion entails the motivation to promote the
interests of others through a concern for the welfare of others in everyday
interactions and through a more universalized concern for others beyond
the primary reference group and for ecological preservation. Agency and
communion, in this conceptualization, explicitly juxtapose promoting the
interests of self and others and thus frame the issue of moral motivation.
r Openness to change (self-direction, stimulation, hedonism2 versus con-
servation (tradition, conformity, security): Openness to change entails
1. Schwartz’s label for the agency quadrant is self-enhancement and, for the communion
quadrant, self-transcendence.
2. Hedonism straddles the boundary between self-enhancement and openness to change.
MPQ 61.3_05.indd 415 28/05/15 2:43 AM
416 Merrill-Palmer Quarterly
motivation for new ideas, actions, and experiences, whereas conserva-
tion entails motivation for self-restriction, order, and the status quo.
These two quadrants entail values that are less relevant to moral moti-
vation as conceptualized here.
A variety of models of agency and communion are extant (Paulhus &
Trapnell, 2008) but, conceptually, the foregoing definitions of agency as
the self-enhancing values of power and achievement and of communion
as the self-transcending values of benevolence and universalism best cap-
ture the oppositional dualism that, we posit, may be reconciled in moral
maturity.
Frimer and colleagues (2011) found that moral exemplars tended to
coordinate agentic and communal themes within their personal strivings
and in the flow of their life stories, more so than comparison participants
and even when controlling for baseline levels of motivation. Agency and
communion may function dualistically for most people, but exemplars
have overcome the tension by integrating their personal ambitions with
their moral concerns. This suggests that the end-point goal for moral moti-
vation is the integration of agency and communion in a form of enlightened
self-interest. Morality can and should be self-regarding if it is to have moti-
vational oomph (Walker, 2013).
The Structure of Agency–Communion Integration
The evidence, thus, is that agentic and communal motivation frequently
co-occur in the functioning of moral exemplars, but how are these motives
integrated? In particular, what is the direction of the integrated relationship
between agency and communion? Two possibilities, which carry vastly dif-
ferent moral weight, exist: One possibility would be the reliance on agentic
motives to advance communal ends (e.g., “I desire to use my social standing
to help others”); the other would be the use of communal motives to pur-
sue agentic ends (e.g., “I want to help the poor so as to enhance my social
status”). Rokeach (1973) first articulated this notion of instrumental and
terminal value motivation. Instrumental motivation represents a means to
some other end, whereas terminal motivation represents an ultimate value,
an end in itself. Thus, agentic and communal motivation can be understood
within an instrumental–terminal framework.
To investigate the direction of the instrumentality between agentic and
communal motives, Frimer, Walker, Lee, Riches, and Dunlop (2012) exam-
ined the moral motivation of a different sample of exemplars: influential
figures of the past century as identified by Time magazine. Expert raters
judged the moral character of this set of target subjects. The top-ranking
MPQ 61.3_05.indd 416 28/05/15 2:43 AM
Trajectories of Agency and Communion 417
figures were classified as moral exemplars (including several Nobel Peace
Prize laureates); the bottom-ranking figures comprised a comparison group
of similar renown and familiarity but who were not noted for their moral
qualities. These historical figures were not available for direct research par-
ticipation, so these researchers content-analyzed archival materials (their
speeches and interviews) for agency and communion in an instrumental–
terminal framework.
The comparison subjects evidenced strong agentic motivation at both
the instrumental and the terminal levels: Agency was the means to attain-
ing more power and achievement. They ubiquitously advanced notions of
agency. The exemplars similarly evidenced strong agentic motivation at
the instrumental level; after all, these were incredibly influential figures.
However, in sharp contrast to the comparison group, exemplars’ terminal
motivation was predominantly communal. Thus, exemplars’ moral motiva-
tion was hierarchically integrated, with agency instrumental to terminal
communal ends. These findings reinforce the importance of distinguishing
between instrumental and terminal aspects of motivation and, further, sug-
gest that replacing agency with communion as the characteristic terminal
motive represents a goal for moral development.
How Do Agency and Communion Develop?
Instrumental motives seem to be predominantly agentic (for both exem-
plar and comparison participants). This represents the inherent motiva-
tional structure of goal-directed activity. Goals require motives of power
and achievement for their attainment. Terminal motives, in contrast, may
be more variable and have a stronger relation with moral behavior. But do
instrumental and terminal motives vary across developmental levels?
Dunlop, Walker, and Matsuba (2013) reported the first study to com-
pare the motivational profiles of exemplars at different points in the life
span. Their participants were young adult exemplars (who had been identi-
fied for their extraordinary moral commitment to social service agencies),
along with a matched comparison group, who were then contrasted with the
influential historical figures from Frimer et al.’s (2012) study. The young
adult participants completed an interview, which, among other aspects,
focused on life goals that were coded for agentic and communal motiva-
tion within an instrumental–terminal framework.
The older comparison figures strongly typified unmitigated agency; the
younger comparison participants also had agency predominate at both the
instrumental and terminal levels, but the effect was attenuated at the termi-
nal level. The older exemplars showed a pattern of instrumental agency but
MPQ 61.3_05.indd 417 28/05/15 2:43 AM
418 Merrill-Palmer Quarterly
terminal communion. The younger exemplars were not quite there: They
showed predominantly instrumental agency, but, at the terminal level, they
were in a divided state, pursuing agentic and communal goals with equal
vigor. Thus, Dunlop et al.’s (2013) study provides preliminary evidence
that the relation between the moral motives of agency and communion does
change over the course of development.
The Present Study
The purpose of the present study was to map developmental trajecto-
ries of agentic and communal motives within an instrumental–terminal
framework. The study involved a cross-sectional research design, using a
community-based sample comprising four age groups from a wide portion
of the life span (8–45 years), to assess different phases that may mark tran-
sitions (childhood, adolescence, emerging adulthood, and mid-adulthood).
Because patterns of agency and communion may vary across cultural con-
texts (e.g., individualistic and collectivistic), we restricted our sample to
the most populous ethnic group in the local region and deferred questions
about cultural variability to future research (see the Discussion section).
We assessed agentic and communal motives with a semistructured inter-
view, asking each participant about the value motivation in various aspects
of their everyday lives and prompting for the instrumental–terminal struc-
ture in their thinking.
Extrapolating from the findings of previous research, we predict that
agency will be the dominant instrumental motive across the life span from
childhood to mid-adulthood, but that the hegemony of agency as the termi-
nal motive will decline over age as communal ends increase in significance.
