What Does It Mean to Assess Identity Development?
Assessing identity development is the systematic study of how people explore, establish, evaluate, reconsider, narrate, and coordinate commitments concerning who they are and how they wish to live. A defensible assessment identifies a particular dimension of identity, specifies the life domain and time scale to which it applies, selects a measure whose interpretation has been supported in the relevant population, and examines the evidence for change. A questionnaire score, an interview classification, and a coded life story provide different kinds of evidence. Taken together, they can illuminate developmental processes that any one approach captures incompletely.
The expression identity development assessment refers primarily to psychological research and appropriately qualified applied assessment. It does not designate a medical test, clinical diagnosis, universal score of identity maturity, or standardized developmental examination required of adolescents. The central research question is what a person's reported or observed identity processes reveal within a defined model and context. The practical question is whether the selected evidence is strong enough for the conclusion being drawn. The distinction is consequential: a person can report uncertainty while exploring meaningful possibilities, and an apparently stable commitment can be rigid, unexamined, or constrained by circumstance. The value of an assessment depends on interpretation rather than the impression that certainty is inherently healthier.
Contemporary identity research draws on the developmental tradition associated with Erik Erikson and James Marcia, multidimensional process models, and narrative approaches to autobiographical meaning. Marcia's original interview-based identity-status framework introduced the examination of exploration and commitment; later instruments differentiated commitment making from identification with commitments, exploration in breadth from exploration in depth, and reflection from ruminative searching. Narrative researchers examine how people connect life events to a developing sense of self. These methods answer related but distinct questions Marcia (1966), Luyckx et al. (2006), McAdams (2001).
Identity development as a multidimensional research construct
A person's identity includes commitments and experiences in domains such as education, work, relationships, values, religion, community, culture, and social roles. The salience of these domains changes across the life course and between social environments. Developmental identity research is concerned with the processes through which individuals engage with these domains, their capacity to make and revise commitments, and the meanings attached to these decisions. Assessments should therefore distinguish the content of an identity commitment from the process through which it is reached and sustained. Two people may express the same career aspiration while differing sharply in the amount of exploration, felt ownership, available alternatives, and stability of commitment.
Identity development is also distinguishable from adjacent psychological constructs. Self-concept represents organized self-beliefs; self-esteem evaluates personal worth; self-concept clarity concerns the perceived consistency and definition of self-beliefs; identity status summarizes configurations of exploration and commitment; narrative identity concerns autobiographical meaning and continuity. Instruments targeting one construct cannot be treated as interchangeable measures of the others. For the developmental framework itself, see Identity Development: Exploration, Commitment, and Change Across Life; for a separate explanation of classification, see Identity Status Theory: Achievement, Moratorium, Foreclosure, and Diffusion.
What an assessment can legitimately conclude
A well-designed identity study may estimate the strength of a commitment, the degree of active exploration, the pattern of reconsideration, the coherence or agency expressed in autobiographical narratives, the likelihood of belonging to a statistical profile, or change in these features over repeated assessments. It can also test associations with functioning, relationships, educational adaptation, and reported distress. These conclusions remain conditional on the model, instrument, sample, language, comparison group, measurement quality, and study design. The result should be expressed as a finding about measured psychological processes in a context, with uncertainty proportional to the available evidence.
Theoretical Frameworks That Determine What Is Measured
Eriksonian theory and Marcia's identity statuses
Eriksonian developmental theory treated identity as an evolving psychosocial achievement involving self-continuity and social recognition. Marcia translated part of this broad account into a research procedure focused on two criteria: exploration of alternatives, originally often termed crisis, and commitment. Their combination produced the familiar statuses of achievement, moratorium, foreclosure, and diffusion. The original research used interviews in a restricted sample of college-age men, making the historical context of validation important when considering contemporary use across genders, ages, and cultures Marcia (1966).
Identity-status classifications are useful summaries of answers to specific questions, not natural categories established independently of the instruments that generate them. Achievement describes an explored commitment under a defined model; moratorium indicates exploration without settled commitment; foreclosure indicates commitment with limited exploration; diffusion indicates weak commitment and exploration according to the operational thresholds used. A respondent may show different patterns across education, relationships, faith, and occupation. A global label can conceal those differences, and a status recorded at one interview may change. An instrument cannot infer an enduring personality type or psychiatric condition from the status designation.
Dual-cycle models of formation and evaluation
The four-dimensional model differentiated commitment making, identification with commitment, exploration in breadth, and exploration in depth. An extension added ruminative exploration, distinguishing persistent, distress-associated indecision from active consideration of possibilities. This distinction matters because the ordinary word exploration covers behaviors with different emotional and functional associations. A high overall exploration score is ambiguous when the underlying items combine information seeking, reflection, and repetitive worrying Luyckx et al. (2006), Luyckx et al. (2008).
The Utrecht tradition models identity as cycles of commitments, in-depth exploration of current commitments, and reconsideration when alternatives become salient. This approach emphasizes the maintenance and revision of commitments already present, rather than assuming that development consists of a single choice followed by permanent completion. The resulting Utrecht-Management of Identity Commitments Scale, or U-MICS, is especially informative in studies examining specific life domains repeatedly Crocetti, Rubini, and Meeus (2008), Crocetti and Meeus (2026).
Identity styles and processes of self-revision
Identity-style approaches ask how an individual tends to handle identity-relevant information. Informational, normative, and diffuse-avoidant approaches concern the ways people search, evaluate, conform to expectations, or defer decisions. They address characteristic processing strategies rather than the same dimensions as identity-status scales. A person with firm commitments can still approach new evidence thoughtfully; another may embrace a social norm while managing other decisions differently. Measurement choices should reflect the distinction between the process of handling information and the content or firmness of one's current commitments Berzonsky (1992).
Narrative identity and autobiographical meaning
Narrative approaches examine identity through personal stories rather than closed response options. Participants may describe turning points, difficult experiences, important relationships, or imagined futures. Researchers study how these accounts express continuity, change, agency, communion, autobiographical reasoning, and connections between events and personal values. The narrative tradition contributes information about meaning and context that process scales typically omit. Because stories are told to someone, at a particular time, under particular cultural conventions, narrative findings are also shaped by the elicitation setting and interpretive framework McAdams (2001), McLean, Pasupathi, and Pals (2007).
Structured Identity Interviews
The identity-status interview: questions, domains, and coding
A structured or semistructured identity-status interview asks respondents to explain commitments and the ways they considered alternatives. In educational identity, an interviewer might ask what direction the participant has chosen, what other paths were considered, which influences mattered, and what would lead them to reconsider. In ideological, interpersonal, or occupational domains, equivalent questions must reflect the actual choices available to the respondent. Interviewers examine the reasoning and developmental history behind an answer rather than merely checking whether a commitment exists. Status coding is performed using an explicit manual and documented decision rules Marcia (1966).
