Abstract
The purpose of this study was to evaluate the psychometric evidence for Achievement Motivation and General Mattering scores among Latinx college students in Mexico and the United States based on degree of measurement invariance. Validating instruments in cross-cultural contexts is essential to ensure that psychological constructs are measured reliably across populations with distinct cultural experiences, such as family dynamics, schooling, mental health, and acculturation. A total of 505 participants, with 263 from Mexico and 242 from the United States, participated in the study (U.S. sample: 71.9% female, ages 18–57, M = 23.65; Mexican sample: 55.9% female, ages 15–56, M = 26.69). The results indicated that the Spanish version of the Achievement Motivation scale had partial strict invariance, and the General Mattering scale had metric invariance when used with these two groups of Spanish-speaking Latinx college students. This study highlighted the need of using accurate, trustworthy, and culturally sensitive measures to comprehend the psychological experiences of Latinx college students.
Validation of instruments in other languages is a significant aspect of cross-cultural research. This type of validation provides that results collected from different cultural and linguistic backgrounds are comparable, thereby contributing to the building of knowledge across nations, cultures, and languages. Furthermore, validation of measures for use in different languages is beneficial for practitioners and researchers who are interested in conducting research across or within cultures to address various social phenomena (Karaman et al., 2017, 2019; Leyva-Moral et al., 2021).
In recent years, researchers (Chen & Lu, 2015; Dueñas & Gloria, 2017; Elfreich, 2019; Karaman, 2023) conducted studies on the psychological constructs of achievement motivation and general mattering due to their impact on mental health, educational outcomes, and overall well-being. Researchers have demonstrated that measurement construct validity can vary across different cultural and national contexts. Instruments that may work effectively in one cultural setting may not fully capture the experiences of individuals in another due to variations in cultural norms, values, and social structures (Chen & Lu, 2015). For instance, constructs such as achievement motivation and mattering may be interpreted differently across cultures, particularly between Latinx students in the United States and Mexico. These differences are influenced by distinct educational systems, socioeconomic conditions, and familial expectations, which shape students’ perceptions of academic success and personal significance (Gonzalez, 2024; Karaman et al., 2017, 2020; Suarez-Orozco, 1993).
In Mexico, for example, students may experience achievement motivation as closely tied to family obligations and the need to contribute to the household (Aguirre-Zazueta & Ontiveros-Herrera, 2024). Aguirre-Zazueta and Ontiveros-Herrera (2024) found that cultural expectations around family and community played a significant role in nursing students’ academic motivation and performance. In contrast, Latinx students in the United States may experience achievement motivation differently due to their navigation of bicultural identities, acculturation, and the pressures of being first-generation college students (Cervantes Arreola et al., 2018). Furthermore, general mattering can be shaped by distinct cultural narratives about collectivism and individualism (Cienfuegos-Martínez et al., 2016). In Mexican culture, the sense of mattering may be more strongly linked to familial and community ties. For instance, Díaz Gaspar et al. (2024) emphasizes how study habits and motivation are connected to familial expectations in Mexican public universities. On the other hand, Dueñas and Gloria (2020) stated that Latinx students in the United States may experience mattering in a more individualistic framework, shaped by the broader societal emphasis on self-actualization and independent achievement.
For Latinx students, particularly those in the United States and Mexico, both constructs are significant in understanding their academic motivation and psychosocial well-being. However, despite the importance of these constructs, researchers developed and validated existing instruments on predominantly White, English-speaking populations (Karaman et al., 2019). This lack of cultural adaptation raises concerns about the reliability and validity of these scales when used with Latinx populations, who may have different cultural and linguistic contexts.
Understanding the factor structure of psychometric instruments is beneficial when working with Latinx students who might have unique cultural experiences and perspectives as related to family (Kwak et al., 2017; Villarreal et al., 2005), language practices (Cavazos & Karaman, 2023), K-12 schooling experiences with teachers and school counselors (Vela et al., 2015), mental health (Vela et al., 2020), and acculturation (Cuellar et al., 1995; Lindsey et al., 2022).
The current study focuses on the validation of the Achievement Motivation Measure (AMM; Smith et al., 2020) and General Mattering Scale (GMS; Marcus, 1991) in Mexican Spanish. The rationale of the current study is grounded in the importance of achievement motivation and mattering among Latinx population who speak Mexican Spanish. The validation of AMM and GMS in Mexican Spanish may ensure cultural appropriateness, enhance inclusivity, and improve communication within diverse Latinx cultures in the U.S. and Mexico.
