1 Introduction
The way in which children and youths come to understand who succeeds in society and why is a relevant aspect of the socialization process (Mijs, 2016; Resh & Sabbagh, 2014). Schools are important actors in this regard, as they use grading, tracking, and selection to apply meritocratic criteria based primarily on individual talent and effort (Darnon, Wiederkehr, et al., 2018; Traini, 2022). This early exposure to meritocratic principles is consequential, as it shapes students’ understanding of fairness and incorporates notions of social mobility and social hierarchies. At the same time, schools are environments where students encounter evidence that outcomes are shaped by factors other than meritocratic ones, such as family background, social connections, and structural advantages (Bourdieu & Passeron, 1990; Goudeau & Croizet, 2017; Tang et al., 2025). These background attributes can be considered privileges due to heritage-based advantages over which people have no control, unlike effort. The potential tension between meritocratic ideals and the weight of privilege factors is a complex issue for students to navigate and is central to their understanding of fairness and inequality. Yet, there is still little knowledge about how students conceive meritocracy and to what extent such conceptions challenge the role of privilege in society.
Although meritocracy is often treated as a single dimension, mostly related to individual effort, the literature suggests that it is a multidimensional construct. On the one hand, this refers to the distinction between perceptions of how meritocracy operates in society and preferences about how it should operate (Castillo et al., 2023). The former refers to the beliefs about the actual functioning of meritocracy, while the latter refers to the normative beliefs about how society should reward talent and effort. On the other hand, meritocratic beliefs can be distinguished from privilege-based beliefs, which emphasize the role of family wealth and social connections in shaping rewards in society (C. Liu & Wang, 2025; Tang et al., 2025). Such beliefs also encompass perceptions of how society functions and preferences about people’s normative inclinations. In this sense, a multidimensional approach incorporates perceptions and preferences about both meritocratic and privilege-based explanations of social outcomes, allowing for a more comprehensive understanding of how individuals interpret social inequalities and fairness.
Previous research has shown that privilege and meritocratic beliefs represent different and not necessarily opposite dimensions, therefore they can coexist in people’s minds (C. Liu & Wang, 2025; Tang et al., 2025). This means, for instance, that a person can believe that society rewards talent and effort while also acknowledging that family wealth and social connections play a role in shaping outcomes. This coexistence of beliefs has important implications for how individuals understand social mobility, fairness, and inequality, as well as for how they evaluate policies aimed at promoting equal opportunities. However, attempts to study these beliefs have been limited by the lack of instruments that capture their multidimensionality and by the absence of longitudinal studies that examine how these beliefs evolve over time. In the present study, we aim to contribute to the field of meritocracy research by applying a multidimensional model to measure meritocratic and privilege beliefs in a sample of Chilean students and by examining the stability and measurement equivalence of these beliefs across school stages and over time. Our main argument is that it is possible to identify and distinguish meritocratic and privilege beliefs even at an early school age, and that these beliefs have a structure similar to that found in a previous study that applied the meritocracy framework in the adult population (Castillo et al., 2023). Furthermore, we argue that these beliefs are not static, but rather evolve over time as students gain more experience and exposure to different social contexts.
This study draws on a two-wave panel dataset of Chilean students in 6th and 9th grade (in wave 1). Studying meritocracy and privilege in the context of Chilean schools provides a unique opportunity to explore how students navigate the tension between meritocratic ideals and the reality of social inequality. Chile’s educational system combines high stratification, market-based competition, and explicit meritocratic discourse in school culture (Castillo et al., 2024; Valenzuela et al., 2013), alongside persistent socioeconomic inequality and a deep marketization of social welfare (Ferre, 2023), features shared by many Latin American countries and increasingly present in European and East Asian systems (Busemeyer, 2014; Gingrich, 2011). This makes Chile a productive case for studying how students internalize, question, or reconcile meritocratic ideals with the reality of privilege, while the dimensional structure and longitudinal patterns identified here are likely to travel to other highly stratified educational contexts.
The research questions guiding this study are as follows:
Do school-aged students distinguish between meritocratic and privilege-based principles, as well as between perceptions (“what is”) and preferences (“what should be”)?
Are the perceptions of privilege detrimental for the endorsement of meritocratic beliefs, or do they coexist in students’ minds?
Are these dimensions longitudinally stable within students over time, and does their measurement equivalence hold across school stages, or does educational experience reshape the meaning of meritocratic beliefs?
In the following sections, we review the theoretical background on meritocratic beliefs in school settings, present the methodology and results, and discuss the implications of our findings for understanding how meritocratic socialization develops during adolescence and for civic education, school climate, and equity-oriented policy.
2 Theoretical and empirical background
2.2 Conceptualizing and measuring meritocratic and privilege beliefs
The research agenda on meritocratic beliefs has faced challenges in conceptualizing and measuring meritocracy. Some studies have relied on narrow operationalizations that equate meritocracy with the social valuation of effort (Mijs, 2021; Wiederkehr et al., 2015), often reflecting simplified readings of Young’s (1958) formulation (Merit = Effort + Intelligence). Other approaches treat meritocratic beliefs as interchangeable with broader constructs such as system justification (Day & Fiske, 2017; Jost et al., 2004), beliefs about social mobility (Deng & Wang, 2025; McCoy & Major, 2007), or generalized support for equal opportunity (Batruch et al., 2022; Darnon, Wiederkehr, et al., 2018). As a result, attempts to measure meritocracy with items battery frequently collapse indicators that tap different dimensions, making it difficult to assess what meritocracy actually means in practice.
A related, more specific limitation to study meritocracy concerns its links with privilege-based factors such as family wealth, social connections, and other structural advantages. As Castillo et al. (2023) notes, much survey research relies on item batteries that ask respondents to rate the importance of effort, talent, family background, networks, or luck for “getting ahead”, and then reduces these into a single score. This becomes especially problematic when researchers construct continuum measures in composite indicators, treating meritocratic and privilege-based explanations as opposite ends of the same scale (e.g., by subtracting “non-merit” from “merit,” as in Reynolds & Xian, 2014). Such operationalizations embed a strong zero-sum assumption, namely that endorsing merit necessarily entails rejecting structural or relational advantages, thereby ruling out, by design, the empirical possibility that both types of beliefs coexist. Recent debates, including the exchange between Mijs (2026) and Wiesner & Sachweh (2026), underscore this point by arguing for measures that capture meritocratic beliefs alongside privilege-based ones, rather than as a residual of the relative importance assigned to structural factors. Complementary empirical work has raised similar concerns: C. Liu & Wang (2025) shows that meritocratic endorsement may remain stable even as recognition of structural constraints increases; Tang et al. (2025) motivates dual consciousness but notes that operationalizations forcing relative trade-offs risk obscuring it; Kwon & Pandian (2024) identify three attitudinal clusters in Europe, instrumental, idealized merit, and ambivalent, showing that support for merit can coexist with recognition of social connections in distinct configurations; and Zhu (2025) argues directly that meritocratic and structural explanations are not zero-sum but often accumulate rather than replace one another. Unidimensional indices, and especially subtraction scores, can therefore manufacture artificial neutral positions and blur substantively distinct belief profiles that carry different educational consequences.
In response to these limitations for the comprehensive measurement of merit and privilege beliefs, Castillo et al. (2023) proposed a multidimensional framework that decomposes meritocratic beliefs along two analytically independent axes: preferences versus perceptions, and meritocratic versus privilege-based (originally called “non-meritocratic”) allocation principles. Preferences capture normative ideals about how rewards should be distributed (e.g., whether effort and ability ought to determine life chances); perceptions capture descriptive evaluations of how society actually works (e.g., whether observed inequalities reflect merit-based processes) (Janmaat, 2013). Meritocratic elements (effort, talent) and privilege-based elements (family wealth, networks) are treated as distinct components rather than as opposite poles of a single continuum. This yields four non-redundant constructs, allowing belief combinations that older measures tend to collapse, such as endorsing meritocracy as an ideal while simultaneously recognizing the weight of family background and connections. The proposal is deliberately minimalist, making it well-suited for large-scale surveys with limited questionnaire space, though this parsimony entails a trade-off: each factor is measured with only two items, which can constrain reliability and content coverage.