Interview methods offer depth, flexibility, and the opportunity to explore contradictions across contexts. They can reveal an apparent commitment that is still tentative, a family expectation presented as a personal decision, or a period of exploration shaped by constrained opportunities. Their limitations include training demands, participant burden, interviewer influence, culturally situated assumptions, memory errors, and the need for reliable coding. The original classification can be difficult when a narrative combines domains or contains evidence supporting more than one status. Studies should report the interview format, coder qualifications, whether coding was performed independently, the level of agreement, and procedures for resolving disagreement.
Avoiding a premature identity-status label
When a participant provides conflicting evidence, an interviewer's first task is to document the conflict. A person may show occupational achievement but relationship moratorium, or may make a commitment after prolonged reflection while currently reconsidering it. Coders should be permitted to retain domain-level results, distinguish historical exploration from ongoing exploration, and state that an overall classification is uncertain. Forcing ambiguous narratives into a single category can create precision on paper while reducing substantive validity.
Interviewers must also account for unequal opportunities for exploration. Economic insecurity, displacement, family obligations, discrimination, legal restrictions, and restricted educational access affect which alternatives can realistically be tried. An instrument that asks whether participants pursued options may conflate their developmental processes with opportunity structures. A credible interview records both personal choices and external constraints before interpreting the absence of reported exploration.
Questionnaires and Rating Scales
Ego Identity Process Questionnaire: separate exploration and commitment
The Ego Identity Process Questionnaire (EIPQ) was developed to provide separate continuous self-report scores for exploration and commitment. Its original 32 items covered occupation, religion, politics, values, family, friendships, dating, and sex roles, with initial validation in college samples. Continuous process scores preserve some information that categorical status assignments compress. Nevertheless, domain coverage and item language reflect the historical context of the original measure, and the scale should not be transplanted into new populations without evaluating content relevance and psychometric performance Balistreri, Busch-Rossnagel, and Geisinger (1995).
The EIPQ helps answer how much exploration and commitment a participant reports within its operational scope. It does not directly measure narrative coherence, clinical identity disturbance, or the quality of each identity commitment. Researchers deriving statuses from questionnaire thresholds must identify the algorithm and report how alternative cutoffs affect classification. An apparently objective scoring rule may reflect a particular sample's distribution rather than universal developmental boundaries.
Extended Objective Measure of Ego Identity Status
The Extended Objective Measure of Ego Identity Status, including revised forms, translated status concepts into questionnaire responses spanning ideological and interpersonal concerns. The revision by Bennion and Adams specifically addressed limitations in interpersonal items, illustrating that broad status instruments can perform differently across domains. Before selecting any EOM-EIS form, researchers must identify the exact version, item content, scoring manual, normative context, and evidence of reliability for their sample. Scores and categories from different revisions should never be pooled as though the instruments were equivalent Bennion and Adams (1986).
Dimensions of Identity Development Scale
The Dimensions of Identity Development Scale (DIDS) operationalizes multiple dimensions: commitment making, identification with commitment, exploration in breadth, exploration in depth, and ruminative exploration. The differentiation is analytically valuable because ongoing information seeking and repetitive indecision may have different correlates. The DIDS is particularly suited to questions about broad future orientation and the balance between exploration and commitments. It is less well suited to detailed claims about the content of a specific social identity unless researchers add domain-focused measures. Studies must verify how translated versions reproduce the expected factor structure rather than assume that the original dimensional solution holds everywhere Luyckx et al. (2008).
DIDS scores should be treated dimensionally before considering any person-centered classification. Identifying a cluster with high ruminative exploration and low commitment in one sample does not justify labeling every similar respondent in another population without testing profile replication. Item wording, response scales, recruitment, and culture can change how profiles emerge. Claims about developmental improvement require longitudinal data, not a one-time comparison of scores for people of different ages.
Utrecht-Management of Identity Commitments Scale
The U-MICS estimates three processes: commitment, in-depth exploration, and reconsideration of commitment. It is often used with domain-specific items about education, interpersonal relationships, or other salient commitments. Commitment captures the presence and perceived significance of current choices, in-depth exploration concerns active reflection and discussion of those choices, and reconsideration concerns evaluating alternatives to them. The original development research involved 1,952 younger adolescents and tested its three-factor structure across relevant groups Crocetti, Rubini, and Meeus (2008).
A 2026 systematic review of the U-MICS examined 194 empirical quantitative references after full-text screening and documented its use across many contexts and domains, including translations and longitudinal applications. The authors discuss factor structure, measurement invariance, classification, and research designs, providing a particularly useful resource for investigators who need to evaluate rather than simply administer the instrument. Importantly, widespread use does not guarantee that every language, domain, and group comparison has been validated. Local evidence and the study's actual measurement objective remain decisive Crocetti and Meeus (2026).
Identity-style inventories and construct overlap
Identity-style inventories are relevant when the question concerns how people cope with identity-relevant decisions. Their information-processing focus may complement status or commitment scales, especially where uncertainty and decision making are under study. Yet a self-report response about a usual style is exposed to the same limitations as other trait questionnaires: social desirability, retrospective reconstruction, language comprehension, and limited sensitivity to rapid within-person fluctuation. Conceptual overlap with coping, reflection, and avoidance must be addressed through discriminant-validity evidence Berzonsky (1992).
Comparing instruments without treating them as synonyms
Selection follows the research question. An identity-status interview is appropriate when a project needs contextual explanations of how a commitment was formed. The EIPQ provides relatively broad exploration and commitment ratings. The DIDS separates different components of exploration and commitment at a more detailed conceptual level. The U-MICS focuses on the regulation and reconsideration of commitments within domains. Identity-style inventories assess characteristic approaches to identity-related information. These instruments are connected by shared developmental theory, yet their numerical scores refer to different operational constructs. Their correlation establishes relatedness and cannot alone establish equivalence.
Researchers should report the version and language of each scale; the exact sample and recruitment setting; item response options; missing-data treatment; whether subscale scores were calculated in the prescribed manner; and any adaptation from original instructions. When a questionnaire is used to infer individual identity status, the classification rule and its stability under alternative rules should be reported separately from the raw scale result. Contemporary methodological reviews emphasize that important cognitive, emotional, behavioral, temporal, and contextual aspects of identity can remain outside the content of conventional exploration-and-commitment items de Moor and Johnson (2026).
Narrative Interviews and Life-Story Methods
Life-story interviews and narrative prompts
Narrative identity research elicits accounts of personally meaningful scenes, relationships, transitions, and turning points. A life-story interview may ask for a high point, low point, significant challenge, formative moment, or imagined future chapter. Another design invites a focused account of an identity-relevant event, such as a move, an educational choice, a change in belief, or a conflict between family expectations and personal values. Both formats study how people represent continuity and change, but the breadth of material differs: a full life story emphasizes a person's attempt to organize a life; event narratives isolate particular episodes that researchers can examine across participants McAdams (2001).