Delineating these trajectories of agency and communion will advance the
process of reverse engineering of moral motivation and set the stage for
subsequent research on the mechanisms underlying these trajectories and
their relationship to behavior.
Method
Participants
Participants were from Greater Vancouver, a cosmopolitan city on the west
coast of Canada. Research assistants recruited prospective participants in
public places in the community (e.g., community centers, coffee shops,
parks) and through a process of snowballing. Aiming to avoid the pos-
sible moderating effects of culture/ethnicity, we limited the sample to the
MPQ 61.3_05.indd 418 28/05/15 2:43 AM
Trajectories of Agency and Communion 419
most populous local ethnicity; thus, only individuals who self-identified as
White/Caucasian, had English as their first language, and fit into one of the
four age groups of interest (childhood, adolescence, emerging adulthood,
and mid-adulthood) qualified for the study. Limiting the study to one eth-
nicity also helped reduce the pragmatic challenge of recruiting demograph-
ically matched age groups. Participants received an honorarium of $25.
The sample comprised 140 participants across four target age groups:
childhood (8–12 years, M = 10.8, SD = 1.5, n = 20 boys and 20 girls),
adolescence (14–18 years, M = 16.8, SD = 1.9, n = 20 boys and 21 girls),
emerging adulthood (20–28 years, M = 23.4, SD = 2.7, n = 20 men and 19
women), and mid-adulthood (35–45 years, M = 40.9, SD = 3.4, n = 9 men
and 11 women).
Interview
After providing written informed consent (parents for their children and
adults for themselves), participants completed a self-report measure (not of
interest here) and then responded to an individual audio-recorded interview
that implicitly measured value motives of agency and communion. One of
three trained female research assistants conducted the interviews, typically
in a university lab. These interviews, averaging 30 minutes in length, were
subsequently transcribed verbatim for blind coding.
The semistructured interview (inspired by Damon and Hart’s [1988]
self-understanding interview and by Frimer and Walker’s [2009] self-
understanding interview–transmogrified) asked participants about various
aspects of their lives but within an instrumental–terminal framework, allow-
ing their value-laden motives to emerge in an implicit fashion (McAdams,
1993). We contend that this implicit approach to assessing value motivation
is not merely a redundant and labor-intensive means of measuring the same
individual differences as would be provided by self-report measures (such
as the Schwartz Value Survey; Schwartz, 1992). First, extant self-report
measures do not appropriately tap the critical distinction between instru-
mental and terminal values and so do not assess their hierarchical integra-
tion that is of interest here. We rely on implicit measures because they
much more clearly distinguished moral exemplars from comparison par-
ticipants in previous research (Walker & Frimer, 2007). Second, implicit
and explicit measures access different psychological systems, with implicit
measures being more predictive of life outcomes than are explicit mea-
sures and less predictive of behavior in any specific context (McClelland,
Koestner, & Weinberger, 1989). Not surprisingly, then, implicit and explicit
measures have been found to correlate weakly (Thrash & Elliot, 2002)
MPQ 61.3_05.indd 419 28/05/15 2:43 AM
420 Merrill-Palmer Quarterly
perhaps because they access different selves (agents vs. actors; McAdams,
2013) with different moral qualities (Frimer, Schaefer, & Oakes, 2014).
Seed questions. The interview had five seed questions regarding sev-
eral domains of daily life: (a) Do you go to school? What do [did] you
study? (b) Which of your activities are most important to you? (c) Who
are the most significant people or groups of people in your life? (d) What
is important to you in terms of your physical characteristics? (e) What are
your responsibilities? These general questions were simply intended to
focus attention on different but important domains of everyday life.
Instrumental–terminal chain co-construction. Each time a participant
provided a response, the interviewer asked a series of follow-up prompts
intended to co-construct, with the participant, an instrumental–terminal
chain of moral motives. An instrumental–terminal chain is a sequence of
concepts wherein any one concept is the result of the previous, ending in a
terminal value that a person describes as being worthwhile in its own right.
For example, if concept X will bring about concept Y, which will, in turn,
bring about concept Z, then X is instrumental to Y, and Y is instrumental to
Z (symbolically, X → Y → Z). For example, a person may choose to study
math (X) to get a good job (Y) to be able to support a family (Z). See the
Appendix for annotated examples of two interview excerpts.
To co-construct a chain, the interviewer asked participants to explain
the significance of (and thereby unwittingly give value content to) a first
link. For example, if the response to a seed question was “I study math,”
the interviewer would ask, “Why is studying math important to you?” The
interviewer then subtly prompted the participant to explain the instrumen-
tality of the link by asking, “What has been the result of [studying math]?”
or “What do you hope to achieve by [studying math]?” The process was
then repeated, establishing the value content and instrumentality of each
link until the participant identified a terminal motive and had nothing fur-
ther to add. And, when the interviewer suspected that the participant had no
further links to report, she checked this by asking, “Is [supporting a family]
satisfying in itself or do you hope that something else will come from it?”
The chain was complete when the participant (a) had no further response
or (b) offered a generic, value-neutral statement (e.g., “Just because” or “It
will make me happy”3); otherwise, the co-construction of the chain contin-
ued. Across participants, each chain averaged 4.2 (SD = 1.0) links, indicat-
ing that several instrumental motives were typically implicated before a
terminal motive was articulated.
3. Schwartz et al. (2001) found that happiness belongs at the neutral point of the typology
of 10 universal values.
MPQ 61.3_05.indd 420 28/05/15 2:43 AM
Trajectories of Agency and Communion 421
Participants often provided several responses to each of the five seed
questions in the interview. The interviewer followed up on each response
and co-constructed a respective chain. Across participants, interviews aver-
aged 10.7 (SD = 2.8) chains, providing a good sampling of value motiva-
tion in daily life. Thus, the interview allowed participants to express the
structure of their moral motivation in several domains of everyday life.
Coding
Trained research assistants coded the transcribed interviews, first for inter-
view structure and then for motives at each location within the structure.
Structure coding. Structure coding involved carefully reading the tran-
script and identifying the exact text that comprised each link of each instru-
mental–terminal chain. The coder circled the text that made up a link and
delineated each link’s instrumental–terminal location by drawing arrows
from one link to the next. The coder then summarized each link with a stem
(i.e., a few words that captured the main thrust of the link) and entered this
information into a computer spreadsheet.