Evidence from adult samples using confirmatory factor analysis supports a four-factor structure (perceptions and preferences of both meritocracy and privilege (Castillo et al., 2023). Besides, it shows systematic associations among dimensions: stronger perceived privilege tends to co-occur with stronger meritocratic preferences (Castillo et al., 2023), a pattern that aligns with Zhu’s (2025) notion of dual consciousness, here understood as the simultaneous endorsement of meritocracy as a normative ideal alongside descriptive recognition that rewards are shaped by structural advantage. Regarding school-based research, these measurement challenges have received limited attention. Chauvin et al. (2026) propose a multidimensional measure of school meritocracy that distinguishes between equity and ability dimensions, but their scale applies only to teachers and remains restricted to perceived meritocracy. Castillo et al. (2024) apply the four-factor framework to a sample of Chilean 10th graders and report that students differentiate perceptions from preferences and meritocratic from privilege-based principles; however, that study relies on item-level descriptives and bivariate associations rather than a formal factorial measurement model. Beyond this, to our knowledge no study has examined whether these dimensions are stable across age cohorts or over time, which limits understanding of how meritocratic beliefs develop through schooling and whether observed differences across stages reflect substantive change or measurement non-equivalence.
2.3 This study
Building on the previous evidence, we state the following research hypotheses:
H_1 (Dimensional differentiation): Students’ meritocratic and privilege beliefs can be represented and measured by four distinct factors: perceived meritocracy, perceived privilege, meritocratic preferences, and privilege acceptance.
H_2 (Equivalence): The four-factor model is equivalent across the two waves of the panel survey and across two age cohorts (6th and 9th grade in the first wave).
H_3 (Beliefs compatibility): merit and privilege beliefs are not mutually exclusive, but rather compatible orientations that can coexist within the same individual.
This third hypothesis is particularly important because it challenges the assumption that meritocratic and privilege-based beliefs are zero-sum, and it motivates the use of a multidimensional measurement model that allows for the coexistence of these orientations. Nevertheless, it opens the question of how these beliefs are organized within people’s minds, and whether they can be grouped into distinct profiles. Individuals do not always adhere to a single ideal, but can apply mutiple beliefs simultaneously. In this regard, while we expect that meritocratic and privilege-based beliefs are compatible, it is possible that some students endorse both types of beliefs strongly, while others endorse one type more than the other. This raises the possibility of identifying different groups of students based in their belief profiles, which could shed new light on how beliefs about merit and privilege are formed among the student population. Thus, as a additional analysis, we will test whether there are different profiles of beliefs regarding meritocracy and privilege using Latent Class Analysis, a method that allows us to identify and group individuals with similar beliefs based on their response patterns.
2.4 The Chilean educational context
Chile’s educational system has undergone a profound market-oriented transformation since the early 1980s under the military dictatorship (1973-1989), when a comprehensive reform introduced school vouchers, decentralization, and private provision as organizing principles of public education (Valenzuela et al., 2013). The result is a system in which families formally choose among schools competing for enrollment, but in which that competition reproduces and amplifies existing inequalities: school quality is strongly stratified by socioeconomic status, ability tracking begins early, and access to higher-quality institutions depends heavily on academic performance and, in many cases, on family resources and social networks Otero et al. (2017). Chile consistently ranks among the most educationally stratified countries in the OECD, with large and persistent gaps in learning outcomes across income groups that the expansion of educational access has not resolved (Ferre, 2023; OECD, 2025). Within this structure, meritocratic discourse is actively promoted through institutional design and public policy: effort and individual achievement are framed as the legitimate and sufficient route to social mobility, and recent debates over school admissions reform have reinforced this, with competing governments successfully installing concepts of merit and academic selection at the center of public debates about educational fairness, displacing alternative framings centered on inclusion and equity (Cabalin et al., 2023).
Within this context, students, teachers, families, and adults who have navigated the system all tend to combine meritocratic ideals with recognition of structural advantage, though the form this takes varies by social position and school type. Among secondary students, awareness of educational inequality is widespread, yet meritocratic principles are simultaneously invoked to evaluate and naturalize one’s own position within it (Neut Aguayo et al., 2025). Among teachers and parents, merit is defined less through academic outcomes than through attitude and visible effort, with failure attributed to individual will rather than structural conditions, a pattern that varies by school type and was accentuated after the pandemic (Moyano Dávila et al., 2025). Even among upwardly mobile professionals, success is rarely attributed to individual merit alone: effort is understood as a collective family project, and luck, networks, and institutional support are recognized as equally decisive (Fercovic, 2022). Together, these findings suggest that the coexistence of meritocratic and privilege-based accounts is a durable feature of Chilean culture, making the school system the primary arena in which these orientations are first encountered and gradually consolidated.
Beyond its national specificity, Chile’s case speaks to broader patterns in comparative educational research. The combination of market-based competition, ability tracking, explicit meritocratic discourse, and persistent socioeconomic stratification that characterizes the Chilean system is not unique to Latin America: similar configurations have emerged in European countries that have pursued school choice and accountability reforms (Busemeyer, 2014; Gingrich, 2011), and in China, where high-stakes selection mechanisms produce strong meritocratic narratives alongside intense competition and visible class-based advantage (C. Liu & Wang, 2025; Tang et al., 2025). This makes Chile not a peripheral case but a particularly clear instance of a more general phenomenon: the coexistence of meritocratic ideology and structural inequality in formally open educational markets. How meritocratic beliefs are structured, how stable they are across school stages, and whether they can be measured equivalently across student populations are questions that any research program on meritocratic socialization must confront, regardless of national context. Findings on how to measure these beliefs reliably in a highly stratified system can therefore inform similar efforts elsewhere and contribute to building a comparative basis for understanding how schooling shapes students’ orientations toward merit, privilege, and educational opportunity.
3 Method
3.1 Participants
This study draws on secondary data from the Panel Survey on Education and Meritocracy (EDUMER) for the 2023 and 2024 waves, covering school-age students1. The primary objective of the survey was to examine students’ beliefs, attitudes, and behaviors concerning meritocracy, inequality, and citizenship. The questionnaire was administered to sixth-grade and first-year secondary students from nine schools in the Metropolitan and Valparaíso regions of Chile. The sampling strategy was non-probabilistic and lacked quota controls; however, a minimum sample size of 900 students was established to ensure adequate statistical power. After data processing, the final analytical sample comprises 846 students for Wave 1 (386 girls, 421 boys, 39 identifying as other), and the second wave followed 662 of them (303 girls, 338 boys, 21 identifying as other), yielding a retention rate of 78.3% (attrition = 21.7%). The primary causes of attrition were school transfers between academic years and student absenteeism, reported to be high in several schools. Missing data within waves were handled via listwise deletion. To assess whether attrition introduced systematic bias, we compared retained and non-retained students on all Wave 1 scale items and cohort membership using independent-samples t-tests and chi-square tests. No statistically significant differences were found between groups on any of the meritocracy dimensions or on cohort distribution (all pp p > .05), suggesting that attrition was largely non-systematic with respect to the constructs under study. Results of these comparisons are reported in the Supplementary Material.
3.2 Procedure
Data collection was conducted by a professional research firm using a Computer-Assisted Web Interviewing (CAWI) approach, based on online questionnaires. All participants received an informed consent form that was reviewed and validated by a parent or legal guardian before participation. To enable longitudinal linkage while protecting confidentiality, the survey firm generated anonymous unique identifiers for each student; no directly identifying information was collected for students or their guardians in the analytic files.