Narrative elicitation needs explicit prompts, consistent procedures, informed consent, and care in handling sensitive autobiographical material. An open question creates different material from a highly guided task; spoken interviews produce different discourse from written responses. Participant language, literacy, comfort with autobiographical disclosure, cultural conventions, and the presence of an interviewer all shape the result. Researchers should record what was asked, how much time respondents received, whether follow-up questions were permitted, and whether narratives were audio recorded, transcribed, translated, or edited before coding.
Meaning-making, agency, communion, and coherence
Narrative coders may study autobiographical reasoning: the explanation a narrator gives of how an experience affected beliefs, priorities, or understanding of self. Agency concerns experienced influence and capacity to direct a life, while communion concerns connections, care, and relationships. Coherence can refer to temporal ordering, causal explanation, or thematic integration; these forms must be operationalized separately. Some studies code redemption and contamination sequences, emotional tone, or the complexity of reflection. The unit of analysis may be an event, a passage, a whole life story, or the individual across multiple stories. A coder cannot assume that a more polished story reflects a more developed identity without considering linguistic ability and sociocultural narrative conventions McLean, Pasupathi, and Pals (2007).
Longitudinal narrative evidence can capture changing self-interpretations. In a study of adults entering psychotherapy, repeated coded narratives showed that changes in agency preceded associated improvements in mental health within the observed design. This is a valuable example of using temporal ordering, but it does not establish that improving narrative agency causes clinical recovery. The study's sample, repeated-task context, and concurrent treatment limit the scope of generalization Adler (2012).
Reflexive thematic analysis and qualitative interpretation
Qualitative thematic analysis identifies patterns of meaning across accounts, often attending to contradictions, social power, cultural identity, and context. Reflexive approaches require researchers to articulate their theoretical assumptions and interpretive role. Their credibility is evaluated through transparency, analytic depth, grounding in data, and the fit between research aims and the chosen approach. They should not be judged solely by numerical coder agreement when the approach explicitly treats interpretation as situated rather than as an attempt to recover one objective code. A 2026 methods paper outlines how reflexive thematic analysis can represent identity as dynamic, relational, and intersectional while requiring disciplined reflexivity Watson, Gutierrez, and Azmitia (2026).
The difference between coding reliability and interpretive validity
When a narrative study uses a fixed codebook, independent coders and agreement statistics help determine whether the scheme can be applied consistently. Agreement is necessary for replicable quantitative content coding but does not demonstrate that the categories have psychological validity. For interpretive qualitative studies, the criterion may be the strength and transparency of the analytic argument rather than frequency counts. Researchers should name their epistemological approach, distinguish codebook coding from reflexive thematic analysis, and avoid presenting them as identical techniques. Translation and automatic transcription add further interpretive decisions that require documentation.
Social, Cultural, and Domain-Specific Identity Measures
Educational, occupational, and relational identities
A domain-specific measure targets a clearly identified area of life. Educational identity may concern a student's relationship to an educational pathway, while occupational identity involves commitments about work, professional roles, and vocational values. Interpersonal identity may address romantic relationships, friendship commitments, or family roles. The same scale structure can produce different developmental patterns across these domains because opportunities for choice and their subjective importance differ. A student's high educational commitment during a constrained school system is not directly comparable to a working adult's strong occupational commitment in a labor market with multiple alternatives. Domain selection is therefore a theoretical decision requiring explicit justification.
A multi-domain design should ask respondents how central each domain is to their sense of self and which decisions are currently live. It may be reasonable to exclude a domain that is irrelevant to participants, but comparisons then become limited to the remaining shared content. Researchers should avoid silently combining domains with different numbers of items into an undifferentiated identity-development score. Cross-domain patterns are themselves a meaningful finding: identity can be stable in one role while another role is undergoing active reconsideration. These complexities are recognized in the broad applications of the U-MICS literature Crocetti and Meeus (2026).
Ethnic-racial identity and culturally situated meanings
Ethnic-racial identity measurement addresses the meaning, exploration, affirmation, and commitment associated with membership in ethnic or racialized communities. These processes are influenced by cultural histories, discrimination, migration, family socialization, and social classification. Measures such as the Multigroup Ethnic Identity Measure–Revised (MEIM-R) were developed for questions specific to ethnic identity and require a validation logic different from general future-oriented identity instruments. Comparisons across groups are especially vulnerable to differences in how labels, group membership, and ethnic belonging are understood Phinney and Ong (2007).
Researchers should state what identity construct is being measured and whose understanding of membership guides item wording. Racial identity, ethnic identity, national belonging, cultural practices, and attitudes toward a group's place in society can overlap without being interchangeable. Survey questions about exploration can function differently for participants whose group identity is constantly made salient by social institutions than for participants who can choose when to engage with that identity. Interpretation benefits from cognitive interviews with members of the intended population, participation in measure design, and comparative evidence that items work as intended.
Recent research illustrates both innovation and continuing limitations. A 2026 scale for retrospective perceptions of ethnic-racial identity stability and change reported preliminary evidence for proposed factors but did not establish measurement invariance across its analytic groups. Accordingly, its results are informative for research within its intended context while group mean comparisons remain unwarranted. Such careful reporting is an example of the difference between a promising new measure and a validated universal comparison tool Ho and Tang (2026).
Gender, sexuality, religion, community, and intersectionality
Identity development can also be investigated through gender, sexual, religious, political, community, professional, and disability-related identification, depending on the project's objective. These domains have their own conceptual traditions and ethically relevant contexts. Applying a general exploration-and-commitment scale to them may capture only a narrow portion of lived identity. Some identities are made salient by external classification; others become central through affiliation, belief, personal meaning, or a developing relationship to community. Research designs should not presume that every identity unfolds through a common sequence of discovery followed by fixed commitment.
Intersectional identities pose additional questions about how roles and social positions interact. Participants may experience particular combinations of group membership as inseparable in everyday life, whereas separate subscale scores treat them as distinct. Qualitative interviews, social identity mapping, and participant-defined categories can provide complementary evidence. A 2026 methodological comparison highlighted how different research tools reveal different aspects of identity multiplicity and contextual movement; because the comparison was exploratory and used a specific student sample, it serves as a methodological demonstration rather than proof of a best universal instrument Mancini and Imperato (2026).
Psychometric Foundations of Responsible Interpretation
Reliability: consistency for a specific purpose
Reliability indicates the consistency or precision of measurements under defined conditions. Internal consistency examines relations among items intended to reflect the same dimension. Test–retest reliability concerns stability across administrations, though genuine development between time points can reduce observed stability without indicating a defective instrument. Interrater reliability is relevant when interviews or narratives are scored by independent coders. These are different properties, and no single reliability coefficient establishes the suitability of a measurement procedure for every use.