A second trained coder determined the reliability of structure by cod-
ing a random subset (25%) of the interviews. To avoid compounding errors
in subsequent steps and thus underestimating the reliability of coding, we
divided the reliability coding into three sequential steps that map the logic
of the coding process. The reliability coder followed the same procedure
as the primary coder save for the modifications described in the following
paragraph.
After coding the interview for a participant, the reliability coder
counted the number of chains and recorded this in a spreadsheet. This
first step of coding produced near-perfect agreement, r(34) = .96. In the
rare cases of disagreement, the reliability coder adopted the primary cod-
er’s response and revised the reliability coding as needed. After doing
so, the reliability coder, in the second step, counted the number of links
for each chain. This second step also produced near-perfect agreement,
r(370)=.92. Again, in the rare cases of disagreement, the reliability coder
then adopted the response of the primary coder and revised the reliability
coding accordingly. In the third and final step, the reliability coder identi-
fied a stem (afew words that captured the essence) for each link. A second
reliability coder judged stem agreement by ascertaining whether the 1,572
pairs of stems that each of the primary and reliability coders had produced
captured the same concept. Reliability on this third step was, again, excel-
lent with 96% agreement. We relied on data from the primary coder for all
subsequent coding and analyses.
MPQ 61.3_05.indd 421 28/05/15 2:43 AM
422 Merrill-Palmer Quarterly
Motive coding. Once the structure coding of the interview was
complete, a trained research assistant undertook motive coding by
analyzing the content of each stem. For each stem, the coder judged the
single most strongly implied value motive by matching concepts to the
Values Embedded in Narrative (VEiN) Coding Manual (Frimer, Walker,
& Dunlop, 2009), which taps Schwartz’s (1992) typology of values in
narrative text. Coding was performed at the level of the 10 VEiNs (viz.,
power, achievement, hedonism, stimulation, self-direction, universalism,
benevolence, tradition, conformity, and security). As noted before, the
10 values are situated around a circumplex, summarized by two bipolar
dimensions, forming four quadrants: agency (power and achievement)
versus communion (universalism and benevolence), and conservation
( tradition, conformity, and security) versus openness to change (self-
direction, stimulation, and hedonism). With our theoretical interest being
in two value quadrants—agency (as power and/or achievement) and com-
munion (as universalism and/or benevolence)—we recoded the VEiNs to
be agency, communion, or neither. A second trained coder determined the
reliability of this classification by coding a random subset (25%) of the
interviews. Reliability was substantial with 86% agreement and κn = .80.
Validity of motive coding. The VEiN methodology has been validated
in several studies: Frimer and Walker (2009) found that VEiN coding dis-
tinguished participants in terms of their self-reported and observed moral
behavior; and Frimer et al. (2011, 2012) and Dunlop et al. (2013) found
that VEiN coding readily distinguished different sets of moral exemplars
(who have engaged in a “moral career”) from matched comparison groups.
And one of the significant advantages of the VEiN methodology is that its
coding can be applied to any sort of textual material, including life-story
interviews (Dunlop et al., 2013; Frimer et al., 2011), self-understanding
interviews (Frimer & Walker, 2009), personal strivings and personal proj-
ects (Dunlop et al., 2013; Frimer et al., 2011), and speeches and media
interviews (Frimer et al., 2012).
Analytic Strategy
The number of chains that participants produced in response to questioning
in the interview was unrelated to age, r(139) = –.08, p = .33, and there was
a nonsignificant trend for the number of links per chain to decrease slightly
with age, r(139) = –.16, p = .06. Counter to the intuition that perhaps
adults have more to say than children or are more likely to confabulate, the
data weakly suggest the opposite. Variability in verbal production is not
of prime interest in this study; however, to remove individual differences
MPQ 61.3_05.indd 422 28/05/15 2:43 AM
Trajectories of Agency and Communion 423
in verbal productivity, we operationalized motives in terms of proportions
(dividing the relevant frequencies by the total opportunities). For example,
we calculated the proportion of terminal agency as the number of terminal
concepts expressed that endorsed agentic (power or achievement) values,
divided by the total number of chains in the interview. With this operation-
alization, participants who spoke more in the interview (and therefore had
more chains or more links per chain) would be no more or less likely to
evidence any particular motives than less talkative participants.
The proportions of terminal agency and terminal communion are
straightforward to determine since each chain of responses has a single
terminal motive. The proportions of instrumental agency and instrumen-
tal communion are less straightforward to determine since the number of
instrumental links varies across chains. To determine the proportions of
instrumental agency and instrumental communion, averages were calcu-
lated based on the number of links in each chain. For example, if a chain
of responses had five instrumental links (leading to a terminal value) with
three links being coded as agency, one as communion, and one as nei-
ther, the proportion of instrumental agency for that chain would be .6 and
the proportion of instrumental communion would be .2. These proportions
were then averaged over the chains of responses in the interview.
To summarize, we examined value motives within an instrumental–
terminal framework by interviewing people from four age groups about
various aspects of their lives and asking them to explain how their cur-
rent endeavors might be instrumental to more ultimate (terminal) goals. We
later analyzed the content of the value motives that emerged in their chains
of responses in terms of agency (power and/or achievement) and commu-
nion (benevolence and/or universalism).
Results
Omnibus Analysis
Our primary interest was in developmental trajectories of moral motivation
as expressed within an instrumental–terminal framework. Does agency
or does communion dominate? Does it depend on the level within this
framework? How do they relate? Our main analysis took the form of a 4
(age group: childhood, adolescence, emerging adulthood, mid-adulthood)
× 2 (gender: male, female) × 2 (motive: agency, communion) × 2 (level:
instrumental, terminal) mixed-model analysis of variance (ANOVA) that
used proportional scores as the dependent variable. We included gender
as a variable because agency is sometimes aligned with masculinity and
MPQ 61.3_05.indd 423 28/05/15 2:43 AM
424 Merrill-Palmer Quarterly
communion with femininity (Wiggins, 1991). However, in this analysis,
the main effect of gender and all interactions involving gender were not
significant (ps > .21 and ηp
2s < .04). Thus, we consider gender no further
in these analyses.
This omnibus analysis revealed a main effect for motive,
F(1,132)=128.76, p < .001, ηp
2 = .49, indicating that, overall, people
express more agency than communion. A main effect for level was also
revealed, F(1, 132) = 23.53, p < .001, ηp
2 = .15, indicating that people
expressed more agency and communion at the instrumental than at the ter-
minal level. Significant two-way interaction effects for age group × motive
and for motive × level were also revealed, but were qualified by the critical
three-way age group × motive × level interaction, F(3, 132) = 3.58, p = .02,
ηp
2 = .07.