3.3 Measures
3.3.1 Scale of perceptions and preferences of meritocracy
Meritocratic and privilege-based perceptions and preferences were measured using the same items proposed in the original scale (Castillo et al., 2023). Each of the four dimensions of the scale is measured by two items. The dimensions are:
Meritocratic perceptions: the extent to which effort and ability are rewarded in Chile,
Privilege perceptions: the extent to which success is perceived as linked to connections and family wealth,
Meritocratic preferences: agreement with that those who work harder or are more talented should be better rewarded,
Privilege acceptance: agreement that it is acceptable for individuals with better connections or wealthy parents to achieve greater success (see Table 1).
Each item is rated on a four-point Likert scale ranging from “strongly disagree” (1) to “strongly agree” (4), therefore higher scores indicate stronger endorsement of the corresponding perception or preference.
Each factor is measured by only two indicators, the minimum required for identification in a Confirmatory Factor Analysis framework. This is a deliberate trade-off in the original scale design (Castillo et al., 2023), which prioritizes parsimony for large-scale survey use over content breadth and reliability. As a consequence, internal consistency cannot be computed in the conventional sense, and the two items per dimension may not fully capture the construct’s full theoretical range. Factor-level estimates should therefore be interpreted with appropriate caution throughout the results and discussion.
3.3.2 Cohort level
To distinguish students’ academic level across waves for conditional and multi-group invariance analyses, we created a cohort indicator capturing whether the respondent belonged to the primary (sixth grade) or secondary (first-year secondary) cohort at the time of each survey wave. Descriptive statistics2 by cohort are reported in Table 2.
3.4 Analytical strategy
3.4.1 Measurement model and estimation
To evaluate the underlying structure of the scale, we employed Confirmatory Factor Analysis (CFA) based on a measurement model with four latent factors, using Diagonally Weighted Least Squares with robust correction (WLSMV) estimation. This estimator is particularly suitable for ordinal data, such as four-point Likert-type scales, as it avoids the bias associated with treating categorical data as continuous (Kline, 2023).
Model fit was assessed following the guidelines of Brown (2015) using a triangulated approach that considers multiple indices jointly rather than applying fixed pass/fail cutoffs. The Comparative Fit Index (CFI) and Tucker-Lewis Index (TLI) values above 0.95 indicate good fit, with values between 0.90 and 0.95 considered acceptable; the Root Mean Square Error of Approximation (RMSEA) indicates good fit below 0.06 and acceptable fit below 0.08; and the Chi-square statistic provides an additional reference, with p > 0.05 and a ratio of Chi-square to degrees of freedom below 3 suggesting adequate fit. Because these indices capture different aspects of model-data correspondence and can diverge in practice, fit evaluation follows the overall pattern across indices rather than any single criterion (Kline, 2023).
3.4.2 Measurement invariance
A key contribution of this study is its assessment of measurement stability through factorial invariance testing (Davidov et al., 2014). We implemented two complementary strategies: cross-group invariance across school levels (primary vs. secondary students) and longitudinal invariance across two panel waves. Given the ordinal nature of the items, we used estimation procedures appropriate for ordered categorical indicators, following Liu et al. (2017) for longitudinal invariance and Svetina et al. (2020) for cross-group tests. For the longitudinal analyses, we adopted the standard hierarchical sequence of four nested models—configural (same factor structure), metric (equal loadings), scalar (equal thresholds/intercepts), and strict (equal residual variances). For the cross-group analyses, we estimated a configural model, then a threshold-invariant model, and finally a model jointly constraining thresholds and factor loadings across groups.
In addition to the traditional chi-square difference test, we assessed invariance using changes in approximate fit indices. Following Chen’s (2007) recommendations for adequate sample sizes, we treated \DeltaCFI as the primary criterion and \DeltaRMSEA as a supplementary criterion. Noninvariance was indicated by a decrease in CFI of at least .010 (i.e., \DeltaCFI \leq −0.010), together with an increase in RMSEA of at least .015 (i.e., \DeltaRMSEA \geq 0.015).
All analyses were performed using the lavaan package in R version 4.5.2. The hypotheses of this research were pre-registered on the Open Science Framework (OSF).
4 Results
4.1 Measurement model
Figure 1 presents the standardized factor loadings and the overall fit indices of the general measurement model estimated on the pooled analytical sample across cohorts and waves. The model shows a reasonable, though not fully satisfactory, fit to the data (CFI = 0.97; TLI = 0.94; RMSEA = 0.07). Although the CFI falls within commonly accepted ranges, the RMSEA and TLI remain below conventional thresholds, indicating that fit is acceptable in some respects but not uniformly strong across indices. Overall, these results suggest that the proposed four-factor structure captures the main covariance pattern in the combined sample reasonably well 3.
All standardized factor loadings are statistically significant and generally moderate to strong, though their magnitude varies across dimensions. For meritocratic perceptions, both indicators load positively on the latent factor, with effort showing a particularly strong loading (\beta = 0.88) and talent a moderately strong loading (\beta = 0.62). Meritocratic preferences exhibit more even, but somewhat weaker, loadings for effort (\beta = 0.59) and talent (\beta = 0.55). For perceived privilege, both indicators load strongly, especially contacts (\beta = 0.88), followed by coming from a rich family (\beta = 0.72).Privilege acceptance also shows strong and relatively balanced loadings, with contacts (\beta = 0.84) and rich family (\beta = 0.72), indicating that both indicators are similarly representative of this latent dimension.
The latent correlations reveal a differentiated structure among the four dimensions. Meritocratic perceptions and meritocratic preferences are positively correlated (r = 0.14), consistent with some alignment between what students perceive is rewarded and what they believe should be rewarded. Meritocratic perceptions and privilege perceptions are negatively correlated (r = −0.21), suggesting that students who perceive greater meritocracy in society tend to view a weaker role for connections and family wealth. At the same time, privilege perceptions are positively related to meritocratic preferences (r = 0.48, strongest correlation), suggesting that perceiving society as rewarding contacts and wealth may coincide with stronger endorsement of meritocratic ideals. Finally, meritocratic preferences and privilege acceptance are modestly positively correlated (r = 0.15), and privilege perceptions and preferences are also positively associated (r = 0.22), indicating that recognizing privilege in practice is linked to a greater acceptance of their legitimacy, albeit to a moderate extent.
4.2 Invariance analysis
We assessed measurement invariance through a hierarchical sequence of nested confirmatory factor analysis models, imposing progressively stricter equality constraints to evaluate whether the measurement model was comparable across time and cohorts. For longitudinal invariance, we followed the approach in Y. Liu et al. (2017) and estimated the model simultaneously across the two waves. We began with a configural model that specified the same factor structure at each wave while allowing item parameters (factor loadings, thresholds/intercepts, and residual variances) to vary freely; this model served as the baseline for subsequent comparisons. We then tested metric (weak) invariance by constraining factor loadings to equality across waves, followed by scalar (strong) invariance by additionally constraining item thresholds (and, where applicable, intercepts) to equality over time—an essential step with ordered-categorical indicators to ensure that response categories map onto equivalent points on the latent continuum and thus permit meaningful comparisons of latent means. Finally, we evaluated strict invariance by further constraining residual variances to equality across waves. To assess invariance between cohorts, we followed the sequential procedure proposed by Svetina et al. (2020) for ordinal indicators: we estimated a configural model, then a threshold-invariant model, and finally a model that jointly constrained thresholds and factor loadings across groups.
In the following sections, we proceed as follows. First, we present descriptive statistics for the meritocracy scale among the comparison groups. Second, we report the results of the measurement invariance analyses, emphasizing the highest level of invariance achieved in each comparison. Finally, we examine potential sources of non-invariance to identify which items or parameters are responsible for the lack of equivalence.
