Running head: TRANSITIONS IN CIVIC ENGAGEMENT ACROSS ADOLESCENCE
Examining Developmental Transitions in Civic Engagement across Adolescence: Evidence from
a National U.S. Sample
International Journal of Developmental Science, 8, 95-104
Laura Wray-Lakea*, Wendy M. Rotea, Celina M. Benavidesb, and Christine Victorinoc
aUniversity of Rochester
bClaremont Graduate University
cUniversity of California, Riverside
* Address for correspondence:
Laura Wray-Lake, Department of Clinical and Social Sciences in Psychology, University of
Rochester, P.O. Box 270266. Rochester, NY 14627-0266. email@example.com
TRANSITIONS IN CIVIC ENGAGEMENT ACROSS ADOLESCENCE 2
Describing how much and what type(s) of change are evident in civic engagement across
adolescence is a fundamental starting point for advancing developmental theory in the civic
domain. Using five annual waves of data from a large national U.S. sample spanning 8th-12th
grades, our study describes civic engagement typologies and transitions in and out of typologies
across adolescence. Four distinct civic typologies were identified across indicators of civic
values, behaviors, and future expectations. Two-thirds of youth demonstrated ipsative continuity,
i.e., within-class stability over time. Transitions indicated gradual stepwise change in both
upward and downward directions and thus provided only modest support for age-related gains.
Our study has the potential to spur theoretical progress regarding civic development by
documenting developmental change as a series of transitions that vary across people. Results
help to clarify the diverse civic pathways that youth experience across adolescence.
types of civic engagement, Longitudinal Study of American Youth, latent transition analysis,
civic development, civic transitions
TRANSITIONS IN CIVIC ENGAGEMENT ACROSS ADOLESCENCE 3
Despite the vast implications of civic engagement for individuals and society (Sherrod,
Torney-Purta, & Flanagan, 2010), we know very little about systematic longitudinal change in
civic engagement. Thus, fundamental lifespan developmental concepts such as continuity and
discontinuity remain understudied in this key domain of life. Our study can enhance civic
developmental theory by utilizing data from the Longitudinal Study of American Youth spanning
8th through 12th grades and taking a person-oriented approach to identify the prevalence, type,
and timing of developmental transitions in civic engagement across adolescence.
Normative Developmental Change in Civic Engagement
Developmental scholars have long been preoccupied with questions of stability versus
change, and often processes of change are ignored in favor of investigating stability (Overton &
Reese, 1981). Lifespan longitudinal research on civic engagement has emphasized rank-order
stability (i.e., differential continuity), demonstrating that civic engagement in adolescence
positively predicts civic engagement in adulthood (e.g., Hart, Donnelly, Youniss, & Atkins,
2007; Youniss & Yates, 1999). More sophisticated developmental models of change have been
introduced to understand adults’ civic engagement, such as the political life cycle model positing
growth in civic engagement across adulthood (Kinder, 2006). Empirical evidence has shown
growth, decline, and episodic ups and downs in civic engagement across adulthood, with varying
conclusions likely due to different measures, samples, and historical times (e.g., Boehnke &
Boehnke, 2005; Jennings & Stoker, 2004; Neundorf, Smets, & García-Albacete, 2013). Thus,
little empirical consistency has emerged regarding developmental change in civic engagement,
which in turn leads to theoretical confusion regarding how civic engagement changes
normatively with age. Even less developmental theorizing on civic engagement has been applied
TRANSITIONS IN CIVIC ENGAGEMENT ACROSS ADOLESCENCE 4
to adolescence, despite the period being prime for civic awakening given the convergence of
sociocognitive developmental skills with identity and value development (Flanagan, 2004).
Describing how much and what type(s) of change are evident in civic engagement across
adolescence is a fundamental starting point for advancing developmental theory in this domain
Despite a dearth of longitudinal studies of civic engagement in adolescence, other
research suggests expectations for developmental change in levels of civic engagement. For
example, with age, adolescents show gains in abstract thinking, reasoning skills, future planning,
perspective taking, and autonomy (Smetana & Villalobos, 2009). Given that these skills and
capacities are thought to underlie civic engagement, civic engagement may show continuous,
gradual upward change across adolescence in conjunction with such sociocognitive gains (Wray-
Lake & Syvertsen, 2011). We call this the normative growth hypothesis. There is some empirical
support for this hypothesis in the related domain of prosocial behaviors, with some studies
showing expected increases from childhood to adolescence (Eisenberg, Fabes, & Spinrad, 2006).