Motives for Each Age Group
We decomposed this three-way interaction with a motive × level ANOVA
for each age group separately, addressing this question: Does the structure
of moral motivation differ across developmental periods? Figure 1 shows
the proportion of instrumental agency and communion and of terminal
agency and communion for each age group.
In childhood, only main effects of motive (with agency being domi-
nant over communion) and of level (with stronger motivation at the
instrumental level than the terminal level) emerged, Fs(1, 39) = 126.49
and 14.24, ps<.001, ηp
2s = .76 and .27, respectively. For each of the
three other age groups, a motive × level interaction was found: for ado-
lescence, F(1,40)=8.88, p= .005, ηp
2 = .18; for emerging adulthood,
F(1,38)=17.11, p<.001, ηp
2 = .31; and for mid-adulthood, F(1, 19)=8.77,
p=.008, ηp
2=.32.
In adolescence, follow-up analyses indicated that agency predominated
over communion at both the instrumental and the terminal levels, but that
the effect was noticeably weaker at the terminal level, Fs(1, 40)=62.28
and 7.23, ps < .001 and = .01, ηp
2s = .61 and .15, respectively.
Likewise, in emerging adulthood, analyses following up on the
motive× level interaction indicated that agency predominated over com-
munion at both levels, but that the effect was again weaker at the terminal
level, Fs(1,38) = 86.12 and 9.80, ps < .001 and = .003, ηp
2s = .69 and .21,
respectively.
In mid-adulthood, follow-up analyses indicated that, at the
instrumental level, as with the other age groups, agency was more
MPQ 61.3_05.indd 424 28/05/15 2:43 AM
Trajectories of Agency and Communion 425
prevalent than communion, F(1, 19) = 7.53, p = .01, ηp
2 = .29. However,
at the terminal level, the prevalence of agentic and communion moti-
vation did not differ, F(1, 19) = 0.43, p = .52, ηp
2 = .02; indeed, the
direction of the difference had reversed, with more terminal communion
than agency.
In summary, in all age groups from childhood to mid-adulthood,
instrumental motivation was primarily expressed in agentic terms,
which is not surprising given its instrumentality. In terms of termi-
nal values, agency was also the dominant motive in the younger age
groups, but the effect progressively weakened and, by mid-adulthood,
had dissipated.Indeed, an analysis of the relations between terminal
motives and exact age indicates that terminal agency decreased over
age,r(139)= –.21, p = .01, whereas terminal communion increased
with age, r(139) = +.40, p < .001. The rise of terminal communion is
particularly pronounced, increasing from 8% in childhood to 26% in
mid-adulthood.
.0
.1
.2
.3
.4
.5
.6
instrumental terminal
proportion
motive
childhood
agency
communion
.0
.1
.2
.3
.4
.5
.6
instrumental terminal
proportion
motive
adolescence
.0
.1
.2
.3
.4
.5
.6
instrumental terminal
proportion
motive
emerging
adulthood
.0
.1
.2
.3
.4
.5
.6
instrumental terminal
proportion
motive
mid-adulthood
Figure 1. Proportional scores for instrumental agency and communion and for
terminal agency and communion for each of four age groups. Error bars indicate
95% confidence intervals.
MPQ 61.3_05.indd 425 28/05/15 2:43 AM
426 Merrill-Palmer Quarterly
Patterns of Initial Instrumental and Terminal Motivation
Although revelatory in many respects, the previous set of analyses has
a couple of limitations that can be overcome by examining the data in a
different manner. The previous analyses assessed instrumental agentic
and communal motivation by averaging over the links within a chain of
responses. This may best reflect the “weight” of the instrumental motiva-
tion, but it obscures the relationship between the initially expressed motive
and the terminal value, which would be more consistent with the analytic
strategy of previous research (Frimer et al., 2012). In the following set of
analyses, various patterns in the relationship between the initial instrumen-
tal motive (the first link in the chain of responses) and the terminal motive
are assessed: (a) initial instrumental agency culminating in terminal agency
(agency → agency; i.e., agency for agency’s sake), (b) initial instrumental
agency for a communal end (agency → communion), (c) initial communion
leading to agency (communion → agency), and (d) initial communion cul-
minating in terminal communion (communion → communion). The pro-
portion of each participant’s chains of responses reflecting each of these
four patterns was determined.
The previous set of analyses examined the relationship between the
type (agency, communion) and level (instrumental, terminal) of moral
motivation within each age group—and found that these relationships
indeed differed across age groups—but the analyses did not directly
assess developmental trajectories. In the following set of analyses, devel-
opmental trends in the four patterns of relationships between initial and
terminal motive are examined. Figure 2 illustrates these developmental
trajectories.
For each pattern of relationships, we conducted a trend analysis by
using age group as the independent variable. For agency → agency, a
significant linear trend was found, F(1, 136) = 8.22, p = .005, ηp
2 = .06, indi-
cating that this pattern of motivation generally declines across age groups
(see Figure 2); indeed, the correlation with exact age was r(139)=–.21,
p =.01. However, a cubic trend was also significant, F(1, 136) =8.67,
p= .004, ηp
2 = .06, meaning that this pattern reasserts itself somewhat in
emerging adulthood.
For agency → communion (arguably the ideal form of integrated
motivation), a significant linear trend was found, F(1, 136) = 8.60, p =.004,
ηp
2 = .06, indicating that the pattern of instrumental agency for communal
ends clearly increases across age groups. The correlation between the
proportional scores for this pattern and exact age was r(139) = +.24, p = .004.
MPQ 61.3_05.indd 426 28/05/15 2:43 AM
Trajectories of Agency and Communion 427
.0
.1
.2
.3
childhoodadolescence emerging
adulthood
mid-adulthood
proportion
age group
agency agency
.0
.1
.2
.3
childhoodadolescence emerging
adulthood
mid-adulthood
proportion
age group
agency communion
.0
.1
.2
.3
childhoodadolescenceemerging
adulthood
mid-adulthood
proportion
age group
communionagency
.0
.1
.2
.3
childhoodadolescence emerging
adulthood
mid-adulthood
proportion
age group
communion communion
→
→→
→
Figure 2. Proportional scores, across age groups, for four different patterns of
links from initial to terminal motive. Error bars indicate 95% confidence intervals.