Achievement Motivation
McClelland et al. (1953) defined achievement motivation as a psychological drive to excel in one’s work and endeavors. McClelland’s early publications on achievement motivation outlined the hypothesis that action is motivated by a desire for success and a fear of failure (McClelland, 1965; McClelland et al., 1953). McClelland’s achievement motivation theory has had a wide impact, laying the groundwork for future research (see Karaman, 2023; Karaman et al., 2019; Karaman & Watson, 2017; Smith & Troth, 1975; Vu et al., 2022). Alschuler and Irons (1973) expanded achievement motivation research by including achievement motivation teaching strategies, arguing that educational environments could generate better achievement motivation among students. Alschuler and Irons (1973) stated that such educational techniques can better equip individuals to face problems and strive for greater success later in life. Smith (2011, 2015) expanded on this concept by presenting an evidence-based method to achievement motivation training and claiming that it can improve performance. By offering a contextual approach to achievement motivation, Smith (2015) argued that understanding an individual’s environment and personal context is critical to evaluating their motivation levels.
Researchers (Elliott & Bempechat, 2002; Karaman et al., 2017; Trumbull & Rothstein-Fisch, 2011) addressed the cultural context of achievement motivation from different perspectives. Karaman et al. (2017) emphasized how achievement motivation varied across different cultures (i.e., Arabic, Turkish, U.S.), suggesting the significance of cultural factors in shaping an individual’s drive to succeed. Martin et al. (2017) and Vu et al. (2022) stressed the cyclical nature of achievement motivation in learning. They explored a dynamic model where motivation and achievement interact and impact each other over time, providing a new direction for research in achievement motivation. In Bataeva’s (2018) study, McClelland’s achievement motivation theory was also applied to analyze the motivational content of educational materials. Bataeva’s (2018) work demonstrated the usefulness of McClelland’s theory in understanding the motivational context of educational resources, which can have significant implications for educational practices. Moreover, in a neuroscientific study, Rybnicek et al. (2019) demonstrated the biological basis of McClelland’s theory, providing a new perspective on how individual needs impact motivation effects.
In addition to the aforementioned studies, the scale developed and validated by Smith et al. (2020) with undergraduate and graduate students has also opened the door to many studies (see Atik et al., 2024; Karaman, 2023; Sari & Yetkiner, 2020). Specifically, the consideration of achievement motivation from both cognitive and behavioral perspectives allows for the application of these dynamics in various fields of social sciences. The scale, which was also used in our current study, with its Mexican Spanish adaptation, will provide opportunities for studies conducted with the Latinx population in both Mexico and the United States. Achievement motivation is a complex concept with important theoretical and practical ramifications. Understanding its origins, impact, and application can help to guide treatments targeted at improving individual accomplishment and performance. It is beneficial to create and validate psychometrically sound and culturally adapted instruments to attain this goal.
Achievement Motivation in Latinx Culture
Understanding the dynamics of achievement motivation, particularly in the context of Latinx culture, has philosophical and intellectual implications in modern society. The need to achieve goals, overcome challenges, and strive for perfection is embodied by achievement motivation. Although extensive studies on achievement motivation in diverse populations (e.g., S. Y. Chen & Lu, 2015; Karaman, 2023; Karaman & Watson, 2017) have been conducted, there is a lack of literature understanding its dynamics within the Latinx culture.
Researchers (Ibañez et al., 2004; Suarez-Orozco, 1993; Triandis et al., 1982) from different perspectives used diverse variables to explain the relationship among social, psychological, and educational constructs in Latinx culture and achievement motivation. Suarez-Orozco (1993) emphasized the relevance of cultural and familial influences in developing achievement motivation among Mexican immigrant and Mexican American teenagers. Similarly, Ibañez et al. (2004) identified perceived school experiences and parental participation as important determinants in Latinx adolescents’ motivation, particularly those who were English-speaking or US-born.
In recent years, studies conducted with adult groups have focused on the relationships between university students’ motivation and social, psychological, and educational constructs. Cervantes Arreola et al. (2018) explored differences in academic self-efficacy, psychological well-being, and achievement motivation among university students with high and low academic performance in Mexico. The study showed that students with higher academic performance tended to have higher self-efficacy and achievement motivation, which were related to their sense of personal significance or mattering within their educational and social contexts. It also emphasized the role of psychological well-being as a mediator between motivation and performance. In another study, Díaz Gaspar et al. (2024) found that study habits were influenced by students’ perceptions of support from family and their motivation to succeed. The research highlighted that Mexican students’ sense of achievement motivation was often tied to societal expectations and academic pressures.