4.2.1 Longitudinal invariance
Figure 2 displays the response distributions for the meritocracy scale, distinguishing perceptions (Panel A) from preferences (Panel B) across waves. In Wave 1, perceptions show high levels of agreement across all items, especially for contacts (77%), talent (74%), and having rich parents (71%), with effort also widely endorsed (67%); at the same time, sizeable minorities disagree (33% for effort, 29% for rich parents, and 27% for talent), indicating a dual recognition in which students strongly acknowledge privilege-based determinants (contacts and parental wealth), while within the meritocratic dimension they view talent as more influential than effort, which is perceived as the least rewarded. Wave 1 preferences show the inverse pattern: effort is overwhelmingly endorsed as the legitimate basis for rewards (87%), far exceeding its perceived role in practice; views on talent are split (54% agree, 45% disagree); parental wealth remains nearly evenly divided; and contacts are relatively more accepted than rejected (61% vs. 39%), suggesting that this factor is not uniformly considered illegitimate in the “ideal” society.
Wave 2 largely replicates these patterns with modest shifts. In perceptions, endorsement of effort declines (62%), whereas contacts (76%), talent (70%), and parental wealth (70%) remain consistently high. In preferences, support for effort remains highly stable (88%), while support for talent declines, with disagreement rising to 51%. Attitudes toward parental wealth remain evenly divided, and contacts continue to be more accepted than rejected (59% vs. 41%), albeit far below the near-consensus around effort. Overall, both waves reveal a stable gap: students perceive social outcomes as strongly shaped by privilege-based factors (especially contacts and family wealth) and, to a lesser extent, talent, yet they overwhelmingly endorse an effort-centered meritocratic ideal. This persistent divergence between perceived reality and normative preferences highlights a key paradox—strong support for meritocracy as a principle alongside skepticism about its realization in practice.
Table 3 reports the results of the longitudinal invariance tests. Overall, the scale achieved the strongest level of invariance (strict invariance) across the two waves. The strict model showed good fit, \chi^2(84) = 130.17, CFI = 0.991, RMSEA = 0.031 (90% CI: 0.020–0.040). Moreover, compared with the strong (scalar) model, the deterioration in fit was negligible: \Delta \chi^2(4) = 1.635, \DeltaCFI = 0.000, and \DeltaRMSEA = -0.002. These differences fall well within commonly used cutoffs for invariance evaluation (Chen, 2007), indicating that constraining residual variances in addition to loadings and thresholds/intercepts is tenable. Substantively, this implies that the item–factor relations are stable over time (metric invariance) and that, for a given level of the latent trait, respondents are expected to endorse the same response categories across waves (scalar invariance), supporting the comparability of scores and latent parameters within students over time.
As an additional diagnostic, we examined potential sources of localized misfit using univariate score tests showing the increase of misfit at each step of the invariance hierarchy (configural, metric, scalar, and strict). Across steps, the score tests provided no evidence of meaningful violations of the imposed equality constraints, indicating that a small subset of parameters did not drive the constraints. The supplementary materials report the global score tests and the smallest p-values observed at each step.
4.2.2 Cohort invariance
Figure 3 shows response distributions by cohort, separating perceptions and preferences. In primary education, perceived rewards are similar for effort and talent (both 76% agree). Contacts (67%) and a rich-family background (59%) are also seen as rewarded, and these privilege-based factors elicit the most disagreement within the cohort (41% disagree with rich-family background; 33% with contacts). In preferences, effort is clearly dominant (84% agree), whereas talent and a rich family background split evenly. Notably, 62% consider it as good that those who have good contacts do better in life (39% disagree).
In secondary education, perceptions shift: disagreement with the idea that effort is rewarded increases (45%), and agreement decreases (55%). Disagreement about talent also increases (31%). The largest differences concern non-meritocratic factors, which are overwhelmingly perceived as rewarded (82% for rich-family background; 86% for contacts), with a stronger concentration in the highest agreement category. Preferences again strongly favor effort (92%), while views on talent, rich-family background, and contacts remain mixed (roughly 45% disagree to 58% agree across items).
Table 4 summarizes the cross-cohort invariance tests. The scale achieves configural invariance between primary and secondary students, indicating the same four-factor structure across groups, but does not support threshold invariance and therefore also fails to meet the prerequisites for metric or scalar invariance. Following the indications by Wu & Estabrook (2016) and Svetina et al. (2020) for group invariance testing, we estimate a first invariance level restricting the tresholds, and further a second invariance level with tresholds and loadings restricted. Imposing equality constraints on thresholds led to a deterioration in model fit (\DeltaCFI = −.013; \DeltaRMSEA = .013) that exceeds Chen’s (2007) recommended cutoffs for threshold/intercept invariance. This pattern indicates systematic differences in how response categories relate to the underlying latent constructs across cohorts. Substantively, primary and secondary students do not use the response scale equivalently when evaluating meritocratic and privilege-based criteria, which limits the interpretability of direct cross-level comparisons of latent means or other latent-score differences.
To locate the source of cross-cohort non-invariance, we inspected univariate score tests from lavTestScore for the threshold-invariant model. The largest localized violations were concentrated in the thresholds of the perceived-effort item, especially the second threshold (\chi^2=10.12, p=.001), with the first threshold also among the top contributors (\chi^2=7.55, p=.006). Substantively, the younger and older cohorts appear to use the response options differently when judging whether effort is rewarded in society, consistent with socialization processes through which older students may adopt a more critical view of how far effort translates into success.
4.2.3 Conditional longitudinal invariance by cohort
Despite the lack of cohort measurement invariance, we examined whether plausible cohort-related heterogeneity could compromise longitudinal invariance by re-estimating the longitudinal invariance sequence while controlling for cohort (dummy-coded) as a covariate predicting each latent factor. Model fit remained good across all steps—from the configural baseline to the strict model—and consistently met conventional standards; for instance, the strict invariance model with cohort control showed \chi^2(92) = 131.64, CFI = .991, and RMSEA = .027. Importantly, including cohort as a predictor did not materially alter model fit or the invariance conclusions, suggesting that the equality constraints over time are tenable even after accounting for cohort differences in average latent factor levels.
This result has two main implications. First, it strengthens the claim that the measurement properties of the scale are temporally stable within individuals: observed over-time comparisons are not driven by shifts in how primary versus secondary students use the response categories, but reflect stability (or change) in the underlying latent constructs. Second, and substantively, cross-cohort non-invariance should not be read merely as a technical limitation on comparability. Rather, it is an informative finding in its own right: the fact that threshold non-equivalence is concentrated in the perceived-effort item suggests that the meaning of “effort being rewarded” shifts with educational experience. Primary and secondary students do not simply differ in how much they perceive effort to be rewarded—they appear to interpret the question through different reference frames, likely reflecting cumulative exposure to school-based sorting and the growing visibility of structural constraints as students progress. Cross-level comparisons of latent means should therefore be made with caution or handled via partial invariance or cohort-specific models, while within-person longitudinal analyses pooling both cohorts remain well-supported.
5 Discussion
The first hypothesis proposed that students’ meritocratic beliefs would be best represented by four distinct factors: perceived meritocracy, perceived privilege, meritocratic preferences, and privilege acceptance. The confirmatory factor analyses support this expectation: the scale performs well in the school-aged population, with satisfactory model fit, strong factor loadings, and meaningful correlations among factors. Consistent with evidence from adult populations (Castillo et al., 2023), these results indicate that even at an early stage of school socialization, students distinguish between perceptions and preferences, and between meritocratic and privilege-based principles. This four-factor structure also resonates with qualitative evidence from Chilean schools, where students across socioeconomic groups simultaneously invoke meritocratic ideals and recognize structural advantage (Neut Aguayo et al., 2025), and where merit is often defined by attitude and visible effort rather than by academic outcomes (Moyano Dávila et al., 2025). This last finding is relevant for interpreting the privilege acceptance dimension: in contexts where merit is primarily understood as behavioral disposition, accepting that privilege-based advantages lead to better outcomes may not be experienced as a contradiction but as a separate normative judgment about a different domain, a reading that, while going beyond what Moyano Dávila et al. (2025) directly establish, is consistent with their evidence. This suggests that the coexistence of these orientations reflects how students actually navigate competing narratives about fairness and reward, rather than an artifact of survey design.