The communion → agency pattern was rare for any age (averaging
just 2%), and the trend analysis indicated no significant effects, F(3, 136)
= 0.34, p = .80, ηp
2 = .01.
Finally, for communion → communion, a significant linear trend
was found, F(1, 136) = 6.06, p = .015, ηp
2 = .04, indicating that the moti-
vational pattern of communion for communion’s sake increases over age
groups (as with agency for communion’s sake). The correlation between
the proportional scores for this pattern and exact age was r(139) = +.25,
p = .005.
Discussion
All of us possess, in some measure, the desires to promote our own
interests as well as the interests of others, but these desires are often
in conflict, with the advancement of one resulting in the diminution of
the other, typically with the interests of the self prevailing. Advancing
MPQ 61.3_05.indd 427 28/05/15 2:43 AM
428 Merrill-Palmer Quarterly
the interests of the self (self-enhancing agency) and advancing the
interests of others (self-transcending communion) are typically con-
ceptualized as an oppositional dualism. Agentic motivation, as defined
here, is concern for social power, dominance, material resources, and
achievement, whereas communal motivation is concern for the welfare
of others in everyday interactions, as well as a more universalized con-
cern for thosebeyondthe primary reference group, including ecological
preservation.
Moral exemplars, however, have seemingly surmounted this dualism,
with agency and communion becoming hierarchically integrated such
that self-interests are fulfilled through the pursuit of communal concerns
(Colby & Damon, 1992; Dunlop et al., 2013; Frimer et al., 2011, 2012;
Walker & Frimer, 2007). When agency is channeled into communal goals,
the personal interest at stake provides the motivational impetus to ener-
gize judgment into action (Bergman, 2004; Blasi, 1984; for a review, see
Walker, 2014). Thus, the “finished product”—developmentally mature
moral exemplars—evidence a pattern of moral motivation that synergisti-
cally integrates agency into communion.
We suggest that this integration represents a developmental achieve-
ment, with typical adults having traversed some, but not all, of the devel-
opmental trajectory from childhood to moral exemplarity. We conducted
a cross-sectional study with four age groups (childhood, adolescence,
emerging adulthood, and mid-adulthood), covering much of the life span
(8–45 years). We assessed moral motives of agency and communion
within an instrumental–terminal framework by interviewing participants
about various aspects of their lives and asking them to explain how their
current endeavors might be instrumental to their terminal goals. This
enabled a developmental mapping of trajectories of moral motivation that
may lead to the end-point integration of agency and communion. Tapping
the instrumental–terminal distinction in value motivation became clearly
evident as an important conceptual and methodological innovation of
thiswork.
Agency was the dominant instrumental motive from childhood
through mid-adulthood, representing the inherent and ubiquitous motiva-
tional structure of goal-directed activity. Agency was also the dominant
terminal motive early in development, but, with increasing age, its hege-
mony as a terminal value diminished, albeit far from being completely
extinguished (consistent with some of the previous research on devel-
opmental trends in motivation reviewed in this article’s introduction).
Thus, the typical pattern from childhood through emerging adulthood
MPQ 61.3_05.indd 428 28/05/15 2:43 AM
Trajectories of Agency and Communion 429
indicated a movement toward, but failure to achieve, integration; indeed,
the dominant pattern was agency as the means to more agency. This pat-
tern was particularly evident in emerging adulthood, a period that may
represent a motivational turning point in the life span (Arnett, 2007;
Erikson, 1968) as young adults prepare for careers and choose among
certain social roles and lifestyles.
Communion as a terminal value increased markedly in significance
from childhood to mid-adulthood, and the pattern of instrumen-
tal agency in service to terminal communion—the critical form of
motivational integration—similarly increased in frequency. By mid-
adulthood, terminal motives tended to be about equally agentic and
communal.
Integrating with previous research (Dunlop et al., 2013; Frimer
etal., 2011, 2012), several relevant observations can be made. The moti-
vational profiles of children in this study and the nonmoral influential
figures in Frimer et al.’s (2012) study were strikingly similar: Both
treated agency as a means to more agency. The emerging adults in this
study and the young-adult comparison participants in Dunlop et al.’s
(2013) study evidenced the same profile, which is as would be expected
given the similar age range and which provides a partial replication.
The midlife adults in the present study evidenced parity between agency
and communion as a terminal motive, which is the same pattern as the
younger moral exemplars in Dunlop et al.’s study (indicating that those
young-adult exemplars are developmentally advanced), but both groups
fall short of the strong pattern of instrumental agency for terminal com-
munion exemplified by the moral exemplars of historical renown (Frimer
et al., 2012). In other words, the accumulated evidence is that the devel-
opment of moral motivation covers most of the life span and is typically
not complete in most adults.
The basic issue addressed by the present study focused on how moral
motivation develops across the life span. The findings of previous studies
in this program of research indicated that developmental maturity, as indi-
cated by the functioning of moral exemplars, entails the synergistic inte-
gration of the typically antagonistic motives of agency and communion.
The contribution of the present study was to further the process of reverse
engineering by charting the developmental trajectory of the relationship
between agentic and communal motivation that may lead to this end-
point integration that fosters moral excellence. Empirically, our innovative
interview methodology that elicited participants’ framing of their instru-
mental motives in relation to terminal (ultimate) values was particularly
MPQ 61.3_05.indd 429 28/05/15 2:43 AM
430 Merrill-Palmer Quarterly
consequential because it revealed that the hierarchically integrated pattern
of agentic motivation in service to communal concerns increased across
age groups from childhood to mid-adulthood and implied that displacing
agency with communion as the terminal value should be a fundamental
goal for moral development.
Limitations and Future Directions
A limitation of the present study is that inferences regarding
developmental trajectories were made on the basis of cross-sectional
data. Only a longitudinal design has the potential to rule out cohort
effects and to assess intraindividual patterns of change in the relationship
between agency and communion. Interestingly, generational differences
on related variables have been reported. For example, Twenge, Campbell,
and Freeman (2012) analyzed national survey data from high school and
university students collected at different points in time, thus represent-
ing different birth cohorts: Boomers (born 1943–1961), Generation
X’ers (born 1961–1981), and Millennials (born 1982–1999). Overall
trends indicated that later cohorts more strongly endorsed extrinsic and
narcissistic life goals than did earlier cohorts (suggesting a pattern of
increasing agency across generations) while, at the same time, later
cohorts were less likely to endorse goals reflecting concern for others,
community feeling, and empathy (suggesting a concomitant pattern
of decreasing communion). Untangling age- and time-related effects
remains a focus for future research.