Because most achievement motivation measures were created and validated with non-Latinx White groups, researchers (Gizdarska, 2017; Herrero, 2014) emphasized the need for additional research. This calls into question the cultural relevance and sensitivity of these methods for the Latinx population. Hence, research on achievement motivation in the Latinx culture can help to inform the development of culturally sensitive assessment methods, thereby improving their psychometric properties. By recognizing and appreciating cultural differences in achievement motivation, this research can greatly contribute to encouraging diversity and tolerance in educational and work contexts.
Mattering
Mattering, as defined broadly, is the state of significance and importance (Rosenberg & McCullough, 1981). Social scientists conceptualized mattering as being integral to our self-perception by attributing an element of importance to our existence in the lives of others and the world at large in the past two decades (G. Elliott et al., 2004).
Rosenberg initially introduced the construct of mattering in the early 1980s, suggesting that individuals experience a sense of general mattering (our significance to society) and interpersonal mattering (our significance to specific individuals) (Rosenberg & McCullough, 1981). In their pioneer study, Rosenberg and McCullough (1981) found a positive association between mattering and self-esteem and a negative association with depressive symptomology, anxiety, negative affect, and deviant behavior. They determined this sentiment originates from four sources: attention, importance, dependence, and ego-extension. Attention is the perception that others notice what we do, while importance is the perception that our actions affect others. Dependence is the self-perception that others need us for their well-being. Ego-extension refers to an individual’s belief that others care about them and their actions significantly affect others’ emotions. If they fall ill, win an award, or pass away, the feelings of others will be greatly impacted (Rosenberg & McCullough, 1981).
Recent longitudinal and evaluative studies have explored mattering as a positive experience that enhances the sense of well-being and is inversely related to depression and worry (France & Finney, 2009; Taylor & Turner, 2001). Elliott et al. (2004) conducted an empirical validation study wherein the constructs of awareness, importance, and reliance defined mattering similarly to Rosenberg’s and Rosenberg and McCullough’s (1981) original conceptualization. Their findings aligned with Rosenberg’s work, identifying mattering as a primary motivator in self-concept (Rosenberg & McCullough, 1981). Drawing from this information, Rayle (2006) proposed significant implications for counselors, emphasizing that the phenomenon of mattering is vital for the counseling relationship and requires further exploration.
Mattering in Latinx Culture
Mattering has a positive impact on an individual’s self-esteem, motivation, and mental health (Dueñas & Gloria, 2017; Elfreich, 2019; Elliott et al., 2011). The idea of mattering has diverse interpretations and connotations within the sphere of Latinx culture, which is distinguished by its specific social practices, beliefs, and community orientation. A thorough study of what matters in Latinx culture is needed and may provide significant insights for promoting cultural competency, inclusion policies, and psychological treatments, resulting in a more profound understanding of this heterogeneous population.
In recent years, researchers have conducted studies on the Latinx population, focusing on how feelings of significance or “mattering” relate to various psychological and social factors. These studies have laid the groundwork for further research, as seen in the present study. For example, Dueñas and Gloria (2017, 2020) found belonging and cohesiveness play an important role in establishing a sense of belonging among Latinx undergraduate students. Researchers have revealed the relationship among university belonging, collective self-esteem, and mattering by analyzing psychological, social, and cultural aspects, with belonging mediating the relationships of cohesiveness and congruity with mattering, respectively. A sense of belonging was important for first-generation Latinx students, who had different college experiences and perspectives on higher education than their continuing-generation peers (Gloria & Castellanos, 2012).
Silas-Casillas and Perales-Franco (2014) enhanced general mattering and focused on how social relationships students’ educational experiences and outcomes in Mexico. The quality of relationships among school principals, teachers, parents, and students significantly influenced academic performance and retention rates in marginalized communities. The study highlighted the importance of generative relationships—where school actors engage in collaborative, supportive interactions that foster a shared sense of purpose and commitment to student success. In schools with generative processes, students were more likely to experience a sense of mattering, as they feel valued and supported by their teachers, peers, and parents. This feeling of importance, in turn, enhanced achievement motivation by creating an environment in which students believe that their efforts were recognized and appreciated.
Purpose of the Study
The purpose of the current study was to find evidence of construct validity and psychometric properties for the Spanish version of the AMM and GMS, which have been used widely in the fields of psychology and education, in Spanish Mexican. While conducting this study, validity and reliability procedures were carried out with the Latinx group living in two different countries, in accordance with American Educational Research Association, American Psychological Association, & National Council of Measurement in Education (2014) standards. In line with the goal of the study, it was also aimed to test the measurement invariance of the measurement tools and the differences between the groups. In this vein, answers to the following research questions have been sought:
Are the scores of the Mexican Spanish version of AMM and GMS valid and reliable?