The empirical separability of these dimensions has substantive implications for how meritocratic beliefs are understood and measured (Kwon & Pandian, 2024; Mijs, 2026; Zhu, 2025). Perceived and preferred meritocracy are already distinguishable among school-aged students, consistent with the distinction between “what is” and “what should be” and with earlier attempts to operationalize it (Castillo et al., 2023; Duru-Bellat & Tenret, 2012). This matters because the same normative endorsement of merit can carry different meanings depending on perceived reality: high perceived meritocracy is more compatible with affirming existing arrangements, whereas low perceived meritocracy can reflect dissatisfaction and a demand for greater fairness. At the same time, meritocratic and privilege-based principles coexist rather than forming opposite poles of a single continuum, directly challenging zero-sum measurement strategies that presuppose endorsing merit entails rejecting privilege (Reynolds & Xian, 2014; Wiesner & Sachweh, 2026).
The correlational pattern observed in our data supports the coexistence of meritocratic and privilege-related factors. First, perceived privilege is positively associated with stronger meritocratic preferences, which suggests a normative-reactive response to perceived unfairness (Castillo et al., 2023). Perceiving privilege entails recognizing structural inequalities, that is, a social order in which a person’s background shapes their chances of success. Adherence to meritocratic criteria can thus be read as a rejection of a world governed by privilege. Second, meritocratic perceptions are positively correlated with preferences for privilege. Prior research has linked meritocratic perceptions to the self-interest hypothesis, which holds that those who occupy higher social positions are more likely to adopt a meritocratic worldview. Extending this framework to incorporate privilege reveals a more complex picture: perceiving society as meritocratic does not merely reflect a particular reading of how society works, but may also conceal support for privilege. In other words, the coexistence of these factors suggests that the justification of inequality is embedded in the meritocratic narrative itself, such that the defense of privilege emerges alongside the belief that the world is meritocratic.
The positive association between the normative dimensions of meritocracy and privilege provides further evidence that merit and privilege coexist in people’s views. While support for privilege could be interpreted as a justification for inequality, its co-occurrence with support for meritocratic norms warrants closer examination. This pattern suggests that inherited or relational advantages are accepted as legitimate drivers of success, and so are effort and talent. From this perspective, success is conceived as a goal that warrants mobilizing every available resource, whether earned or inherited. The negative correlation between perceptions of meritocracy and perceptions of privilege, however, points to a different dynamic. At the descriptive level, individuals appear to hold a dominant account of how the world works, in which either merit or privilege prevails. The inverse relationship between them indicates that these accounts do not coexist harmoniously but instead compete with one another. One possible explanation for this contrast between the normative and descriptive levels is that preferences are more readily evaluated in light of one’s own situation and interests, which makes them more malleable. Perceptions of how the world works, by contrast, are shaped primarily by material conditions of existence and opportunities for socialization, which makes them a less flexible interpretive framework. Finally, the correlations within each domain, between the two meritocracy factors, and between the two privilege factors, are positive and more intuitive. They indicate that beliefs about how society functions are aligned with beliefs about how it should function. This pattern is consistent with a mutually reinforcing relationship between descriptive and normative beliefs, although our cross-sectional design does not allow us to establish the direction of that relationship.
This results are consistent with Zhu’s (2025) argument that meritocratic and structural explanations accumulate rather than displace one another, and with accounts in which merit endorsement may remain stable as recognition of structural constraints increases (C. Liu & Wang, 2025; Tang et al., 2025). It is also worth noting how this finding relates to typological approaches: Kwon & Pandian (2024) document three distinct attitudinal clusters in Europe, showing that coexistence takes different forms across individuals. The present study complements that work by showing how the same coexistence can be represented within a single coherent measurement model that supports longitudinal inference, rather than requiring person-centered classification. This has concrete implications for how educators and policy designers interpret students’ orientations toward fairness. A student who strongly endorses merit as an ideal while recognizing that privilege shapes outcomes is not disengaged or fatalistic: they hold a normative demand for fairness that the system is not meeting. Conflating this profile with one of simple disbelief in effort leads to misdiagnosis, whether in classroom practice or in policy design. Distinguishing which dimension is at stake, endorsement of merit, recognition of privilege, or acceptance of it as legitimate, matters for how schools interpret students’ sense of fairness and respond to it through civic education, classroom climate, and efforts to build trust in the fairness of school itself.
These results also have implications for how the existing literature on school meritocracy should be interpreted. A substantial body of research has linked stronger meritocratic beliefs among students to inequality legitimation, weaker support for redistribution, and greater acceptance of unequal access to educational resources (Batruch et al., 2022; Castillo et al., 2024; Wiederkehr et al., 2015), while among teachers and parents, similar beliefs predict competitive classroom practices and lower demand for equity-oriented policies (Darnon et al., 2023). A particularly relevant finding is that perceived, but not preferred, school meritocracy predicts lower support for egalitarian and inclusive educational interventions (Darnon, Smeding, et al., 2018), and that belief in school meritocracy is associated with greater acceptance of social-class achievement gaps (Darnon, Wiederkehr, et al., 2018). However, most of this research relies on unidimensional or undifferentiated measures, which means it cannot establish whether these effects are specific to particular dimensions or reflect a more general meritocratic orientation. Our findings suggest that these documented effects are likely dimension-specific: perceived meritocracy is not equivalent to preferred meritocracy, and neither carries the same consequences as perceived privilege or privilege acceptance. Chauvin et al. (2026) similarly highlight multidimensionality by distinguishing ability and equity, but their measure remains confined to perceived meritocracy and applies only to teachers. What the present study adds is a validated four-factor structure applicable to students, which provides the measurement foundation needed to examine whether distinct dimensions of meritocratic belief carry distinct consequences for the educational environment, from students’ sense of justice in school to civic attitudes and the broader climate of fairness, a question the existing literature has not yet been able to address with adequate precision.
The second hypothesis stated that the measurement model would be stable over time, and the results support this expectation. Achieving strict longitudinal invariance indicates that the scale functions equivalently across waves, measuring the same latent constructs at each time point. This is crucial in developmental research because without invariance, apparent change may reflect shifts in how students interpret items or use response categories rather than genuine change in underlying beliefs (Y. Liu et al., 2017; Putnick & Bornstein, 2016; Van De Schoot et al., 2015). With strict invariance in place, observed differences over time can be interpreted as substantive within-student change rather than measurement artifact, strengthening longitudinal inference and supporting models such as latent mean comparisons or growth curve analyses (Putnick & Bornstein, 2016). Substantively, the results suggest that meritocratic belief structures are relatively stable over a one-year window in early-to-mid adolescence, which has implications for when educational interventions targeting these beliefs are likely to produce measurable change. This longitudinal invariance also provides the measurement foundation for subsequent studies to examine how distinct dimensions of meritocratic belief relate to outcomes such as academic motivation, civic attitudes, and school well-being. In this regard, Liu & Wang’s (2025) distinction between education operating as legitimation or as enlightenment, depending on the broader inequality context, points to a productive direction: whether perceived privilege leads to stronger meritocratic preferences or to disengagement from institutional narratives may depend on the degree to which the educational system is experienced as fair, a question that requires dimensionally validated instruments to examine adequately.