Researchers should report reliability estimates for the actual analytic sample, ideally with uncertainty intervals, and consider whether assumptions of a chosen estimate are plausible. An apparently high coefficient may reflect redundant items rather than comprehensive coverage. For short subscales, coefficients may be more sensitive to item count. If respondents are changing, a low test–retest coefficient can reflect both random error and meaningful variation; a design that seeks change should separate these sources through repeated observations, latent-variable models, and sensitivity checks where possible.
Content validity and the choice of construct boundaries
Content validity requires correspondence between the intended construct and the items or prompts used to represent it. Questions about career plans assess only part of identity development when the study's theory includes relationships, values, cultural belonging, and life stories. Conversely, broad questions about satisfaction with one's life may drift away from identity formation into well-being. The researcher must decide whether the instrument is intended to measure exploration behaviors, reflective thinking, emotional identification, degree of commitment, or self-evaluative reactions, and then inspect the items for alignment.
Good validation begins with a construct map, literature review, input from domain experts and intended respondents, and cognitive interviewing to examine what participants understand. Some items that appear equivalent to researchers can be experienced differently across educational, cultural, and linguistic settings. A measure's historical use is relevant evidence but cannot replace scrutiny of whether its items still capture current experiences. The 2026 recommendations by de Moor and Johnson emphasize better coverage of the cognitive, affective, behavioral, temporal, and contextual character of exploration and commitment de Moor and Johnson (2026).
Structural validity and factor analysis
Confirmatory factor analysis can evaluate whether observed response patterns are reasonably compatible with a predicted factor structure, such as U-MICS commitment, in-depth exploration, and reconsideration. Exploratory factor analysis is useful when a theoretically expected structure is uncertain, but investigators should ideally test exploratory findings in a separate sample rather than treating the same data as both discovery and independent confirmation. Factor correlations help assess whether dimensions are distinguishable, although statistical separation does not automatically imply psychological independence.
Acceptable numerical fit alone is insufficient. Researchers should inspect item loadings, residual patterns, theoretically plausible alternative models, sample size, category distributions, and response format. Different versions of an instrument can yield different structures. In a 2023 study of a Persian DIDS adaptation, the authors reported a six-factor solution in their sample rather than simply assuming perfect transfer of the original five-factor representation. The finding illustrates why version, culture, and local factor evidence must be disclosed whenever scale scores are interpreted Salehi Yegaei et al. (2023).
Measurement invariance across groups
Measurement invariance asks whether a measurement model functions sufficiently similarly for the comparisons researchers intend to make. Configural invariance concerns the overall pattern of dimensions; metric invariance concerns the comparability of factor loadings; scalar invariance concerns item intercepts or thresholds, often important for comparing latent means. The exact tests depend on whether responses are treated as continuous or categorical and on the estimation procedure. A study should assess the relevant level of invariance before interpreting group differences as differences in underlying identity processes.
Failure to establish scalar invariance makes direct latent-mean comparisons problematic. It does not mean that the measure is useless for studying within-group associations or exploratory patterns. Partial invariance may sometimes support a narrowly specified comparison, but the assumptions and affected items must be reported. Researchers should avoid the common error of assuming that translation, a satisfactory alpha, or broadly similar factor labels establishes cross-cultural equivalence. The 2026 U-MICS review discusses invariance testing and the spread of adaptations across settings; emerging specialized measures can be even more limited in their comparative claims Crocetti and Meeus (2026), Ho and Tang (2026).
Longitudinal measurement invariance
A repeated survey can show that a score changed while leaving open whether the meaning of its items changed. Longitudinal invariance tests address whether the measurement scale operates comparably across waves. This is especially important across transitions such as leaving school, entering work, migration, or changes in partnership, when respondents may interpret commitments differently. A shift in how a question is understood can resemble developmental change. Growth models applied without checking measurement properties can magnify this error by treating a changing instrument as a fixed ruler.
Response shift is a related concern: individuals may redefine what commitment, certainty, or exploration means to them over time. Qualitative follow-ups and cognitive interviews can clarify whether changes in score reflect changes in the trait of interest, reinterpretation of the response scale, or altered relevance of a life domain. Longitudinal researchers should preserve wording when feasible while documenting necessary contextual adaptations and their implications.
Convergent, discriminant, and criterion evidence
A valid identity-process instrument should relate to relevant external constructs in theoretically plausible ways and retain distinguishable meaning from adjacent measures. Commitment may correlate with self-concept clarity or well-being, but that does not make commitment a substitute for either construct. Ruminative exploration may correlate with distress, but a distress association does not convert the scale into a clinical screening tool. Cross-method convergence between interviews and surveys may support the inference that both tap related processes, yet imperfect correlations are expected because the methods represent different behavioral and meaning-based information.
Interpretation should be supported by multiple evidence streams: factor structure, response processes, longitudinal prediction, correlations with neighboring constructs, and replication across populations. No single correlation, classification accuracy estimate, or factor-fit index is a certificate of validity. Instrument validity refers to evidence supporting an intended interpretation of scores in a particular use, and the required evidence becomes more demanding as the stakes of the decision increase.
Measuring Change Across Time
Cross-sectional age differences versus development
Cross-sectional research compares respondents of different ages at one point in time. Such comparisons can reveal age-related patterns but confound age with cohort, historical context, education, socioeconomic conditions, and sampling. A higher mean commitment score among older respondents does not demonstrate that each participant naturally becomes more committed as they age. Development is a within-person process; claims about trajectories require evidence gathered over time or appropriately designed developmental methods.
Research reviewing longitudinal studies has found a broad pattern of increasing organization in some aspects of personal identity across adolescence while also documenting substantial stability and domain differences. The literature does not justify a rigid universal sequence in which everyone moves through the same statuses at the same ages. Identity development can involve return, revision, and continued exploration in adulthood Meeus (2011).
Panel studies and repeated measurement
A longitudinal panel follows the same respondents across several waves. It can estimate within-person change, stability in relative standing, transitions among patterns, and associations between identity processes and later outcomes. Design quality depends on sufficient waves, theoretically appropriate intervals, participant retention, a suitable reference period for each item, and analysis that recognizes repeated observations are not independent. Years may be appropriate for educational trajectories while days or weeks can be more informative for identity fluctuations during an acute transition.
Two-wave studies can estimate simple change but are limited for modeling nonlinear trajectories and developmental cycles. Additional waves enable more nuanced questions about commitment consolidation, reconsideration, and possible returns to exploration. They also create new threats: selective dropout, practice or memory effects, changes in questionnaire meaning, and wave-specific contextual disruptions. The analytic model should follow the substantive question rather than the availability of sophisticated software.