A related limitation of the present study is that portions of the life
span were not represented in the sample. How moral motives are config-
ured in early childhood and whether these motives continue to develop in
later adulthood remain questions for further research. The methodology
employed here may not be amenable to assessing the motivation of young
children, so that presents a further research challenge.
Another limitation of the present study was the restriction of the
sample to one cultural/ethnic group. Indeed, there is some evidence that
patterns of agency and communion may vary between individualistic
and collectivistic cultures (Gebauer, Wagner, Sedikides, & Neberich,
2013). Henrich, Heine, and Norenzayan (2010) have urged caution in
drawing inferences about humanity at large based on data from a single
subpopulation, particularly one that is weird (Western, educated,
industrialized, rich, and democratic). They concluded, in their review,
that members of individualistic Western societies were more likely to
MPQ 61.3_05.indd 430 28/05/15 2:43 AM
Trajectories of Agency and Communion 431
hold independent self-concepts, whereas interdependent self-concepts
were more common in collectivistic non-Western societies (Markus
&Kitayama, 1991), as evidenced by Westerners’ more positive self views,
heightened value on self-direction and choice, and weaker motivation to
conform.
These particular aspects of agency and communion for which there is
evidence of cultural variability seem to pertain primarily to psychological
distance (i.e., individuality–relatedness; Grotevant & Cooper, 1998) rather
than to the promoting-interests aspects that are the focus of the pres-
ent study. In that regard, it is noteworthy that Frimeret al. (2011) found
that agency and communion were integrated in the motivation of moral
exemplars only when these constructs were defined in terms of promoting
interests of self versus others and not when they were defined in terms of
an alternate scheme focusing on psychological distance. Further, Henrich
et al. (2010) noted striking similarities across cultural groups and soci-
eties in some psychological characteristics, notably including personality
structure (e.g., behavioral traits of dominance and nurturance). However,
given our individualistic sample, we make no claim that the effects gener-
alize to other cultural orientations, although recent findings suggest that
they may. Frimer et al. (2014) asked both individualists and collectiv-
ists to describe their goals and then rate the degree to which their goals
promote self- interest (agency) and the interests of others (communion).
Both groups claimed that their goals advanced the self’s interests more so
than the interests of others. This dominance of agency over communion
was evident in both samples, albeit greater for individualists, suggesting
that culturecan only partially moderate (and not eliminate) the domi-
nanceofagencyover communion. Future research should examine cultural
variability inthe interplay betweenagency and communionwithinthe
instrumental– terminal framework that was so informative here.
In a related manner, the findings may be constrained by social-structural
influences that are extant in modern technological societies. Such influences
in children’s and adolescents’ lives seemingly are rather agentic in nature,
including the ubiquitous requirement of school attendance that values
achievement, the structured and often-coerced activities that pervade out-
of-school time, the technological devices that can constrain meaningful
human interaction, and the dominance of a worldview that often reifies
economic development over other concerns. It was not always so, and it
is not so everywhere. Thus, it is important to acknowledge the context in
which the present research is situated and to remain mindful of the poten-
tial impact of the social-structural influences on patterns of motivation.
MPQ 61.3_05.indd 431 28/05/15 2:43 AM
432 Merrill-Palmer Quarterly
Despite its effectiveness in mapping developmental trajectories of
moral motivation, the methodology of the present study presents an
obstacle for further research. The individual interview format and the
painstaking microanalytic coding of structure and motives are very labor
intensive, and more expedient methods for obtaining similar data with
reliability and validity would be advantageous for the research enterprise.
Note, however, that implicit and explicit measures of the same construct
rarely correlate with each other and often relate to behavioral indices in
markedly different ways (McClelland et al., 1989), suggesting that they
are tapping fundamentally different underlying mechanisms. Further
research employing a range of measures of motivation may provide
some clarification in this regard.
The present research examined the motivational function-
ing of ordinary folk. Future research should map and compare the
developmental trajectories of moral exemplars versus other types of
exemplars ( political, business). Such research could test that notion
(Frimer & Walker, 2009)that moral exemplars follow a typical develop-
mental trajectory up until some turning-point phase when they diverge
from their cohort.
Such research begs the question of the developmental mechanisms
underlying divergent developmental trajectories, which is the
subsequent step in the process of reverse engineering. What influences
the variousturning points, progressions, stagnations, and end states?
How do these different developmental trajectories relate to behavior?
We anticipate in that regard that a moral lifestyle will be more strongly
related to patternsof moral motivation than idiosyncratic single
behaviors.
The evidence from the present study is that the typical developmental
trajectory is for motives of terminal communion to increase and that
instrumental agency puts wind into these sails. We can speculate
regarding—and proffer for future research—various social and intra-
psychic factors that might contribute to this pattern of moral motivation:
for example, developing social awareness and expanded worldviews
(Selman, 1980), resolution of the tension between agentic and communal
motivation (Frimer & Walker, 2009), meaningful contact with a prosocial
cause (Hart& Matsuba, 2009), and the changing demands and opportuni-
ties associated with different phases of the life span (e.g., generativity;
McAdams, Diamond, de St. Aubin, & Mansfield, 1997). Once the
psychological mechanisms underlying patterns of moral motivation are
better understood, we will be in the place to implement intervention pro-
grams and to effect change.
MPQ 61.3_05.indd 432 28/05/15 2:43 AM
Trajectories of Agency and Communion 433
References
Arnett, J. J. (2007). Emerging adulthood: What is it and what is it good for? Child
Development Perspectives, 1, 68–73. doi:10.1111/j.1750-8606.2007.00016.x
Bakan, D. (1966). The duality of human existence: An essay on psychology and
religion. Chicago: Rand McNally.
Bergman, R. (2004). Identity as motivation: Toward a theory of the moral self. In
D. K. Lapsley & D. Narvaez (Eds.), Moral development, self, and identity
(pp.21–46). Mahwah, NJ: Erlbaum.
Blasi, A. (1984). Moral identity: Its role in moral functioning. In W. M. Kurtines
& J. L. Gewirtz (Eds.), Morality, moral behavior, and moral development
(pp.128–139). New York: Wiley.
Colby, A., & Damon, W. (1992). Some do care: Contemporary lives of moral com-
mitment. New York: Free Press.
Damon, W., & Hart, D. (1988). Self-understanding in childhood and adolescence.
Cambridge: Cambridge University Press.