The third hypothesis proposed invariance across educational levels. The results did not support full metric or scalar equivalence: non-invariance is concentrated in perceived meritocracy, specifically in how students at different school stages interpret whether effort is rewarded in society. Rather than treating this as a measurement failure, we interpret it as a substantive theoretical finding. Cross-cohort non-invariance indicates that primary and secondary students do not merely differ in their level of perceived meritocracy: they attach qualitatively different meanings to the concept of effort being rewarded. This is consistent with developmental research showing that adolescence is precisely the period when abstract reasoning about justice and fairness becomes more sophisticated and when school-based sorting mechanisms, including grades, tracking, and peer comparison, grow increasingly salient (Henry & Saul, 2006; Resh & Sabbagh, 2014), providing students with a richer and more critical experiential basis for evaluating meritocratic claims. A plausible mechanism is that cumulative exposure to these processes produces a progressive reality shock (Allen, 2016; Tang et al., 2025), such that older students evaluate the same item through a more critical and experience-laden lens. The SES-differentiated trajectories documented by Tang et al. (2025) are instructive here: Chinese students from lower socioeconomic backgrounds initially recognize the weight of structural advantage to a lesser extent than their higher-SES peers, but converge progressively as they accumulate educational experience. If a similar dynamic operates earlier, during secondary school in Chile’s highly stratified system, the non-invariance concentrated in the perceived effort item may partly reflect this convergence process, as students across social positions increasingly confront the limits of effort as a sufficient explanation for outcomes. In Chile’s highly stratified system, where the gap between meritocratic discourse and structural reality is particularly visible, students from lower socioeconomic backgrounds have been shown to experience this gap as a moral grievance (Neut Aguayo et al., 2025), which is consistent with the idea that the construct of perceived effort becomes more contested, and therefore less stable across students, as schooling advances. For researchers, this is a call to move beyond assuming that meritocracy scales travel unchanged across school levels: future work should prioritize stage-sensitive item formulations and partial invariance approaches that can accommodate meaning shifts without sacrificing cross-level comparability (Putnick & Bornstein, 2016; Van De Schoot et al., 2012). For practitioners, including teachers and those designing civic education and school-climate initiatives, the finding indicates that the same meritocratic message may be processed through different interpretive frames depending on students’ accumulated school experience, and that efforts to foster a shared sense of fairness in school need to be calibrated to developmental stage rather than applied uniformly across grade levels.
Taken together, the longitudinal and cross-cohort findings are not contradictory but jointly informative. Longitudinal invariance was established within the same students over a one-year window: within that period, the scale measures the same constructs in the same way, and observed change can be interpreted as substantive. Cross-cohort non-invariance reflects a different comparison, two groups separated by approximately two to three years of accumulated schooling, and points to a longer developmental process through which cumulative school experience progressively reshapes how students interpret whether effort is rewarded. The instrument is therefore well-suited for longitudinal inference within the same school stage, while cross-stage comparisons require caution and, ideally, stage-sensitive measurement approaches.
6 Conclusion
Schools socialize students into beliefs about meritocracy, promoting effort and talent as the legitimate basis of reward while simultaneously exposing students to the weight of family resources and social connections in shaping outcomes (Goudeau & Croizet, 2017; Resh & Sabbagh, 2014; Traini, 2022). Students navigate this tension by holding both orientations at once, endorsing merit as an ideal while recognizing that privilege structures actual outcomes, a configuration the literature describes as dual consciousness (C. Liu & Wang, 2025; Tang et al., 2025). Capturing this complexity requires instruments that distinguish how students perceive meritocracy to work from how they prefer it should work, and that treat merit and privilege as potentially coexisting rather than mutually exclusive, a standard most available measures do not meet. Extending Castillo et al.’s (2023) multidimensional framework to adolescents, this study tests whether students distinguish merit from privilege and perceptions (“what is”) from preferences (“what should be”), using panel survey data from two Chilean cohorts (8th and 10th graders) across two waves (2023-2024), and whether these dimensions are stable over time and comparable across school stages. The contribution is a modest but foundational one: to establish whether meritocratic beliefs can be measured reliably and equivalently in school-aged populations, as a precondition for studying their developmental trajectory and educational consequences.
The results support a four-factor structure, perceived meritocracy, perceived privilege, meritocratic preferences, and privilege acceptance, showing that Chilean adolescents differentiate perceptions from preferences and meritocratic from privilege-based principles, and that merit endorsement can coexist with recognition of structural advantage. Consistent with this, perceived privilege is positively associated with stronger meritocratic preferences, suggesting that recognizing unfair advantages can intensify demands that rewards should follow merit, a normative-reactive pattern that unidimensional instruments would obscure (Castillo et al., 2023). The measurement structure is longitudinally invariant within students, providing a reliable basis for interpreting over-time change as substantive rather than artifactual. At the same time, cross-cohort non-invariance is itself a substantive finding rather than merely a technical caveat: comparisons across educational stages require caution because the perception that “effort is rewarded” does not carry the same meaning for primary and secondary students. This measurement heterogeneity is theoretically consistent with developmental accounts in which cumulative schooling experience progressively reshapes how students interpret meritocratic signals (Allen, 2016; Tang et al., 2025), and it calls for stage-sensitive item formulations in future work rather than treating the current scale as directly comparable across levels.
More broadly, these findings sharpen the way the literature should interpret the effects of meritocracy in educational contexts. When measures conflate perceptions with preferences or collapse merit and privilege into a single continuum, they blur substantively different belief profiles and yield ambiguous conclusions about whether meritocracy is high, low, or declining (Reynolds & Xian, 2014; Zhu, 2025). Part of what is often interpreted as a change in belief in meritocracy may instead reflect shifting recognition of structural constraints rather than a straightforward decline in merit endorsement, an interpretation consistent with recent education-focused accounts of dual consciousness (C. Liu & Wang, 2025; Tang et al., 2025). While typological approaches have documented the coexistence of apparently contrasting beliefs in educational settings (Kwon & Pandian, 2024; Zhu, 2025), the present study shows how that coexistence can be represented within a coherent measurement model that supports longitudinal inference. The combination of strict longitudinal invariance and cross-level non-invariance establishes a key boundary condition: within-student comparisons over time are warranted, whereas comparisons across school stages must account for the shifting meaning of perceived meritocracy.
Although this study draws on a single country, its contributions carry implications beyond the Chilean case. Three findings are likely to travel across educational systems. First, the conceptual distinction between perceived and preferred meritocracy, and between meritocratic and privilege-based principles, reflects general cognitive and normative capacities that develop during adolescence regardless of national context; the four-factor structure should therefore replicate in other highly stratified systems. Second, the coexistence of merit endorsement and recognition of privilege, the core pattern underlying dual consciousness, has been documented in China (C. Liu & Wang, 2025; Tang et al., 2025), the United States and Finland (Zhu, 2025), and Europe (Kwon & Pandian, 2024), suggesting it is not a Chilean peculiarity. Third, the logic of longitudinal invariance testing as a prerequisite for developmental inference is methodologically general and applies to any panel study of meritocratic beliefs. By contrast, the specific pattern of cross-cohort non-invariance, concentrated in perceived effort, may be more system-dependent: it likely reflects the particular intensity and timing of school-based sorting in Chile’s tracked and marketized system. Future research in countries with different tracking regimes and stratification profiles will help clarify which aspects of measurement equivalence are universal and which are conditioned by institutional features.
The findings carry distinct implications for three audiences. For researchers, the results underscore the importance of treating meritocratic beliefs as multidimensional and of routinely testing measurement invariance before making developmental or cross-group comparisons. This matters particularly for equity-oriented research: if measurement models are not equivalent across socioeconomic or generational groups, observed differences in meritocracy may partly reflect differential item interpretation rather than genuine attitudinal divergence, misattributing to individuals what may be a structural artifact of how inequality shapes the meaning of merit. For practitioners, including teachers, school counselors, and those who design civic and citizenship education, the coexistence of merit endorsement and recognition of privilege suggests that students are not passive recipients of meritocratic ideals but actively navigate tensions between what schools tell them about fairness and what they observe in practice. This has direct relevance for civic education and for school climate: rather than simply reinforcing effort-based narratives or, conversely, dismantling them, educators can engage students’ existing dual awareness as a starting point for discussing how merit and privilege coexist, helping cultivate a more critical and grounded sense of justice. Because the meanings students attach to effort and reward shift across school stages, such efforts need to be calibrated to students’ developmental moment rather than applied uniformly. For policymakers concerned with equity, the finding that perceived privilege intensifies meritocratic preferences carries a specific message: students who recognize that rewards do not follow merit do not abandon the ideal of fairness but demand it more strongly. Educational systems that maintain a strong meritocratic discourse while visibly reproducing inequality risk widening the gap between what they promise and what students experience, with potential consequences for students’ sense of justice in school, their trust in educational institutions, and their broader civic engagement.