Growth curves and trajectories
Latent growth models estimate average change and individual differences in initial levels and rates of change. Nonlinear models may be justified when theory suggests acceleration, stabilization, or reversal. Researchers should check whether the observed time points support such complexity, whether measurement invariance is plausible, and whether assumed functional forms fit the trajectories. A favorable group-average slope can conceal important subgroups or periods of instability; plotting individual trajectories and evaluating heterogeneity can reveal patterns missed by aggregated models.
Measurement error, missing data, and selective attrition must be investigated. Participants experiencing distress, migration, or social disadvantage may be especially likely to leave longitudinal research, biasing the observed trajectory. Missing-data methods rely on assumptions that should be stated and tested in sensitivity analyses. A model's precision depends on those assumptions as well as sample size, and a statistically significant slope does not by itself indicate an important developmental change.
Cross-lagged models and causal interpretation
Cross-lagged panel models examine whether identity measures predict later values of another construct after adjusting for earlier observations. They can support hypotheses about temporal association, but many conventional models combine stable between-person differences with within-person fluctuation. Random-intercept variants attempt to separate these sources under their own assumptions. Neither approach by itself removes unmeasured confounding, measurement error, reverse pathways, or the influence of common environmental events. Associations involving identity, self-esteem, and depression are especially sensitive to these issues.
A systematic review and meta-analysis of longitudinal work on identity synthesis or confusion and social-emotional symptoms found that narrative synthesis suggested possible bidirectional relations, while meta-analyses of a smaller subset did not yield significant pooled associations for anxiety or depression. Included studies varied in quality and measurement. The correct lesson is that identity and mental health can be meaningfully studied together while precise effect estimates and causal interpretations require care Potterton et al. (2022).
Person-centered profiles, latent transitions, and classification
Person-centered analyses identify groups of participants who share patterns of scores. Cluster analysis and latent profile analysis can produce useful representations of combinations, such as strong commitments with low reconsideration or high exploration with unresolved commitments. A latent transition model can examine movement between profiles across observations. Nevertheless, profiles are estimated from a particular sample, set of indicators, scaling decisions, and model-selection criteria. Different choices can change the number and shape of the profiles.
Researchers should report how many candidate solutions were examined, how the final model was selected, classification uncertainty, sample size per profile, and the stability of findings across reasonable alternatives. A posterior probability indicates uncertainty about profile membership; it does not make a profile a natural diagnostic category. Replication across cohorts and validation of predictive meaning are preferable to giving profiles memorable labels without checking their substantive interpretation. The U-MICS literature explicitly discusses the appropriate use of process scores in person-centered identity classification Crocetti and Meeus (2026).
Experience sampling and intensive longitudinal designs
Experience-sampling and diary methods collect short repeated reports in participants' everyday contexts. Such designs can examine identity-relevant certainty, belonging, exploration, affect, and role salience as they vary within a person over days or weeks. They are appropriate when the question concerns immediate contextual influences or short-term fluctuations rather than lifetime developmental status. Prompt frequency, item brevity, missing responses, participant fatigue, reactivity, and privacy become central design issues.
Repeated reports of situational uncertainty should not automatically be interpreted as evidence of enduring identity diffusion. An individual may feel confident about broad commitments while questioning a specific decision after receiving new information. Multilevel analysis can distinguish within-person departures from a person's typical score from stable differences between participants, although the measure still requires evidence that its short-form items represent the intended situational construct.
Integrating Quantitative and Qualitative Evidence
Mixed-method designs answer complementary questions
A strong mixed-method study can combine process scales, interviews, and narrative analysis. A questionnaire may indicate high reconsideration of educational identity, while an interview reveals whether this reflects intellectual exploration, family pressure, a forced relocation, or a lack of available educational options. Narrative coding may show how the participant connects the decision to personal values and future aspirations. These findings need not be collapsed into a single score. Their joint value lies in explaining convergence, disagreement, and contextual meaning.
Integration can occur through purposive qualitative sampling from different quantitative profiles, longitudinal interviews following scale changes, or joint displays that organize scores and narrative themes without erasing their differences. The study should specify which result has priority for each inference, how discrepant cases are handled, and whether qualitative data were collected before or after researchers knew participants' scale scores. Otherwise, interviews risk becoming illustrations selected to confirm the preferred quantitative result.
What disagreement between methods can reveal
Discordance is often theoretically informative. A participant may endorse commitment items strongly because a decision has already been made, yet tell a narrative dominated by doubt or coercion. Another may have low questionnaire certainty but a coherent autobiographical account of deliberate exploration. The divergence may reflect scale limitations, different time horizons, response bias, genuine complexity, or an error in data collection. Method triangulation should investigate these alternatives rather than assuming one method has privileged access to the participant's true identity.
Digital traces, social identity mapping, and computation
Researchers increasingly study digital self-presentation, participation networks, and narrative language. Online profiles and posts can reveal situated representations of self; network mapping can show affiliations and role connections; computational text methods can extract patterns from large corpora of open-ended responses. Each observes behavior or language in a particular context and should be interpreted within the data-generating process. Public-facing self-presentation is influenced by audience expectations, platform design, moderation, and social rewards. A change in digital behavior is therefore evidence of changed online expression, not sufficient proof that a person's psychological identity has changed.
Computational text analysis and machine learning may help researchers organize large collections of open-ended identity narratives, but an algorithmic output must be treated as an interpretation of linguistic patterns rather than an independent determination of a person's identity. Applications require external validation, scrutiny of data leakage and subgroup performance, attention to cultural and linguistic context, and protection of consent and privacy. These are proposed methodological safeguards, not established evidence of high-stakes individual assessment accuracy. Watson et al. (2026) emphasize the interpretive commitments that remain essential in qualitative identity inquiry.
Research Design, Sampling, and Statistical Pitfalls
Sampling and representativeness
Identity studies frequently recruit students, adolescents in school systems, or convenience samples of emerging adults. Their findings may illuminate important processes while leaving unresolved how those processes apply to working adults, older populations, nonstudents, people experiencing displacement, and communities excluded by recruitment criteria. Samples limited to highly educated settings may overrepresent choices treated as voluntary and underrepresent identities shaped by economic obligation or structural barriers. Report demographic and contextual characteristics relevant to the research question rather than describing convenience samples as general population evidence.
Recruitment should consider variation in family structures, language proficiency, school access, migration histories, socioeconomic resources, community affiliations, and cultural contexts. Sampling across settings provides stronger grounds for generalization, but the same questionnaire may require adaptation and invariance testing. Large samples increase statistical precision for a narrow population; they cannot alone establish representativeness.
Differential item functioning and translation
An item may mean different things in different groups even when respondents share the same underlying level of identity commitment. For example, a question about personally selecting a career may reflect individual choice in one educational setting and family-level planning in another. Differential item functioning methods can examine whether an item behaves differently after conditioning on the underlying construct. Translation is best treated as an iterative process involving language experts, population representatives, cognitive testing, and statistical checks rather than literal word substitution.