Person-Oriented Approach to Studying Civic Development
A focus on identifying a single, normative pattern of civic development is likely
insufficient, however, as civic engagement can take many distinct forms and adolescents may not
show the same patterns of change across them. Moreover, a multidimensional perspective
conceptualizes civic engagement as constituted by behaviors as well as values, attitudes, and
knowledge (Amnå, 2012; Flanagan & Faison, 2001; Sherrod & Lauckhardt, 2009). This
multidimensional conceptualization is especially relevant and appropriate for research with
younger populations because adolescents (especially those without legal adult status) are
sometimes behaviorally limited in their civic engagement and vary widely in the ways they
TRANSITIONS IN CIVIC ENGAGEMENT ACROSS ADOLESCENCE 5
choose to express civic commitments (Sherrod & Lauckhardt, 2009).
Person-oriented methodological approaches are well-poised to empirically incorporate
this multidimensional conceptualization of youth civic engagement because these methods can
identify qualitatively distinct typologies, or different ways that individuals combine types of
civic engagement (Finlay, Flanagan, & Wray-Lake, 2011; Pancer, Pratt, Hunsberger, & Alisat,
2007; Voight & Torney-Purta, 2013; Weerts, Cabrera, & Mejía, 2014; Westheimer & Kahne,
2004). Incorporating attitudes and behaviors into person-oriented models can distinguish
between high and low engagement and also identify youth who hold civic attitudes yet do not
engage in action. For example, Voight and Torney-Purta (2013) examined adolescents’ civic
typologies of civic behaviors and attitudes and three groups emerged: “Social justice actors”
were high on both behaviors and attitudes, “civic moderates” were relatively lower on both
behaviors and attitudes, and “social justice sympathizers” reported high civic attitudes but low
Utilizing person-oriented methodology longitudinally can lead to refinements in
developmental theory through identifying ipsative continuity or discontinuity, that is, stability or
change in the way variables are configured within individuals over time (Caspi & Roberts,
2001). Although relatively less common in developmental science to date, longitudinal person-
oriented methods such as latent transition analysis can describe stability in individuals’
classifications for some while simultaneously identifying progressive (upward) or regressive
(downward) patterns for others (Masyn, 2013). This methodological approach has led to
advancements in identity theory by supporting developmental hypotheses about progressive
growth in identity statuses with age (e.g., Meeus, Van de Schoot, Klimstra, & Branje, 2011). In
civic engagement research, a latent transition analysis across the transition to adulthood garnered
TRANSITIONS IN CIVIC ENGAGEMENT ACROSS ADOLESCENCE 6
support for the political life cycle model and demonstrated that participation in the AmeriCorps
program predicted upward transitions in civic engagement over time (Finlay et al., 2011).
Our study examines adolescents’ future civic expectations, current civic values, and
multiple civic behaviors in tandem to holistically understand developmental change in typologies
of adolescents’ civic engagement. By conducting latent transition analysis using five waves of
data from a national U.S. sample spanning 8th-12th grades, our study describes civic engagement
typologies and transitions in and out of these typologies over time. Our analyses allow us to
identify the prevalence of ipsative continuity, describe various types of transition patterns (e.g.,
progressive, regressive), and examine the developmental timing of these transitions.
The Longitudinal Study of American Youth (LSAY; Miller, 2014) is a national study of
American adolescents primary focused on interest and achievement in science and math. We
utilized Cohort Two, in which 7th grade youth started the study in 1987 and were surveyed in
school for six years, participated in a telephone interview in 1994 (i.e., the year after high
school), and completed a phone interview as adults in 2007. The present study primarily utilized
data from five annual fall semester surveys from 8th-12th grade, with select civic measures taken
from spring semester waves. The study utilized a two-stage stratified probability sampling
strategy for initial recruitment of public high schools; feeder middle schools were recruited using
random sampling with replacement. Random samples of 60 students per school were recruited
with replacement. Student response rates for fall student surveys ranged from 72 to 88% (M =
82%). Sample weights account for unequal probability of attrition across schools and regions.
Sample size was 3,701 (52.2% male). Race/ethnicity distributions were as follows: White
TRANSITIONS IN CIVIC ENGAGEMENT ACROSS ADOLESCENCE 7
(69.5%), Black (11.2%), Hispanic (9.1%), Asian (3.6%), Other (1.5%), and Missing (5.1%). The
sample was socioeconomically diverse. According to mother reports, 71% of mothers had a
high-school degree or less, 10% had some college or a two-year degree, 17% had a bachelor’s
degree or higher, and 2% of responses were missing.
Latent class indicators. All civic variables were used as individual dichotomous
indicators in determining latent classes and trajectories. Having all indicators dichotomous made
variable variances more comparable and facilitated interpretation of latent classes. All items
were asked using identical wording and response scales across waves.