Does the Mexican Spanish version of AMM and GMS have factorial invariance by nationality (Mexican and U.S. Latinx)?
Is there a difference between Mexican Latinxs and U.S. Latinxs based on GMS and AMM scores?
Method
Participants and Procedure
The current study included a total of 505 participants from a university located at the northwest Mexico and a Latinx Serving Institution located at the southwest U.S. The Mexican sample consisted of 263 participants, with a majority being women (55.9%, n = 147) and men comprising 43.7% (n = 115). One person did not complete the gender question. The participants’ ages varied between 15 and 56 years (M = 26.69, SD = 7.37). The U.S. sample included 242 participants, predominantly women (71.9%, n = 174), with men representing 28.1% (n = 68). The participants’ ages varied between 18 and 57 years (M = 23.65, SD = 5.97).
The research protocol for this study was approved by the institutional review board of the university where the data was collected. Prior to participation, all participants, including those identified as part of the Mexican and U.S. samples within the university, were fully informed about the study’s purpose, procedures, potential risks, and benefits. Informed consent was obtained from each participant, ensuring comprehension and voluntary participation. The consent process adhered to the ethical standards set forth by the university, upholding the integrity of the research and the rights of all participants. The translation process included forward and backward methods. Initially, a counseling graduate student who was a Mexican American and could write, read and communicate in Mexican Spanish fluently translated the AMM and GMS from English to Spanish. A colleague who was a faculty member at the Spanish Translation and Interpreting under the Department of Writing and Language studies reviewed the Spanish version for accuracy. Then, a sworn translator translated the document from Spanish to English. The researchers checked the backward translation and approved for data collection.
Measures
Achievement Motivation Measure
The Achievement Motivation Measure was created from the Achievement Motivation Inventory (Smith, 1972). The Achievement Motivation Measure encompasses 13 items divided into two factors, aligning with the theoretical principles of Achievement Motivation outlined by McClelland et al. (1953). These two factors assess the achievement-oriented thought processes (called achievement thoughts [AT] with 9 items) and behaviors characteristic of high performers (called achievement behaviors [AB] with 4 items) as detailed by McClelland (1961). The Achievement Motivation Measure has a 5-point Likert-type scale, ranging from 0 (never) to 4 (always). The lowest possible score is 0, while the highest possible score is 52 (Smith et al., 2020). Higher scores are indicative of strong achievement motivation, while lower scores suggest low levels of motivation. The 13-item scale includes statements like “I can keep my mind on a task for a long period of time” and “I like to undertake projects that involve some risk.” The AMM takes approximately 5 min to complete.
Smith et al. (2020) validated the AMM instrument through two separate analyses using different data sets. The first study, which included 303 graduate and undergraduate students from a university setting, conducted an Exploratory Factor Analysis (EFA) using principal component analysis with a Promax rotation. A Confirmatory Factor Analysis (CFA) was run on the second data set, which comprised 329 participants, to assess the model from the previous analysis. The CFA results suggested a satisfactory model fit.
For additional validation, Smith et al. (2020) applied Cronbach’s Alphas to evaluate reliability estimates for the normative sample. The results showed moderate reliability coefficients for Achievement Thoughts (0.80) and a lesser degree for Achievement Behavior (0.60). The overall alpha coefficient for the AMM was .83.
General Mattering Scale
The General Mattering Scale was developed by Marcus (1991) to measure the extent to which individuals perceive their importance to others. The instrument employs a 5-point Likert scale ranging from “Very Much” to “Not at All,” producing a single scale score that is representative of an individual’s perceived mattering. Scores range from a minimum of five to a maximum of twenty, with a higher score indicating a higher sense of mattering.
The degree of mattering is determined by responses to questions such as, “How important do you feel you are to other people?” and “How interested are people generally in what you have to say?” The reliability of the GMS, as evidenced by Haktanir et al. (2016), had a moderate Cronbach’s alpha of .74. Rayle and Myers (2004), in their evaluation of the original scale, found Cronbach’s alpha coefficients ranging from .74 to .86 within a population of college students. This range demonstrates the reliability and stability of the GMS when applied to diverse student populations.