Researchers should identify whether translated items capture equivalent temporal periods, roles, family obligations, and social labels. Back-translation can expose obvious discrepancies but does not establish construct equivalence. If meaningful adaptation changes a scale, the revised version must be described and psychometrically evaluated as such. These concerns are particularly pressing for cross-national comparisons and instruments concerning ethnicity, religion, gender, and contested social categories.
Social desirability, demand characteristics, and self-presentation
Identity is normatively charged. Participants may feel pressure to appear autonomous, decisive, reflective, loyal to a community, or confident about their future. Anonymous survey procedures and supportive interview settings can reduce some pressures, but response biases cannot be assumed away. Item phrasing that treats exploration as maturity or uncertainty as failure encourages evaluative answers. Questions should describe experiences without encoding the researcher's preferred developmental outcome.
Response-style differences can further complicate cross-group results: some participants favor extreme response categories, whereas others prefer middle options or agree with statements independent of their content. Balanced items, cognitive interviewing, appropriate models, and sensitivity analyses may help. No adjustment can recover data that were never meaningfully elicited, which is why item design remains more important than post hoc statistical correction.
Construct contamination and circular inference
If a scale includes distress, indecision, and dissatisfaction within its definition of identity difficulty, a correlation between that scale and depression is partly expected from overlapping item content. Researchers should inspect whether predictor and outcome measures ask essentially the same question under different labels. Similarly, including commitment satisfaction in an identity score can inflate apparent links with self-esteem. Discriminant validity and item-level review are therefore central to scientific interpretation.
Research teams must distinguish a developmental construct from a symptom checklist. Identity confusion can appear in normal developmental transitions; clinically significant disturbances have different assessment frameworks, thresholds, and functional implications. The findings of identity questionnaires cannot be used to diagnose a personality disorder, depression, or another mental condition. Clinical formulation requires qualified evaluation of symptoms, duration, impairment, history, and relevant diagnostic standards.
Preregistration, reproducibility, and measurement transparency
Strong research practice specifies major hypotheses, measures, eligibility criteria, time points, primary analyses, and planned exclusions before testing confirmatory predictions. Exploratory analysis is valuable when labeled as exploratory. Sharing scoring syntax, codebooks, anonymized derived data when ethically permissible, and analysis scripts makes results more interpretable and easier to reproduce. Sensitive identity narratives, however, may contain names, relationships, and community details that remain identifying even after obvious personal information is removed. Open-science commitments should be paired with robust consent and privacy protections.
For person-centered analyses, report alternatives, model-selection procedures, and robustness checks. For narratives, archive prompts, code definitions, decision logs, and appropriate examples subject to consent. For translated instruments, document item adaptation and invariance evidence. Transparency supports cumulative research because another team can determine whether an apparent disagreement reflects different populations, different instruments, or genuine theoretical differences.
Applying Identity Assessment in Counseling, Education, and Clinical Research
Research information is different from an individual diagnosis
Identity development scales are commonly designed to examine patterns across samples. Even a psychometrically strong group-level instrument may have insufficient individual-level precision for screening or counseling decisions. Interpreting one person's score requires normative evidence, appropriate reference groups, standard errors, and a validated decision framework for the proposed use. Most research questionnaires were not established to make high-stakes individual determinations. Practitioners should treat identity measures as prompts for exploration and formulation within their competence, not as tests that pronounce a person mature, immature, or disordered.
In educational and career settings, structured questions can clarify goals, perceived autonomy, uncertainty, and environmental barriers. A counselor might explore a student's responses through collaborative discussion rather than applying a status label as a verdict. Qualitative accounts of family expectations or financial constraints can be as relevant as the numerical commitment score. An assessment that ignores real opportunities can inadvertently moralize hardship and encourage unhelpful advice.
Identity-related distress and mental health
Identity uncertainty can be accompanied by anxiety, sadness, interpersonal strain, or temporary decreases in confidence. Some research finds associations between dimensions of identity processing and adjustment; these relationships vary by construct, age, study design, and context. Evidence about the direction or clinical importance of such associations is uneven. In the longitudinal systematic review and meta-analysis concerning identity and social-emotional disorders, the narrative and quantitative results did not converge on a simple causal account Potterton et al. (2022).
When distress is clinically relevant, assess symptoms and functional impairment with appropriate clinical methods independently of developmental identity scores. Risk assessment, if indicated, follows established professional procedures. Identity measures can provide context about values, belonging, future orientation, or conflicts, but their scores should not replace evidence-based clinical assessment. The editorial principle is construct specificity: one research indicator is not a diagnosis, and no single identity classification is a treatment decision.
The role of clinical interviewing and collaborative formulation
A sensitive identity interview invites participants to articulate commitments and ambivalence in language they choose. It considers personal history, social context, developmental timing, perceived agency, and consequences for functioning. When a person's narrative involves trauma, discrimination, cultural conflict, or major life changes, an ethically sound process privileges informed consent and the participant's control over disclosure. Assessors can use questions about identity development to enrich a formulation while recognizing the limits of the available tools. An interest in identity is not itself evidence that intervention is necessary.
Interventions and evidence-based boundaries
A developmental assessment may suggest questions for counseling, mentoring, or supportive education, such as what values guide a decision, what options have been available, and whether uncertainty is productive or exhausting. Research on changing identity processes is not equivalent to evidence that an intervention will improve mental health or produce a preferred status. When evaluating an identity-focused intervention, investigators should specify outcomes, use validated measures, include credible comparison conditions where feasible, follow participants over time, and distinguish improvement in a scale from meaningful change in life circumstances.
Some interventions may seek reflection, agency, exploration, or commitment quality rather than maximizing certainty. Studies of narrative processes during psychotherapy can generate useful hypotheses, but improvements observed alongside therapy do not isolate the effects of narrative work. Claims about efficacy require designs capable of identifying treatment effects and should be bounded by the sample and outcomes actually studied Adler (2012).
Choosing an Assessment Strategy
Step 1: Define the unit of identity
Specify whether the project concerns personal commitments, identity status, narrative meaning, ethnic-racial identity, social roles, or the ways a person processes identity-relevant information. State whether the target is a single domain, several specified domains, or a broader life-story organization. Begin with a definition that can be connected to observable interview responses, item-level reports, or reliably coded narratives. Vague aims such as measuring how developed a person is encourage invalid shortcuts.
Step 2: Select the time horizon
Decide whether the objective is a snapshot, developmental change over years, adjustment during a transition, or momentary variations in identity salience. Choose intervals and reference periods appropriate to the processes expected to change. A one-time survey is efficient for associations; panel studies are needed to describe within-person trajectories; diary studies can capture short-term fluctuations; retrospective interviews can illuminate remembered change but remain vulnerable to reconstructive memory.