Future civic expectations. Adolescents responded to a list of behaviors in terms of
expectations at age 40. Items began with, “By the time I’m 40, I will…” and adolescents marked
all that applied. Four items assessed future civic expectations: “Be a regular voter in local and
national elections,” “Run for public office,” “Be a leader in my community,” and “Be well
informed about political and social issues” (marked items = 1; unmarked items = 0).
Current civic values. Adolescents responded to four civic values items with the prompt,
“How important is the following to you in your life?”: “Being a leader in my community,”
“Working to correct social and economic wrongs,” “Keeping up to date with political and social
issues,” and “Helping other people in my community.” Responses were recoded to 0 = not
important and 1 = somewhat or very important.
Civic behaviors. Student government was measured by spring reports of “This school
year I ran for office in student government” (yes/no). Information consumption was measured as
a yes/no dichotomy that combined fall reports of, “Last summer I…” “read a news magazine
most weeks” (yes/no) and “read a newspaper at least 3 times a week” (yes/no) with spring reports
TRANSITIONS IN CIVIC ENGAGEMENT ACROSS ADOLESCENCE 8
of the same two yes/no items measured with the stem “This school year I…”. Political volunteer
was measured as a single yes/no item that combined fall report of “Last summer I did volunteer
work in a political campaign” (yes/no) with spring reports of same yes/no item from past school
year. Thus, information consumption and political volunteer measures were dichotomies that
captured behavior over a 12-month period.
Validation Measures. Two measures from fall of 8th grade were used to validate class
definitions. Adolescents were asked “How interested are you” and “How well informed are you”
about multiple issue areas. Interest in social issues was calculated by summing interest levels
across six social issue topics: minority and women’s rights, foreign and energy policy, and
economic and military issues. Informed on issues was a sum of informed levels for the same six
issues. Response options ranged from (1) not at all to (3) very interested/well informed.
We conducted a cross-sectional latent class analysis (LCA) at each of five waves and a
latent transition analysis (LTA) across waves using Mplus 7.1 (Muthén & Muthén, 1998-2012).
We validated class descriptions by examining whether levels of theoretically meaningful criteria
differed by Wave 1 class membership using the stepwise distal outcome method.
LCA is a confirmatory, person-centered analytic strategy that determines groups of
individuals based on their pattern of scores on a given set of categorical variables (Nylund,
2007). Each class is defined by the likelihood that individuals in the class will endorse each
variable (item probability parameters), and individuals are assigned a probability of belonging to
each class (class probability parameters). The number of classes is empirically determined using
fit indexes, such as the Bayesian Information Criterion (BIC; Nylund, Asparouhov, & Muthén,
2007). Starting with a one-class solution, models are estimated with increasingly more classes
TRANSITIONS IN CIVIC ENGAGEMENT ACROSS ADOLESCENCE 9
until there is no further model improvement (fit indexes show no substantive change or
additional classes are small, conceptually unclear, or slight variations on already identified
classes). LTA is a longitudinal extension of LCA that calculates individuals' patterns of
transitions between latent classes over time (Nylund, 2007). In addition to calculating item and
class probability parameters at each wave, LTA also produces transition probability parameters
for each individual, representing the likelihood of being in a specific latent class at time t given
membership in a specific latent class at the prior time point (t-1). Although each transition
probability only references two waves of data (i.e., t and t-1), these probabilities are combined to
identify an individual's most likely pattern of class membership across a series of waves. We
used chi-square tests with follow-up Bonferroni-corrected pairwise comparisons to analyze
prevalence of transition patterns overall and based on initial class membership. Given the school-
based design, intraclass correlation coefficients (ICCs) were calculated for all variables; all ICCs
were at .05 or below, indicating very little variance due to school membership. Thus, clustering
within schools was not necessary and was not used to avoid model complexity.
At each wave, an LCA on eleven dichotomous civic indicators indicated four classes.
Specifically, a four-class solution demonstrated the lowest BIC scores at each wave while
maintaining classes with substantive meaning and similar interpretations across waves. In some
waves, inclusion of a 5th class lowered BIC values slightly, however this class was inconsistently
present and too small (<5% of sample) to justify inclusion. Thus, a four-class model was
specified for LTA analyses with probability parameters constrained to be equal across waves.
These constraints ensure that class meanings were identical across waves, increasing the
interpretability of class transitions (Meeus et al., 2011); here it also enabled model convergence.