This study has limitations that point to clear directions for future research. The scale is deliberately minimalist, with two items per factor, which constrains content coverage and reliability and means that claims about subtle within-dimension differences should be treated with caution. Future work should expand and refine the item pool, adding at least one additional indicator per factor and testing whether richer item sets alter the invariance conclusions reported here. With only two waves and two school levels, additional waves and broader age coverage are needed to better trace developmental trajectories and identify when shifts in meaning in perceived meritocracy occur. Cross-level non-invariance calls for differential item functioning analyses and stage-sensitive item formulations to support comparability across educational stages. Most importantly, future research should examine how perceptions and preferences of meritocratic and privilege-based principles relate to the outcomes for which they are most likely to matter in school: students’ sense of justice and fairness, civic attitudes and engagement, school climate, and wellbeing, alongside academic motivation. By providing an empirically supported and longitudinally invariant multidimensional measure for adolescents, this study offers a more reliable, if still provisional, foundation for studying how school-based meritocratic socialization shapes students’ interpretations of fairness and inequality.
7 References
Allen, K. (2016). Top girls navigating austere times: Interrogating youth transitions since the “crisis.” Journal of Youth Studies, 19(6), 805–820. https://doi.org/10.1080/13676261.2015.1112885
Batruch, A., Jetten, J., Van de Werfhorst, H., Darnon, C., & Butera, F. (2022). Belief in school meritocracy and the legitimization of social and income inequality. Social Psychological and Personality Science, 194855062211110. https://doi.org/10.1177/19485506221111017
Bell, D. (1972). On equality: I. Meritocracy and equality. The Public Interest, 29, 29–68.
Bellei, C. (2013). El estudio de la segregación socioeconómica y académica de la educación chilena. Estudios Pedagógicos (Valdivia), 39(1), 325–345. https://doi.org/10.4067/S0718-07052013000100019
Bourdieu, P., & Passeron, J. C. (1990). Reproduction in education, society and culture (Second Edition). Sage Publications Ltd.
Brown, T. A. (2015). Confirmatory factor analysis for applied research (Second edition). New York London: The Guilford Press.
Busemeyer, M. (2014). Skills and inequality: Partisan politics and the political economy of education reforms in western welfare states. Cambridge University Press.
Cabalin, C., Saldaña, M., & Fernández, M. B. (2023). Framing school choice and merit: News media coverage of an education policy in chile. Discourse: Studies in the Cultural Politics of Education, 44(6), 927–942. https://doi.org/10.1080/01596306.2023.2218272
Castillo, J. C., Iturra, J., Maldonado, L., Atria, J., & Meneses, F. (2023). A multidimensional approach for measuring meritocratic beliefs: Advantages, limitations and alternatives to the ISSP social inequality survey. International Journal of Sociology, 53(6), 448–472. https://doi.org/10.1080/00207659.2023.2274712
Castillo, J. C., Laffert, A., Carrasco, K., & Iturra-Sanhueza, J. (2025). Perceptions of inequality and meritocracy: Their interplay in shaping preferences for market justice in chile (2016–2023). Frontiers in Sociology, 10, 1634219. https://doi.org/10.3389/fsoc.2025.1634219
Castillo, J. C., Torres, A., Atria, J., & Maldonado, L. (2019). Meritocracia y desigualdad económica: Percepciones, preferencias e implicancias. Revista Internacional de Sociología, 77(1), 117. https://doi.org/10.3989/ris.2019.77.1.17.114
Chauvin, B., Rauscher, C., Saidah, B., & Louvet, E. (2026). School meritocracy: A multidimensional construct? Evidence from the development of a school meritocracy scale by exploratory and confirmatory bifactor modeling. Current Psychology, 45(5), 519. https://doi.org/10.1007/s12144-026-09134-1
Chen, F. F. (2007). Sensitivity of goodness of fit indexes to lack of measurement invariance. Structural Equation Modeling: A Multidisciplinary Journal, 14(3), 464–504. https://doi.org/10.1080/10705510701301834
Darnon, C., Jury, M., Goudeau, S., & Portex, M. (2023). Competitive and cooperative practices in education: How teachers’ beliefs in school meritocracy are related to their daily practices with students. Social Psychology of Education, 26(6), 1789–1805. https://doi.org/10.1007/s11218-023-09824-9
Darnon, C., Smeding, A., & Redersdorff, S. (2018). Belief in school meritocracy as an ideological barrier to the promotion of equality. European Journal of Social Psychology, 48(4), 523–534. https://doi.org/10.1002/ejsp.2347
Darnon, C., Wiederkehr, V., Dompnier, B., & Martinot, D. (2018). “Where there is a will, there is a way”: Belief in school meritocracy and the social-class achievement gap. British Journal of Social Psychology, 57(1), 250–262. https://doi.org/10.1111/bjso.12214
Davidov, E., Meuleman, B., Cieciuch, J., Schmidt, P., & Billiet, J. (2014). Measurement equivalence in cross-national research. Annual Review of Sociology, 40(Volume 40, 2014), 55–75. https://doi.org/10.1146/annurev-soc-071913-043137
Day, M. V., & Fiske, S. T. (2017). Movin’ on up? How perceptions of social mobility affect our willingness to defend the system. Social Psychological and Personality Science, 8(3), 267–274. https://doi.org/10.1177/1948550616678454
Deng, Y., & Wang, F. (2025). It’s my fault, i should try harder! The narratives of self‐made upward mobility sustain belief in meritocracy in low social mobility context. British Journal of Psychology, bjop.70015. https://doi.org/10.1111/bjop.70015
Duru-Bellat, M., & Tenret, E. (2012). Who’s for meritocracy? Individual and contextual variations in the faith. Comparative Education Review, 56(2), 223–247. https://doi.org/10.1086/661290
Fercovic, M. (2022). Disentangling meritocracy among the long-range upwardly mobile: The chilean case. Sociological Research Online, 27(1), 118–135. https://doi.org/10.1177/1360780420963395
Ferre, J. C. (2023). Welfare regimes in twenty-first-century latin america. Journal of International and Comparative Social Policy, 39(2), 101–127. https://doi.org/10.1017/ics.2023.16
García-Sierra, A. (2023). The dark side of meritocratic beliefs: Is believing in meritocracy detrimental to individuals from low socioeconomic backgrounds? Social Justice Research, 36(4), 385–409. https://doi.org/10.1007/s11211-023-00413-x
Gingrich, J. R. (2011). Making markets in the welfare state: The politics of varying market reforms (1st ed.). Cambridge University Press. https://doi.org/10.1017/CBO9780511791529
Henry, P. J., & Saul, A. (2006). The development of system justification in the developing world. Social Justice Research, 19(3), 365–378. https://doi.org/10.1007/s11211-006-0012-x
Heuer, J.-O., Lux, T., Mau, S., & Zimmermann, K. (2020). Legitimizing inequality: The moral repertoires of meritocracy in four countries. Comparative Sociology, 19(4-5), 542–584. https://doi.org/10.1163/15691330-BJA10017
Imhoff, D., & Brussino, S. (2025). Nociones infantiles sobre desigualdad social: atravesamientos ideológicos y procesos de socialización política. Revista Latinoamericana de Ciencias Sociales, Niñez y Juventud, 13(2). https://doi.org/10.11600/1692715x.1329071114
Janmaat, J. G. (2013). Subjective inequality: A review of international comparative studies on people’s views about inequality. European Journal of Sociology, 54(3), 357–389. https://doi.org/10.1017/S0003975613000209
Jost, J. T., Banaji, M. R., & Nosek, B. A. (2004). A decade of system justification theory: Accumulated evidence of conscious and unconscious bolstering of the status quo. Political Psychology, 25(6), 881–919. https://doi.org/10.1111/j.1467-9221.2004.00402.x
Kline, R. B. (2023). Principles and practice of structural equation modeling. Guilford Publications.