Step 3: Match the instrument to the inference
Use a validated process scale when the study needs dimensional comparison of exploration and commitment. Consider a structured interview when the circumstances and reasoning of choices matter. Use narrative methods for autobiographical interpretation, social identity mapping for affiliations and the organization of roles, and mixed methods when numbers require contextual explanation. The instrument should serve the question rather than determining the question merely because it is readily available.
Step 4: Evaluate the population and language
Review evidence for the actual ages, communities, languages, educational settings, and domains in the intended sample. Test translation quality, dimensional structure, score precision, and measurement invariance when comparisons require them. Invite input from people who will answer the questions. If evidence for a proposed population is weak, describe the use as exploratory, consider adaptation and validation work, and narrow the conclusions accordingly.
Step 5: Plan scoring, interpretation, and ethical safeguards
Specify the scoring algorithm, missing-data approach, uncertainty estimates, and analytic strategy before making consequential inferences. Decide how to handle participants whose answers span multiple domains or are inconsistent across methods. Obtain consent appropriate to the sensitivity of identity stories. Limit identifiable data access, establish retention rules, and avoid unconsented reuse of personal narratives. Provide respondents with information about the purpose and limits of the assessment.
Step 6: Preserve the difference between findings and meaning
Interpret results at the level supported by the instrument. A high reconsideration score indicates more reported reconsideration, not the absence of an authentic self. A low commitment score indicates uncertainty or weak endorsement on the items used, not personal failure. A narrative that lacks a conventional coherent structure may reflect culture, language, privacy, or the interview setting. The strongest conclusion is one that names the specific observed pattern and the conditions under which it was observed.
Evidence Strength and Unresolved Scientific Questions
Established foundations
Established findings include the importance of distinguishing exploration from commitment, the feasibility of measuring multiple identity processes with psychometrically evaluated questionnaires, the contribution of structured interviews, and the value of repeated observations for assessing change. Foundational studies validated particular constructs and measurement approaches, and subsequent research extended them across contexts. The existence of a validated scale in one setting is an established methodological fact; its validity for an entirely different group or purpose remains an empirical question Marcia (1966), Crocetti, Rubini, and Meeus (2008).
Strong but conditional evidence
There is substantial evidence that different dimensions of identity development show distinguishable relationships with adjustment and that identity processes can change over time. Their direction, size, and temporal form depend on the model and population. Longitudinal syntheses support developmental change alongside considerable stability, while the newer U-MICS systematic review demonstrates wide methodological use without resolving every cultural or causal question. These conclusions are strong within their documented operational scope Meeus (2011), Crocetti and Meeus (2026).
Preliminary approaches
New domain-specific scales, computational narrative classifications, and novel multi-tool identity mapping approaches are promising when they add information that conventional surveys miss. They often begin with modest or selected samples, limited independent replication, or incomplete evidence of invariance. Their preliminary character should be stated in the article and in every downstream use, particularly when models produce apparently precise categories Ho and Tang (2026), Mancini and Imperato (2026).
Limited or contested inferences
The assumption that one identity status is always optimal, that all cultures favor the same pattern of autonomous exploration, that retrospective reports accurately capture the sequence of past decisions, or that identity questionnaires provide diagnostic categories goes beyond current evidence. Causal claims about identity development and mental health remain especially sensitive to measurement overlap and confounding. Researchers can strengthen this literature through representative longitudinal samples, common item-level reporting, transparent preregistration, adequate measurement invariance, mixed methods, and replication in communities often absent from existing evidence de Moor and Johnson (2026), Potterton et al. (2022).
Frequently Asked Questions
Can identity development be measured with one test?
No single test captures every dimension of identity development. A questionnaire can estimate particular processes such as commitment or exploration, an interview can examine reasons and contexts, and a narrative task can study how a person gives meaning to experience. Researchers select one or several methods according to a defined scientific question. An instrument is useful when the intended inference fits its validated construct.
What is the difference between an identity-status interview and a questionnaire?
An interview elicits explanations, timelines, and contextual details before a coder applies an explicit framework. A questionnaire records answers to standardized items and produces dimension scores or, if justified, derived categories. Interviews require trained coding and are usually more resource-intensive; questionnaires can reach larger samples but provide less contextual depth. Neither method is universally superior.
Is identity achievement the highest score a person should seek?
Identity achievement is a configuration of exploration and commitment in a particular theoretical model, not a universal measure of human worth or a clinical outcome. A person's circumstances, available choices, identity domain, culture, and developmental period determine how reported exploration and commitment should be understood. A classification describes evidence within its operational definitions.
Are high exploration or reconsideration scores always signs of a problem?
Exploration often reflects ordinary consideration of meaningful alternatives. Reconsideration can be appropriate when a commitment no longer fits someone's circumstances. Ruminative exploration may relate to distress, but these constructs should be measured separately. Scores do not diagnose illness, and assessment should consider functioning, context, and how the individual experiences uncertainty.
Can a self-report scale diagnose an identity disorder?
A general identity-development questionnaire is not a diagnostic instrument for a psychiatric disorder. Clinical diagnosis requires a separate assessment of relevant symptoms, duration, impairment, history, and recognized diagnostic criteria. Identity-related research measures can contribute contextual information when appropriately interpreted by qualified professionals.
How often should researchers assess identity development?
Assessment frequency follows the timescale of the hypothesis. Annual waves can illuminate broad developmental trajectories, more frequent panels may capture change during transitions, and diary or experience-sampling designs can study short-lived fluctuations. Respondent burden, measurement invariance, and the possibility that participation itself influences reflection also shape the schedule.
Do identity measures work equally well across cultures?
Cultural translation and widespread use do not establish equality of measurement. Researchers should evaluate item relevance and response meaning, factor structure, reliability, and measurement invariance before comparing groups. Some measures have significant cross-cultural evidence; other adaptations support only limited uses. The strength of comparison depends on the specific validation results.
Can life-story coding measure identity development objectively?
Structured content coding can establish reproducible rules for assigning narrative codes, while reflexive qualitative methods make the interpretive process explicit. Both can contribute scientifically meaningful information, but narratives are shaped by context, memory, language, and cultural storytelling conventions. Interrater agreement establishes consistency of a coding procedure and cannot by itself prove the validity of its psychological interpretation.
Are social media profiles valid measures of identity?
They are evidence of situated self-presentation and participation, shaped by platforms and audiences. They may complement interviews and surveys in studies of digital identity, but profile content alone cannot establish internal commitment, self-knowledge, or long-term development. Ethical access, informed consent, and privacy protection are especially important because digital data may contain third-party information.
Which approach is best for a longitudinal research project?