TRANSITIONS IN CIVIC ENGAGEMENT ACROSS ADOLESCENCE 10
Latent Class Descriptions
--- Figure 1 here ---
Figure 1 displays item probabilities for each latent class and Figure 2 indicates the
prevalence of each class across waves. Class 1 can be considered Civic Sympathizers. Similar to
the sympathizers identified by Voight and Torney-Purta (2013), these youth valued social justice,
staying current on issues, and helping others – but were relatively low on current behaviors and
expectations for future civic behaviors. They made up a relatively small proportion of individuals
(12% of participants across waves) and declined in prevalence over time. Class 2 represents
Unengaged individuals; they were relatively much lower across indicators, including civic
commitments, measured behaviors, and expectations for future engagement, although they were
moderate on expectations for future voting. This group was prevalent (39% of participants across
waves), although less so with age. Class 3 represents Civic Leaders; they showed the highest
levels of current values, future expectations, and civic behaviors compared to other groups. They
were also the only group that expected to run for office or be community leaders in the future
and were most likely to have run for student council in the prior year. This group was relatively
uncommon (14% of participants across waves) but become increasingly prevalent with age.
Finally, Class 4 was named Informed Future Voters; they expected to vote and be well-informed
in the future and were fairly likely to engage in reading the news, but interestingly did not value
being current on issues or holding high levels of other civic commitments. This group was
relatively prevalent (35% of participants across waves), especially later in high school.
--- Figure 2 here ---
Consistent with these definitions, the classes differed significantly in how informed they
were, χ2(3) = 393.73, p < .001; Civic Leaders were most informed (M = 16.62), followed by
TRANSITIONS IN CIVIC ENGAGEMENT ACROSS ADOLESCENCE 11
Informed Future Voters (M = 14.19), then Civic Sympathizers (M = 13.10), and then Unengaged
(M = 11.93), all ps < .001. Likewise, interest in social issues differed among the classes, χ2(3) =
1011.68, p < .001. Also as expected, Civic Leaders were more interested (M = 13.45) and
Unengaged less interested (M = 8.97) than all other classes, ps < .001. Informed Future Voters
(M = 11.13) and Civic Sympathizers (M = 10.93) did not differ in interest, χ2(1) = 1.52, n.s.
Class Transition Patterns
The four obtained classes differed qualitatively in their patterns but also in their general
levels of civic engagement. When describing trends of progressive and regressive civic
engagement over time, we refer to the Unengaged class as least engaged, Civic Sympathizers and
Informed Future Voters as moderately engaged, and Civic Leaders as most engaged. Equality
tests confirmed that classes differed in their overall civic engagement (a sum of all Wave 1 civic
engagement indicators), χ2(3) = 14762.20, p < .001. At Wave 1, Civic Leaders were more
engaged than all other classes, M = 5.91, ps <.001; Unengaged participants were less engaged
than all other classes, M = 0.55, ps <.001, and Civic Sympathizers and Informed Future Voters
did not differ in overall civic engagement, Ms = 2.83, 2.83, respectively, χ2(1) = .005, n.s. Table
1 presents the common class transition patterns present in the data and the proportion of
individuals demonstrating each pattern. The majority (62.5%) of individuals remained stable in
their class membership between 8th and 12th grade. However, stability varied by class, χ2(3) =
398.79, p < .001; stability was more common among individuals starting in the Unengaged,
Informed Future Voters, and Civic Leaders classes (approximately 70% in each) than for those
starting out as Civic Sympathizers (17%; pairwise p’s < .001).
--- Table 1 here ---
Among those transitioning, most individuals changed class membership once, with a
TRANSITIONS IN CIVIC ENGAGEMENT ACROSS ADOLESCENCE 12
small minority oscillating between two classes; only 5% of individuals showed any other pattern
of change. Supportive of the normative growth hypothesis, transitions towards greater civic
engagement were more common than transitions away from it (17.8% vs. 9.9%; χ2(1) = 68.25, p
< .001) and tended to occur later in high school than movement towards less engagement. This
general pattern is also evident in the increasing prevalence of Informed Future Voters and Civic
Leaders over time (see Figure 2). Chi square comparisons indicated that transition probabilities
into the Civic Leaders class were equal across 8th-12th grades, with only one exception (this
transition was less common in 11th than 12th grade, χ2(3) = 9.37, p < .05; pairwise p < .05).