Diehl, M., Owen, S. K., & Youngblade, L. M. (2004). Agency and communion
attributes in adults’ spontaneous self-representations. International Journalof
Behavioral Development, 28, 1–15. doi:10.1080/01650250344000226
Dunlop, W. L., Walker, L. J., & Matsuba, M. K. (2013). The development of moral
motivation across the adult lifespan. European Journal of Developmental
Psychology, 10, 285–300. doi:10.1080/17405629.2012.746205
Ely, R., Melzi, G., Hadge, L., & McCabe, A. (1998). Being brave, being nice:
Themes of agency and communion in children’s narratives. Journal of
Personality, 66, 257–284. doi:10.1111/1467-6494.00012
Erikson, E. H. (1968). Identity, youth, and crisis. New York: Norton.
Frimer, J. A., Schaefer, N. K., & Oakes, H. (2014). Moral actor, selfish agent.
Journal of Personality and Social Psychology, 106, 790–802. doi:10.1037/
a0036040
Frimer, J. A., & Walker, L. J. (2009). Reconciling the self and morality: An
empirical model of moral centrality development. Developmental Psychology,
45, 1669–1681. doi:10.1037/a0017418
Frimer, J. A., Walker, L. J., & Dunlop, W. L. (2009). Values Embedded in Narrative
(VEiN) coding manual. Unpublished coding manual, University of British
Columbia.
Frimer, J. A., Walker, L. J., Dunlop, W. L., Lee, B. H., & Riches, A. (2011). The
integration of agency and communion in moral personality: Evidence of
enlightened self-interest. Journal of Personality and Social Psychology, 101,
149–163. doi:10.1037/a0023780
Frimer, J. A., Walker, L. J., Lee, B. H., Riches, A., & Dunlop, W. L.
(2012). Hierarchical integration of agency and communion: A study
MPQ 61.3_05.indd 433 28/05/15 2:43 AM
434 Merrill-Palmer Quarterly
of influential moral figures. Journal of Personality, 80, 1117–1145.
doi:10.1111/j.1467-6494.2012.00764.x
Gebauer, J. E., Wagner, J., Sedikides, C., & Neberich, W. (2013). Agency–
communion and self-esteem relations are moderated by culture, religiosity, age,
and sex: Evidence for the “self-centrality breeds self-enhancement” principle.
Journal of Personality, 81, 261–275. doi:10.1111/j.1467-6494.2012.00807.x
Grotevant, H. D., & Cooper, C. R. (1998). Individuality and connectedness in ado-
lescent development: Review and prospects for research on identity, relation-
ships, and context. In E. E. Skoe & A. L. von der Lippe (Eds.), Personality
development in adolescence: A cross national and life span perspective
(pp.3–37). Florence, KY: Taylor & Frances/Routledge.
Hart, D., & Matsuba, M. K. (2009). Urban neighborhoods as contexts for moral iden-
tity development. In D. Narvaez & D. K. Lapsley (Eds.), Personality, identity,
and character: Explorations in moral psychology (pp. 214–231). NewYork:
Cambridge University Press. doi:10.1017/CBO9780511627125.010
Henrich, J., Heine, S. J., & Norenzayan, A. (2010). The weirdest people in the world?
Behavioral and Brain Sciences, 33, 61–83. doi:10.1017/S0140525X0999152X
Hogan, R. (1982). A socioanalytic theory of personality. In M. M. Page (Ed.),
Nebraska Symposium on Motivation: Vol. 30. Personality: Current theory and
research (pp. 55–89). Lincoln: University of Nebraska Press.
Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for
cognition, emotion, and motivation. Psychological Review, 98, 224–253.
doi:10.1037/0033-295X.98.2.224
McAdams, D. P. (1988). Power, intimacy, and the life story: Personological inqui-
ries into identity. New York: Guilford Press.
McAdams, D. P. (1993). The stories we live by: Personal myths and the making of
the self. New York: Guilford Press.
McAdams, D. P. (2013). The psychological self as actor, agent, and author. Perspectives
on Psychological Science, 8, 272–295. doi:10.1177/1745691612464657
McAdams, D. P., Diamond, A., de St. Aubin, E., & Mansfield, E. (1997).
Stories of commitment: The psychosocial construction of genera-
tive lives. Journal of Personality and Social Psychology, 72, 678–694.
doi:10.1037/0022-3514.72.3.678
McAdams, D. P., & Olson, B. D. (2010). Personality development: Continuity
and change over the life course. Annual Review of Psychology, 61, 517–542.
doi:10.1146/annurev.psych.093008.100507
McClelland, D. C., Koestner, R., & Weinberger, J. (1989). How do self-attrib-
uted and implicit motives differ? Psychological Review, 96, 690–702.
doi:10.1037/0033-295X.96.4.690
Paulhus, D. L., & Trapnell, P. D. (2008). Self-presentation of personality: An
agency–communion framework. In O. P. John, R. W. Robins, & L. A. Pervin
MPQ 61.3_05.indd 434 28/05/15 2:43 AM
Trajectories of Agency and Communion 435
(Eds.), Handbook of personality psychology: Theory and research (3rd ed.,
pp. 492–517). New York: Guilford Press.
Roberts, B. W., Walton, K. E., & Viechtbauer, W. (2006). Patterns of
mean-level change in personality traits across the life course: A meta-
analysis of longitudinalstudies. Psychological Bulletin, 132, 1–25.
doi:10.1037/0033-2909.132.1.1
Roccas, S., Sagiv, L., Schwartz, S. H., & Knafo, A. (2002). The Big Five personal-
ity factors and personal values. Personality and Social Psychology Bulletin,
28, 789–801. doi:10.1177/0146167202289008
Rokeach, M. (1973). The nature of human values. New York: Free Press.
Schwartz, S. H. (1992). Universals in the content and structure of values: Theoretical
advances and empirical tests in 20 countries. Advances in Experimental Social
Psychology, 25, 1–65. doi:10.1016/S0065-2601(08)60281-6
Schwartz, S. H. (2006). Les valeurs de base de la personne: Théorie, mesures et
applications [Basic human values: Theory, measurement, and applications].
Revue Française de Sociologie, 47, 929–968. doi:10.3917/rfs.474.0929
Schwartz, S. H., Cieciuch, J., Vecchione, M., Davidov, E., Fischer, R.,
Beierlein,C.,… Konty, M. (2012). Refining the theory of basic individ-
ual values. Journal of Personality and Social Psychology, 103, 663–688.
doi:10.1037/a0029393
Schwartz, S. H., Melech, G., Lehmann, A., Burgess, S., Harris, M., & Owens,V.
