Data, variables and method
Data
This study is based on data from the Panel Survey on Education and Meritocracy (EDUMER) from the 2023 (N = 840) and 2024 (N = 840) waves, focusing on students. The database is derived from web-based questionnaires administered to sixth and ninth-grade students from nine schools in the Metropolitan and Valparaíso regions of Chile. Due to the non-probabilistic nature of the data collection process, the design does not allow for representativeness of the target population; therefore, analyses and conclusions can only be drawn at the sample level.
Variables
Scale of Perceptions and Preferences on Meritocracy: The variables included in the measurement model for meritocratic and non-meritocratic perceptions and preferences are operationalized according to the items proposed by Castillo et al. (2023). Perception of meritocracy is measured by two items that assess the level of agreement with the idea that effort and ability are rewarded in Chile, while non-meritocratic perception is measured by two items evaluating the agreement that success is linked to connections and family wealth. Preference for meritocracy is measured by two items that evaluate agreement with the idea that those who work harder or are more talented should be more rewarded. Preference for non-meritocratic aspects is measured by two indicators assessing agreement that it is acceptable for individuals with better connections or wealthy parents to achieve greater success. Each item is rated on a four-point Likert scale ranging from “strongly disagree” (1) to “strongly agree” (4).
| Components | Dimensions | Item (English) | Item original (Spanish) |
|---|---|---|---|
| Perception | Meritocratic | In Chile people are rewarded for their efforts | En Chile las personas son recompensadas por sus esfuerzos |
| In Chile people are rewarded for their intelligence and ability | En Chile las personas son recompensadas por su inteligencia y habilidad | ||
| Non meritocratic | In Chile those with wealthy parents do much better in life | En Chile a quienes tienen padres ricos les va mucho mejor en la vida | |
| In Chile those with good contacts do much better in life | En Chile a quienes tienen buenos contactos les va mejor en la vida | ||
| Preference | Meritocratic | Those who work harder should reap greater rewards than those who work less hard | Quienes más se esfuerzan deberían obtener mayores recompensas que quienes se esfuerzan menos |
| Those with more talent should reap greater rewards than those with less talent | Quienes poseen más talento deberían obtener mayores recompensas que quienes poseen menos talento | ||
| Non meritocratic | It is okay that those who have rich parents do better in life | Está bien que quienes tengan padres ricos les vaya mejor en la vida | |
| It is okay that those who have good contacts do better in life | Está bien que quienes tengan buenos contactos les vaya mejor en la vida |
Preferences for Market Justice: This construct is measured using three variables that address the level of justification for whether access to social services in healthcare, pensions, and education should be income-based. The justification for inequality in healthcare is measured by the item: “Is it acceptable for those who can pay more to have better access to healthcare?” The same question is posed for pensions and education. In all cases, respondents indicate their preferences on a Likert scale ranging from “strongly disagree” (1) to “strongly agree” (4). Additionally, a summary indicator of “preferences for market justice” is included, measured by an average index of all these items (α = 0.83), with values ranging from 1 to 4, where higher values represent greater preferences for market justice.
| Component | Dimensions |
|---|---|
| Market justice preferences | Justice in education |
| Justice in health | |
| Justice in pensions |
Methods
To evaluate the underlying structure of the scale, we employed Confirmatory Factor Analysis (CFA) based on a measurement model with four latent factors, as proposed by Castillo et al. (2023), using Diagonally Weighted Least Squares (DWLS) 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 several indices: the Comparative Fit Index (CFI) and the Tucker-Lewis Index (TLI), with acceptable values above 0.95; the Root Mean Square Error of Approximation (RMSEA), with values below 0.06 indicating good fit; and the Chi-square statistic (acceptable fit indicated by p > 0.05 and a Chi-square/df ratio < 3).
To examine the longitudinal metric stability of the measurement model [@davidov_measurement_2014], we conducted a measurement invariance analysis across the two waves of the study. Following a hierarchical approach as outlined by Liu et al. (2017), we tested four nested models: configural (equivalence in factor structure), metric or weak (equality of factor loadings), scalar or strong (equality of intercepts), and strict (equality of residual variances). This stepwise procedure is essential when using ordered categorical indicators, as failing to account for their non-continuous nature may distort parameter estimates.
In addition to the traditional Chi-square difference test, we used changes in fit indices to assess invariance: a change in CFI (ΔCFI ≥ −0.010) and RMSEA (ΔRMSEA ≤ 0.015) were adopted as more robust and sample-size-independent criteria, as recommended by Chen (2007).
To evaluate the external validity of the measurement model, we conducted regression analyses within a structural equation modeling (SEM) framework. This allowed us to investigate the relationships between latent constructs—meritocratic and non-meritocratic beliefs—and preferences for market-based justice in healthcare, pensions, and education, in line with prior research (Castillo et al., 2024).
All analyses were performed using the lavaan package in R version 4.2.2. The hypotheses of this research were pre-registered on the Open Science Framework (OSF), hosted by the Center of Open Science.