Informed Future Voters and Civic Sympathizers classes both appeared to function as way-
points for change in civic engagement (see Table 1). That is, movement between these and other
classes was quite common. However, the roles of these classes were decidedly different;
individuals rarely (1%) transitioned between the two classes and the trajectories associated with
class membership significantly differed between the two, χ2(1) = 19.99, p < .001. Being an
Informed Future Voter tended to correspond with progressive change towards greater civic
engagement – 70% of Informed Future Voters who made a single transition moved out of the
lowest class or into the leadership class, whereas only 23% did the reverse, primarily regressing
from leadership. In contrast, being a Civic Sympathizer tended to correspond with decreasing
civic engagement – 50% of Civic Sympathizers who made a single transition moved down from
leadership or into the lowest class, most commonly the latter, whereas 37% did the reverse.
Likewise, individuals who increased their civic engagement by becoming Civic Sympathizers
appeared less committed to this moderately engaged status: 27% regressed back to the
Unengaged class before the end of 12th grade whereas only 1% of individuals who moved from
Unengaged to Informed Future Voters did the same, χ2(1) = 61.64, p < .001 (see Table 1).
TRANSITIONS IN CIVIC ENGAGEMENT ACROSS ADOLESCENCE 13
We set out to describe typologies of civic engagement and document developmental
transitions in civic engagement across adolescence to inform civic developmental theory.
Analyses identified four distinct typologies of civic engagement that replicated across five
waves. Our Unengaged group mirrors the uninvolved typology consistently found in other
studies (Finlay et al., 2011; Pancer et al., 2007; Voight & Torney-Purta, 2013), and the relatively
large size and stability of this group suggests that civic disengagement among U.S. youth is an
issue that merits greater attention. Three other typologies differed in their values, behaviors, and
future expectations; these groups had different levels of stability and different types of transitions
over time. Thus, our study highlights the utility of conceptualizing civic engagement holistically
as comprised of multidimensional indicators and of thinking about civic development as a series
We expected substantial civic change across adolescence given that dramatic changes and
transitions characterize this life period, yet a surprisingly large amount of ipsative continuity
emerged. Two-thirds of our sample was consistently categorized in terms of civic engagement
across 8th through 12th grades. Of course, despite being classified as stable, these youth could be
fluctuating in their levels of involvement within these categories, a question to be undertaken in
future research using measures of frequency and quality of participation and variable-oriented
analytic approaches (e.g., growth curves). Our stability findings suggest that person-oriented
approaches may capture moderately enduring individual differences in how youth think about
and engage with society. For example, some youth are drawn to full engagement, others focus on
staying informed, and still others remain unengaged across adolescence.
TRANSITIONS IN CIVIC ENGAGEMENT ACROSS ADOLESCENCE 14
Civic transitions were evident among one-third of our large national sample. Given
previous longitudinal evidence of episodic civic engagement among adults (e.g., Jennings &
Stoker, 2004), it is notable that gradual stepwise change best described the majority of
adolescents’ civic transitions. Indeed, the vast majority of youth who transitioned did so only
once over five years, and nearly all of these transitions were incremental. In other words, fleeting
experimentation and major civic transformations were both extremely rare in our sample. In
addition, upward or progressive transitions became more common later in adolescence,
providing modest support for the normative growth hypothesis that civic engagement increases
with age in correspondence with sociocognitive gains (Wray-Lake & Syvertsen, 2011). The
notion that civic changes are gradual could inform longitudinal research designs in this area and
developmental theories of how and why youth make these civic transitions. For instance, future
research could investigate which contextual supports and experiences as well as which individual
characteristics accumulate to influence progressive civic transitions.
Two distinct transition groups were identified that may signal divergent civic
developmental trajectories in progressive or regressive directions. First, Informed Future Voters
actively followed current events and envisioned themselves as well-informed, voting adults.
Importantly, once youth achieved this status, they either remained stable or progressed to full
engagement; downward movement to behaviorally inactive or uninvolved groups was very
uncommon. Some scholars conceptualize political information consumption as a measure of
civic knowledge (e.g., Schulz et al., 2010). Both civic knowledge and future civic expectations in
adolescence have been shown to predict greater civic involvement in adulthood (Finlay, Wray-
Lake, Warren, & Maggs, 2014; Galston, 2001). Indeed, in our study, knowledge and future
TRANSITIONS IN CIVIC ENGAGEMENT ACROSS ADOLESCENCE 15
thinking appeared to be a powerful positive combination: Beginning as an Informed Future Voter
was the only likely pathway to becoming a Civic Leader, suggesting that combining information
consumption with contemplation of future citizenship may be an important stepping stone in a
progressive civic development trajectory. No age trend was reliably associated with this
progressive transition, suggesting that youth who are Informed Future Voters could be inspired at
any time during adolescence to blossom into more fully engaged citizens.