Kwon, R., & Pandian, R. K. (2024). Multidimensionality in merit attitudes: The role of hard work, skills, and social connections in europe. Social Science Research, 122, 103056. https://doi.org/10.1016/j.ssresearch.2024.103056
Lampert, K. (2013). Meritocratic education and social worthlessness. London: Palgrave Macmillan UK. https://doi.org/10.1057/9781137324894
Liu, C., & Wang, J. (2025). Does education legitimise inequality? Comparative analysis of income inequality, education, and meritocratic beliefs. The British Journal of Sociology, 1468–4446.70029. https://doi.org/10.1111/1468-4446.70029
Liu, Y., Millsap, R. E., West, S. G., Tein, J.-Y., Tanaka, R., & Grimm, K. J. (2017). Testing measurement invariance in longitudinal data with ordered-categorical measures. Psychological Methods, 22(3), 486–506. https://doi.org/10.1037/met0000075
McCoy, S. K., & Major, B. (2007). Priming meritocracy and the psychological justification of inequality. Journal of Experimental Social Psychology, 43(3), 341–351. https://doi.org/10.1016/j.jesp.2006.04.009
Mijs, J. (2016). The unfulfillable promise of meritocracy: Three lessons and their implications for justice in education. Social Justice Research, 29(1), 14–34. https://doi.org/10.1007/s11211-014-0228-0
Mijs, J. (2019). The paradox of inequality: Income inequality and belief in meritocracy go hand in hand. Socio-Economic Review, 19(1), 7–35. https://doi.org/10.1093/ser/mwy051
Mijs, J. (2021). The paradox of inequality: Income inequality and belief in meritocracy go hand in hand. Socio-Economic Review, 19(1), 7–35. https://doi.org/10.1093/ser/mwy051
Mijs, J. (2026). Visualizing belief in merit and privilege, 1930 to 2020: rejoinder. Socius: Sociological Research for a Dynamic World, 12, 23780231261426413. https://doi.org/10.1177/23780231261426413
Moyano Dávila, C., Alarcón-Arcos, S., Angelcos, N., Castillo, J. C., & Salgado, M. (2025). Merit as an attitude: Chilean school communities’ repertoires in chile and the perception of the “good student” in a post-pandemic scenario. Social Justice Research, 38(3), 264–289. https://doi.org/10.1007/s11211-025-00456-2
Neut Aguayo, P., Rivera-Vargas, P., & Miño Puigcercós, R. (2025). Neoliberalism and school inequality in the chilean education system: An analysis from the voices of lower-class, middle-class, and upper-class students. International Studies in Sociology of Education, 1–16. https://doi.org/10.1080/09620214.2025.2558938
OECD. (2025). Trends shaping education 2025. OECD Publishing. https://doi.org/10.1787/ee6587fd-en
Otero, G., Carranza, R., & Contreras, D. (2017). “Neighbourhood effects” on children’s educational achievement in chile: The effects of inequality and polarization. Environment and Planning A: Economy and Space, 49(11), 2595–2618. https://doi.org/10.1177/0308518X17731780
Putnick, D. L., & Bornstein, M. H. (2016). Measurement invariance conventions and reporting: The state of the art and future directions for psychological research. Developmental Review, 41, 71–90. https://doi.org/10.1016/j.dr.2016.06.004
Resh, N., & Sabbagh, C. (2014). Sense of justice in school and civic attitudes. Social Psychology of Education, 17(1), 51–72. https://doi.org/10.1007/s11218-013-9240-8
Reynolds, J., & Xian, H. (2014). Perceptions of meritocracy in the land of opportunity. Research in Social Stratification and Mobility, 36, 121–137. https://doi.org/10.1016/j.rssm.2014.03.001
Sandel, M. J. (2020). The tyranny of merit: What’s become of the common good? (First edition). New York: Farrar, Straus and Giroux.
Sigelman, C. K. (2013). Age differences in perceptions of rich and poor people: Is it skill or luck? Social Development, 22(1), 1–18. https://doi.org/10.1111/sode.12000
Svetina, D., Rutkowski, L., & Rutkowski, D. (2020). Multiple-group invariance with categorical outcomes using updated guidelines: An illustration using m plus and the lavaan/semTools packages. Structural Equation Modeling: A Multidisciplinary Journal, 27(1), 111–130. https://doi.org/10.1080/10705511.2019.1602776
Tang, M., Li, A., & Wu, X. (2025). Meritocratic myth in mind? Socioeconomic backgrounds and shifting beliefs about meritocracy among college students in china. Sociology of Education, 00380407251360432. https://doi.org/10.1177/00380407251360432
Tejero-Peregrina, L., Willis, G., Sánchez-Rodríguez, Á., & Rodríguez-Bailón, R. (2025). From perceived economic inequality to support for redistribution: The role of meritocracy perception. International Review of Social Psychology, 38(1), 4. https://doi.org/10.5334/irsp.1013
Traini, C. (2022). The stratification of education systems and social background inequality of educational opportunity. International Journal of Comparative Sociology, 63(1-2), 10–29. https://doi.org/10.1177/00207152211033015
Valenzuela, J. P., Bellei, C., & Ríos, D. D. L. (2013). Socioeconomic school segregation in a market-oriented educational system. The case of chile. Journal of Education Policy, 29(2), 217–241. https://doi.org/10.1080/02680939.2013.806995
Van De Schoot, R., Lugtig, P., & Hox, J. (2012). A checklist for testing measurement invariance. European Journal of Developmental Psychology, 9(4), 486–492. https://doi.org/10.1080/17405629.2012.686740
Van De Schoot, R., Schmidt, P., De Beuckelaer, A., Lek, K., & Zondervan-Zwijnenburg, M. (2015). Editorial: Measurement invariance. Frontiers in Psychology, 6. https://doi.org/10.3389/fpsyg.2015.01064
Van De Werfhorst, H. G. (2024). Is meritocracy not so bad after all? Educational expansion and intergenerational mobility in 40 countries. American Sociological Review, 89(6), 1181–1213. https://doi.org/10.1177/00031224241292352
Wiederkehr, V., Bonnot, V., Krauth-Gruber, S., & Darnon, C. (2015). Belief in school meritocracy as a system-justifying tool for low status students. Frontiers in Psychology, 6. Retrieved from https://www.frontiersin.org/articles/10.3389/fpsyg.2015.01053
Wiesner, T., & Sachweh, P. (2026). Effort versus advantage: Visualizing (relative) belief in meritocracy, 1930 to 2022—a comment on mijs (2018), 12. https://doi.org/10.1177/23780231261425841
Wu, H., & Estabrook, R. (2016). Identification of confirmatory factor analysis models of different levels of invariance for ordered categorical outcomes. Psychometrika, 81(4), 1014–1045. https://doi.org/10.1007/s11336-016-9506-0
Young, M. (1958). The rise of the meritocracy. New Brunswick, N.J., U.S.A: Transaction Publishers.
Zhu, L. (2025). Meritocratic beliefs in the united states, finland, and china: A multidimensional approach using latent class analysis. The British Journal of Sociology, 76(1), 153–172. https://doi.org/10.1111/1468-4446.13152
Footnotes
Factor loadings and fit indices for the separate analytical samples (i.e., per wave and cohort level) are available in the Supplementary Material.↩︎
Table 2 totals (N = 1,508) correspond to the sum of person-wave observations (846 at T1 + 662 at T2), rather than unique individuals.↩︎
Factor loadings and fit indices for the separate analytical samples (i.e., per wave and cohort level) are available in the Supplementary Material.↩︎