(2001). Extending the cross-cultural validity of the theory of basic human
values with a different method of measurement. Journal of Cross-Cultural
Psychology, 32, 519–542. doi:10.1177/0022022101032005001
Selman, R. L. (1980). Growth of interpersonal understanding: Developmental and
clinical analysis. New York: Academic Press.
Thrash, T. M., & Elliot, A. J. (2002). Implicit and self-attributed achievement
motives: Concordance and predictive validity. Journal of Personality, 70,
729–756. doi:10.1111/1467-6494.05022
Twenge, J. M., Campbell, W., & Freeman, E. C. (2012). Generational differences in
young adults’ life goals, concern for others, and civic orientation, 1966–2009.
Journal of Personality and Social Psychology, 102, 1045–1062. doi:10.1037/
a0027408
Walker, L. J. (2013). Exemplars’ moral behavior is self-regarding. New Directions
for Child and Adolescent Development, 142, 27–40. doi:10.1002/cad.20047
Walker, L. J. (2014). Moral personality, motivation, and identity. In M. Killen &
J.G. Smetana (Eds.), Handbook of moral development (2nd ed., pp. 497–519).
New York: Psychology Press.
Walker, L. J., & Frimer, J. A. (2007). Moral personality of brave and caring
exemplars. Journal of Personality and Social Psychology, 93, 845–860.
doi:10.1037/0022-3514.93.5.845
MPQ 61.3_05.indd 435 28/05/15 2:43 AM
436 Merrill-Palmer Quarterly
Wiggins, J. S. (1991). Agency and communion as conceptual coordinates for the
understanding and measurement of interpersonal behavior. In D. Cicchetti &
W. M. G r ov e ( E ds .) , Thinking clearly about psychology: Essays in honor of
Paul E . M ee hl (Vol. 2, pp. 89–113). Minneapolis: University of Minnesota
Press.
Wiggins, J. S., & Trapnell, P. D. (1997). Personality structure: The return of the Big
Five. In R. Hogan, J. Johnson, & S. Briggs (Eds.), Handbook of personality
psychology (pp. 737–765). San Diego, CA: Academic Press.
Appendix: Examples of Interview Excerpts
CodedforMotives
Here are two illustrative examples of interview coding, with arrows
delineating instrumental–terminal structure and with motive coding of
relevant links in the left margin. Agentic motives are achievement (AC)
or power (PO). Communal motives are universalism (UN) or benevolence
(BE). Also coded here is self-direction (SD), a motive that belongs within
neither agency nor communion.
A first example illustrates three instrumental agentic motives leading
to a terminal communion motive. The participant was a 15-year-old girl
(ID no. 123). Integrated chains such as this one were rare for this individual
(13%); more of her chains terminated in agency (33%).
Do you go to school?
Yes.
Why is it important to go to school?
AC It’s important to learn things so that later on you can
get a good job. . . .
↓
Why is it important to get a job?
PO You’ll probably need to earn some money for your life,
just to buy things. It’s pretty important to get a job, to
basically earn money. And also if you enjoy it, like if
you enjoy your job that could be nice too.
So it is important to have a job that you enjoy?
Yes.
↓
MPQ 61.3_05.indd 436 28/05/15 2:43 AM
Trajectories of Agency and Communion 437
What can you do with the money you earn?
PO Obviously you need to buy some things, like food or
other necessary things. You could also donate it to
important causes, to help other people.
↓
Why is it important to make donations?
UN I think in [my home town], I think I am pretty for-
tunate and if you . . . well, it is important to help the
other people who are less fortunate. I think here we are
a lot better off than some people in Third World coun-
tries. It is important to help them, too, instead of just
living here with a lot of stuff that they don’t have.
Is it satisfying in itself to help others or do you
hope to achieve something else from it? Or do
you hope something else will come from it?
It is satisfying. I mean if you donate money or
something else like that, knowing that you’ve helped
someone . . . they might need help for.
So it’s satisfying knowing that you’ve helped
someone?
Yes.
A second example illustrates agency dominating both instrumental
and terminal levels of motivation. The participant was an 18-year-old boy
(IDno. 48). Over the course of his interview, his terminal motives were
predominantly agentic (54%) and only rarely communal (8%).
Do you go to school?
AC Yes.
↓
Why is it important to you to go to school?
PO I don’t know; I just think it is the only way to get
ahead. Frankly, nowadays it is very difficult to make
anything of yourself without a degree. At least it is
something.
↓
MPQ 61.3_05.indd 437 28/05/15 2:43 AM
438 Merrill-Palmer Quarterly
Why it is a degree important?
SD I think it is like a starting a step to getting possibly a
job or at least finding out what you want to do.
Do you know the jobs that you are into doing?
No, actually. I am thinking of microbiology as my
area. Although a job? Frankly, I don’t know if I want to
do research. That is kind of a hassle. I also want to be a
writer, so that is fairly conflicting.
So right now it sounds like you are here at
school to, like you say, find out what you might
want to do. Why is it important to have this
time to find out what you want to do?
Because I think in about 5 or 10 years, once you get
into a family life or job setting, you really don’t have
the opportunity to take the time and have fun and see
what you want to do. This is really like an once-in-a-
lifetime opportunity.
↓
What would be the result for you of having
this opportunity to take this time?
PO Hopefully I can find something that I at least mildly like
and hopefully I will make some money, as well. I think
ultimately Iwould rather be happy than make money.
↓
So you would rather be happy in terms of a job
or?
I would rather be at a job that I like.
If you were able to get a job that you liked,
what would be the result of that? Why is it
important to you to do what you like?
AC I think people excel in the areas that they like. I don’t
think anybody will put their 100% effort into some-
thing they do just to get the money. It just seems like,
especially if you are going in to something like sci-
ences, like microbiology, you are not going to put in
the extra effort to get the job done when you are only
kind of cashing out because there are other things.
MPQ 61.3_05.indd 438 28/05/15 2:43 AM
Trajectories of Agency and Communion 439
So you think that if you do what you like then
you would excel. Why is it important to you to
excel?
generic I think you only get one life. Frankly, I don’t really want
to waste mine behind a desk doing nothing. I can really
do anything. I don’t really know what it is that I am
going to find what I like. Whatever it is . . . it just seems
like it is really about being happy, u lti mat ely, I t hink .
MPQ 61.3_05.indd 439 28/05/15 2:43 AM