Civic Sympathizers – youth placing high value on social justice, leadership, and helping
others, yet reporting few civic behaviors or future commitments – were the smallest and most
mobile group in our sample. This group is akin to the social justice sympathizers found by
Voight and Torney-Purta (2013), who were high on civic attitudes but low on behaviors.
Importantly, Civic Sympathizers who transitioned over time were most likely to regress to low
engagement, and jumping back and forth between civic sympathizing and low engagement was
also the most common oscillating pattern. These findings are intriguing because values have long
been considered motivational precursors to action, and forming socially responsible and justice-
oriented values are widely thought to underpin youth civic developmental processes (Flanagan,
2013; Wray-Lake & Syvertsen, 2011). We are certainly not suggesting that civic values are
unimportant, as our most behaviorally engaged group (Civic Leaders) also held high civic
values. Yet, our results suggest that youth with a profile of civic values without action merit
further study in terms of their potential for civic growth, as experimentation with values alone
may portend a regressive civic trajectory. Regressive transitions among this group could stem
from cognitive dissonance as a result of experiencing behavior-value discrepancy (Grube,
Mayton, & Ball-Rokeach, 1994) or from inferring changes in personal values based on their lack
of civic behaviors (Bem, 1972). These youth may be doing very well in other domains of life,
TRANSITIONS IN CIVIC ENGAGEMENT ACROSS ADOLESCENCE 16
however; Voight and Torney-Purta (2013) found their social justice sympathizers to also be the
highest academic achievers and have the lowest levels of school disciplinary referrals.
Limitations and Future Directions
The major drawback of our study is the inability to comprehensively measure civic
engagement; a clear measure of volunteering is most conspicuously absent but is an increasingly
prevalent civic behavior among U.S. youth (Syvertsen, Wray-Lake, Flanagan, Briddell, &
Osgood, 2011). The high prevalence of Unengaged youth in our sample, especially, should be
interpreted with caution as civic engagement could be underestimated overall. On the other hand,
our measures may have overestimated political information consumption, as youth could have
been reading non-political material in news magazines or newspapers. Testing whether the same
civic engagement typologies and transitions hold across gender, socioeconomic status, and
ethnicity are important next steps to building developmental theory that is inclusive across
groups. Moreover, similar analyses in other cultural contexts would determine the extent to
which conclusions generalize. Certainly, these analyses should be updated in a more
contemporary cohort to ensure generalizability, as youth in our sample are now adults and civic
engagement may change historically. Fruitful avenues for person-oriented developmental
research include identifying individual and contextual factors that predict stability versus
progressive or regressive transitions, and determining the implications of typology membership
and transition patterns for later civic engagement and functioning in adulthood.
In order to progress, civic developmental theory must address the necessary first question
of how much and what type(s) of change in civic engagement adolescents experience over time.
Our study informs civic developmental theory by taking a person-oriented methodological
TRANSITIONS IN CIVIC ENGAGEMENT ACROSS ADOLESCENCE 17
approach and thus illuminating substantial heterogeneity in adolescents’ patterns of change in
civic engagement. We found ipsative continuity for some adolescents, with stability varying by
type of civic engagement. Gradual incremental change characterized the transitions of other
adolescents, suggesting that the normative growth hypothesis only holds for a subgroup of youth
and thus may not be normative at all. Importantly, distinct transition points may signal which
pathway youth are likely to follow, as two moderately engaged groups in our study showed quite
different patterns of transitions. These findings, taken together, point to a novel and potentially
productive theoretical starting point for future longitudinal civic engagement research: That is,
recognizing that adolescents’ civic development is a person-specific process could bring us steps
closer to understanding how some people come to be actively engaged citizens.
TRANSITIONS IN CIVIC ENGAGEMENT ACROSS ADOLESCENCE 18
Amnå, E. (2012). How is civic engagement developed over time? Emerging answers from a
multidisciplinary field. Journal of Adolescence, 35, 611-627.
Bem, D. J. (1972). Self-perception theory. In L. Berkowitz (Ed.), Advances in experimental
social psychology, Vol. 6 (pp. 1-62). New York: Academic Press.
Boehnke, K., & Boehnke, M. (2005). Once a peacenik – always a peacenik? Results from a
German six-wave, twenty-year longitudinal study. Peace and Conflict: Journal of Peace
Psychology, 11, 337-354.
Caspi, A., & Roberts, B.W. (2001). Personality development across the life course: The
argument for change and continuity. Psychological Inquiry, 12, 49-66.
Eisenberg, N., Fabes, R. A., & Spinrad, T. L. (2006). Prosocial development. In W. Damon & R.