A theoretically matched, validated instrument repeated at meaningful intervals is the foundation for measuring change. Depending on the question, researchers can add interviews, diaries, or narrative tasks. They should address longitudinal invariance, attrition, missing data, within-person versus between-person inference, and the possibility that social circumstances change during the study.
Conclusion
Assessment of identity development is most rigorous when it connects a precisely defined construct to an appropriately validated method, takes cultural and developmental context seriously, and treats measurement as evidence about a process occurring through time. Identity-status interviews illuminate how commitments are formed; process scales quantify specific dimensions of exploration and commitment; narrative methods reveal how people organize experiences into meanings; and longitudinal designs test whether and how these patterns change. Each method contributes a different perspective, and the strength of the final inference depends on the alignment among theory, instrument, data, and context.
For the wider conceptual foundations, see Personal Identity in Psychology: Continuity, Roles, and the Sense of Self and Self-Concept in Psychology: Definition, Structure, and Examples. Scientific progress will be measured by whether identity assessment can distinguish the complexity of development with greater accuracy, comparability, and interpretive care.
Related Articles
Identity Development: Exploration, Commitment, and Change Across Life — A development-focused examination of exploration, commitment, and changing identity across life.
Identity Status Theory: Achievement, Moratorium, Foreclosure, and Diffusion — The theoretical classifications that some interviews and questionnaires operationalize.
Narrative Identity: How Life Stories Shape the Self — The psychological study of autobiographical meaning and continuity.
Identity Diffusion: Meaning, Research, and Clinical Boundaries — Conceptual and clinical boundaries around diffusion and identity-related difficulties.
Self-Concept Clarity: Why a Coherent Sense of Self Matters — An adjacent construct and its distinct measurement questions.
Measuring Self-Concept Clarity: The SCC Scale and Research Uses — Describe the Self-Concept Clarity Scale, measurement purpose, research uses, and validity boundaries.
Rosenberg Self-Esteem Scale: Scoring, Interpretation, and Limitations — Explain the RSES structure, scoring conventions, psychometrics, and its non-diagnostic interpretation.
References
Adler, J. M. (2012). Living into the story: Agency and coherence in a longitudinal study of narrative identity development and mental health over the course of psychotherapy. Journal of Personality and Social Psychology, 102(2), 367–389. https://doi.org/10.1037/a0025289
Balistreri, E., Busch-Rossnagel, N. A., and Geisinger, K. F. (1995). Development and preliminary validation of the Ego Identity Process Questionnaire. Journal of Adolescence, 18(2), 179–192. https://doi.org/10.1006/jado.1995.1012
Bennion, L. D., and Adams, G. R. (1986). A revision of the extended version of the Objective Measure of Ego Identity Status: An identity instrument for use with late adolescents. Journal of Adolescent Research, 1(2), 183–197. https://doi.org/10.1177/074355488612005
Berzonsky, M. D. (1992). Identity style and coping strategies. Journal of Personality, 60(4), 771–788. https://doi.org/10.1111/j.1467-6494.1992.tb00273.x
Crocetti, E., and Meeus, W. (2026). The Utrecht-Management of Identity Commitments Scale (U-MICS): Its origins and applications across domains, contexts, and times. Identity, 26(3), 216–235. https://doi.org/10.1080/15283488.2026.2679957
Crocetti, E., Rubini, M., and Meeus, W. (2008). Capturing the dynamics of identity formation in various ethnic groups: Development and validation of a three-dimensional model. Journal of Adolescence, 31(2), 207–222. https://doi.org/10.1016/j.adolescence.2007.09.002
de Moor, E. L., and Johnson, S. K. (2026). Improving quantitative measurement of identity exploration and commitment processes: Recommendations for using existing measures and creating new ones. Identity. Advance online publication. https://doi.org/10.1080/15283488.2026.2700526
Ho, P., and Tang, X. (2026). The development and validation of the Ethnic-Racial Identity Developmental Dynamics Scale (ERIDDS). Identity. Advance online publication. https://doi.org/10.1080/15283488.2026.2726810
Luyckx, K., Goossens, L., Soenens, B., and Beyers, W. (2006). Unpacking commitment and exploration: Preliminary validation of an integrative model of late adolescent identity formation. Journal of Adolescence, 29(3), 361–378. https://doi.org/10.1016/j.adolescence.2005.03.008
Luyckx, K., Schwartz, S. J., Berzonsky, M. D., Soenens, B., Vansteenkiste, M., Smits, I., and Goossens, L. (2008). Capturing ruminative exploration: Extending the four-dimensional model of identity formation in late adolescence. Journal of Research in Personality, 42(1), 58–82. https://doi.org/10.1016/j.jrp.2007.04.004
Mancini, T., and Imperato, C. (2026). The Rubik's Cube of Identity: A comparative exploration of four empirical tools for capturing identity multidimensionality, complexity, and fluidity. Identity. Advance online publication. https://doi.org/10.1080/15283488.2026.2702024
Marcia, J. E. (1966). Development and validation of ego-identity status. Journal of Personality and Social Psychology, 3(5), 551–558. https://doi.org/10.1037/h0023281
McAdams, D. P. (2001). The psychology of life stories. Review of General Psychology, 5(2), 100–122. https://doi.org/10.1037/1089-2680.5.2.100
McLean, K. C., Pasupathi, M., and Pals, J. L. (2007). Selves creating stories creating selves: A process model of self-development. Personality and Social Psychology Review, 11(3), 262–278. https://doi.org/10.1177/1088868307301034
Meeus, W. (2011). The study of adolescent identity formation 2000–2010: A review of longitudinal research. Journal of Research on Adolescence, 21(1), 75–94. https://doi.org/10.1111/j.1532-7795.2010.00716.x
Phinney, J. S., and Ong, A. D. (2007). Conceptualization and measurement of ethnic identity: Current status and future directions. Journal of Counseling Psychology, 54(3), 271–281. https://doi.org/10.1037/0022-0167.54.3.271
Potterton, R., Austin, A., Robinson, L., Webb, H., Allen, K. L., and Schmidt, U. (2022). Identity development and social-emotional disorders during adolescence and emerging adulthood: A systematic review and meta-analysis. Journal of Youth and Adolescence, 51, 16–29. https://doi.org/10.1007/s10964-021-01536-7
Salehi Yegaei, P., Achenbach, T. M., Trejos-Castillo, E., Mastrotheodoros, S., Izanloo, B., and Habibi Asgarabad, M. (2023). Dimensions of Identity Development Scale: Confirmatory factor analysis, gender invariance, and external validity of the Persian version. Journal of Research on Adolescence, 33(4), 1115–1130. https://doi.org/10.1111/jora.12860
Watson, C., Gutierrez, F., and Azmitia, M. (2026). Honoring complexity: Five methodological commitments for using reflexive thematic analysis in identity research. Identity, 26(3), 202–215. https://doi.org/10.1080/15283488.2026.2656196