M. Lerner (Series Eds.), & N. Eisenberg (Vol. Ed.), Handbook of child psychology: Vol.
3. Social, emotional, and personality development (6th ed., pp. 646-718). New York:
Finlay, A. K., Flanagan, C. A., & Wray-Lake, L. (2011). Civic patterns and transitions over eight
years: The AmeriCorps National study. Developmental Psychology, 47, 1728-1743.
Finlay, A., Wray-Lake, L., Warren, M., Maggs, J.L. (2014). Anticipating their future: Adolescent
values for the future predict adult behaviors. International Journal of Behavioral
Flanagan, C. A. (2004). Volunteerism, leadership, political socialization, and civic engagement.
In R. M. Lerner, & L. Steinberg, (Eds.), Handbook of adolescent psychology (2nd ed.,
pp.721-745). Hoboken, NJ: John Wiley & Sons, Inc.
Flanagan, C. A. (2013). Teenage citizens: The political theories of the young. Harvard Press.
TRANSITIONS IN CIVIC ENGAGEMENT ACROSS ADOLESCENCE 19
Flanagan, C.A., & Faison, N. (2001). Youth civic development: Implications of research for
social policy and programs. Social Policy Report (Vol. 15, no.1). Ann Arbor, MI: Society
for Research in Child Development.
Galston, W.A. (2001). Political knowledge, political engagement, and civic education. Annual
Review of Political Science, 4, 217-234.
Grube, J. W., Mayton, D. M., II, & Ball-Rokeach, S. B. (1994). Inducing change in values,
attitudes, and behaviors: Belief system theory and the method of value self-confrontation.
Journal of Social Issues, 50, 153-174.
Hart, D., Donnelly, T.M., Youniss, J., & Atkins, R. (2007). High school community service as a
predictor of adult voting and volunteering. American Educational Research Journal, 44,
Kinder, D. R. (2006). Politics and the life cycle. Science, 312, 1905-1908.
Jennings, M. K., & Stoker, L. (2004). Social trust and civic engagement across time and
generations. Acta Politica, 39, 342-379.
Lerner, R.M. (1996). Relative plasticity, integration, temporality, and diversity in human
development: A developmental contextual perspective about theory, process, and method.
Developmental Psychology, 32, 781-786.
Masyn, K. (2013). Latent class analysis and finite mixture modeling. In T. Little (Ed.), The
Oxford handbook of quantitative methods in psychology (pp. 375-393). Oxford Press.
Meeus, W., van de Schoot, R., Keijsers, L., & Branje, S. (2012). Identity statuses as
developmental trajectories: A five-wave longitudinal study in early-to-middle and
middle-to-late adolescents. Journal of Youth and Adolescence, 41, 1008-1021.
Miller, J. D. (2014). Longitudinal Study of American Youth, 1987-1994, 2007-
TRANSITIONS IN CIVIC ENGAGEMENT ACROSS ADOLESCENCE 20
2009.ICPSR30263-v2. Ann Arbor, MI: Inter-University Consortium for Political and
Social Research [Distributor]. Retrieved from http://doi.org/10.3886/ICPSR30263.v2
Muthén, L. K., & Muthén, B. O. (1998-2012). Mplus user's guide (7th ed.). Los Angeles, CA:
Muthén & Muthén.
Neundorf, A., Smets, K., & García-Albacete, G. M. (2013). Homemade citizens: The
development of political interest during adolescence and young adulthood. Acta Politica,
Nylund, K. L. (2007). Latent transition analysis: Modeling extensions and an application to peer
victimization. (Doctoral dissertation, University of California, Los Angeles). Retrieved
Nylund, K. L., Asparouhov, T., & Muthén, B. O. (2007). Deciding on the number of classes in
latent class analysis and growth mixture modeling: A Monte Carlo simulation study.
Structural Equation Modeling, 14, 535-569. doi: 10.1080/10705510701575396
Overton, W.F., & Reese, H.W. (1981). Conceptual prerequisites for an understanding of
stability-change and continuity-discontinuity. International Journal of Behavioural
Development, 4, 99-123.
Pancer, S.M., Pratt, M., Hunsberger, B., & Alisat, S. (2007). Community and political
involvement in adolescence: What distinguishes the activists from the uninvolved?
Journal of Community Psychology, 35, 741-759.
Schulz, W., Ainley, J., Fraillon, J., Kerr, D., & Losito, B. (2010). ICCS 2009 international
report: Civic knowledge, attitudes, and engagement among lower-secondary school
students in 38 countries. International Association for the Evaluation of Educational
Achievement: Herengracht 487, Amsterdam, 1017 BT, The Netherlands.
TRANSITIONS IN CIVIC ENGAGEMENT ACROSS ADOLESCENCE 21
Sherrod, L. R., & Lauckhardt, J. (2009). The development of citizenship. In R. M. Lerner, & L.
Steinberg (Eds.), Handbook of adolescent psychology: Contextual inﬂuences on
adolescent development, Vol. 2 (3rd ed., pp. 372-408). Hoboken, NJ: Wiley & Sons.
Sherrod, L., Torney-Purta, J., & Flanagan, C. A. (Eds.) (2010). Handbook of research on civic
engagement in youth. Hoboken, NJ: John Wiley & Sons.
Smetana, J.G., & Villalobos, M. (2009). Social cognitive development in adolescence. In R. M.
Lerner, & L. Steinberg (Eds.), Handbook of adolescent psychology, Vol. 1: Individual
bases of adolescent development (3rd ed., pp. 187-228). Hoboken, NJ: Wiley & Sons.
Syvertsen, A. K., Wray-Lake, L., Flanagan, C. A., Briddell, L., & Osgood, D. W. (2011). Thirty-
year trends in U.S. adolescents’ civic engagement: A story of changing participation and
educational differences. Journal of Research on Adolescence, 21, 586-594.
Voight, A., & Torney-Purta, J. (2013). A typology of youth civic engagement in urban middle
schools. Applied Developmental Science, 17, 198-212.
Weerts, D.J., Cabrera, A.F., & Mejías, P.P. (2014). Uncovering categories of civically engaged
college students: A latent class analysis. The Review of Higher Education, 37, 141-168.
Westheimer, J., & Kahne, J. (2004). What kind of citizen? The politics of educating for
democracy. American Educational Research Journal, 41, 237-269.
Wray-Lake, L., & Syvertsen, A. (2011). The developmental roots of social responsibility in
childhood and adolescence. In C. Flanagan, & B. Christens (Eds.), Youth development:
Work at the cutting edge. New Directions for Child and Adolescent Development, 134,
Youniss, J., & Yates, M. (1999). Youth service and moral-civic identity: A case for everyday
morality. Educational Psychology Review. Special Issue: Moral Development in
Adolescents and Adults, 11, 361-376.
Number and Proportion of Participants Demonstrating Each Transition Pattern
Informed to Leaders
Informed to Unengaged
Informed to Sympathizers
Leaders to Informed
Leaders to Unengaged
Leaders to Sympathizers
Unengaged to Informed
Unengaged to Leaders
TRANSITIONS IN CIVIC ENGAGEMENT ACROSS ADOLESCENCE 23
Note. Transition patterns of 0.0% indicate a non-zero number that rounds to zero in the tenth decimal place; “–“ indicates no pattern is
present. Total N = 3,701.
Sympathizers to Informed
Sympathizers to Leaders
TRANSITIONS IN CIVIC ENGAGEMENT ACROSS ADOLESCENCE 24
TRANSITIONS IN CIVIC ENGAGEMENT ACROSS ADOLESCENCE 25
Figure 1. Likelihood of Individuals Endorsing each Indicator across Classes. Numbers can be
interpreted as percentage of individuals within a class that endorse each item.
Figure 2. Proportion of Participants in Each Class over Time
TRANSITIONS IN CIVIC ENGAGEMENT ACROSS ADOLESCENCE 27
Laura Wray-Lake, Ph.D. Assistant Professor in the Department of Clinical and Social Sciences
in Psychology at the University of Rochester. Her program of research focuses on civic
engagement and social responsibility development among youth, with an emphasis on the role of
social contexts in fostering youth civic development, inequalities in civic engagement, and
longitudinal research methodology.
Wendy M. Rote, Ph.D., Postdoctoral Associate in the Department of Clinical and social Sciences
in Psychology at the University of Rochester. Her research focuses on parenting, parent-child
relationships, and adolescent autonomy development with specific attention to the ways parents
and children differ in their perceptions of parental practices and relational behaviors.
Celina M. Benavides, M.A., Ed.M., Doctoral Student of Positive Developmental Psychology at
Claremont Graduate University, USA completed graduate degrees at Whittier College and
Harvard Graduate School of Education. Her research examines educational achievement and
civic engagement amongst diverse children and youth, settings that promote empowerment, and
the building of assets through community and school activities.
Christine Victorino, Ph.D., Assistant Vice Provost at University of California at Riverside. Dr.
Victorino oversees program evaluation, academic assessment, undergraduate research, and
educational initiatives. Her academic research focuses on issues of faculty and student diversity,
campus racial climate, and civic engagement in higher education. Victorino completed her Ph.D.
(Education) at UC Santa Barbara.