Data Analysis
In this study, we examined univariate and multivariate normality for all items in the two scales across the two samples and then assessed Cronbach alpha as internal consistency for both scales across the two populations. Then, we sequentially ran a series of models to test structural invariances of the scales for both populations. We followed the four steps suggested by Widaman and Reise (1997) for testing measurement invariance. First model is configural model also known as multi-group confirmatory factor analysis (CFA). This model runs CFA for both groups without constraining any parameters to be equal. We used the results of this model to answer research question 1. The second model is metric model which constraints factor loadings to be equal across the groups. The third model is scalar model which constraints factor loadings and intercepts to be equal across the populations. Fourth model is strict model which constrains factor loadings, intercepts, and residual variance to be equal. As commonly used in the literature, we used chi-square difference test (ΔX2) between any of the nested models. It is important to note that when full invariance was not supported, we also tested partially invariant models (e.g., adjusted model) before moving to the next model. We used the results of these models to answer research question 2. When evaluating model fit indices, we used the criteria suggested by Hu and Bentler (1999) (Comparative fit index [CFI] > 0.95; Tucker-Lewis index [TLI] > 0.95; and Root Mean Square Error of Approximation [RMSEA] < 0.06). We used R program and specifically used “lavaan” (Rosseel, 2012) and “semTools” (Jorgensen et al., 2020) packages for the measurement invariance analyses. We used the weighted least squares mean, and variance adjusted (WLSMV) estimator, with scaled chi-square test statistics reported accordingly. Lastly, we conducted independent samples t-test to find differences in AMM and GMS scores between the samples. We used the results of t-tests and Hedges’ g statistics to answer research question 3.
Results
Normality Statistics
We assessed univariate normality for each item by examining skewness, kurtosis, and normal Q-Q plots. In the US sample, skewness values for the AMM and GM scales ranged from −0.482 to 1.236, while kurtosis ranged from −0.294 to 1.791; in the Mexican sample, skewness ranged from −0.729 to 1.531, and kurtosis from −1.102 to 1.631. Normal Q-Q plots showed close alignment with the linear line for all items, and a few multivariate outliers were identified through Mahalanobis distance, but no participants were excluded. Overall, the items followed an approximately normal univariate distribution.
Internal Consistency
For the US sample, the Cronbach’s alpha for the Achievement Thoughts subscale is .81, for the Achievement Behaviors subscale is .63, and the overall scale level alpha is .83. For the Mexican sample, the Cronbach’s alpha for the Achievement Thoughts subscale is .82, for the Achievement Behaviors subscale is .70, and the overall scale level alpha is .87.
For the General Mattering Scale, the Cronbach’s alpha for the US sample is .77, and for the Mexican sample, it is .73. Cronbach alpha values as internal consistency were similar. The results showed that Spanish versions of both scale scores were reliable.
Measurement Invariance for Achievement Motivation Measure
It is important to note that the invariance of AMM was tested as a two-dimensional factor model, with Achievement Thoughts (AT) and Achievement Behaviors (AB) as the two latent constructs. Table 1 shows model fit information across the different models for the achievement motivation scale with scaled chi-square values. Table 2 shows the scaled chi-square model comparisons between the nested models given in Table 1. In the configural model for the Achievement Motivation Measure (AMM), we estimated factor loadings for both latent constructs: Achievement Thoughts (AT) and Achievement Behaviors (AB). For Mexican sample, all factor loadings for the items on the AT and AB constructs were statistically significant (p < .001), with standardized loadings ranging from 0.473 to 0.693 for AT and 0.653 to 0.675 for AB. Similarly, for US sample, the factor loadings were also statistically significant (p < .001), with standardized loadings ranging from 0.399 to 0.710 for AT and 0.493 to 0.610 for AB. Regarding the latent variances, in Mexican sample, the variances of AT and AB were significant (p < .001). For the US sample, the latent variances of AT and AB were also significant (p < .001). The configural model resulted in very good model fit indices and proved that both versions of AMM had the same structure. The results showed that metric model was significantly worse than configural model; ΔX2 = 23.29, df = 10, p = .009. Based on the modification indices, items 9 and 10—both unconstrained in the partial metric and scalar models—belong to the Achievement Thoughts subscale of the Achievement Motivation Measure (AMM). Item 9 (“When working on a committee, I like to see that plans are followed through efficiently”) and item 10 (“I prefer to know how I am progressing by obtaining concrete feedback when working on a task”) reflect key aspects of goal-directed behavior and feedback orientation. The comparison of this partial metric model and configural model indicated that chi-square difference test was no longer significant; ΔX2 = 10.95, df = 8, p = .204. Additionally, the model fit indices for the partial scalar model indicated a very good fit (see Table 1).
Model Fit Information for the Motivation Scale Across the Different Models.
Chi-Square Model Comparison for the Motivation Scale.
Note. ΔX2 is the scaled difference; *statistically significant at the alpha level of .05.
Even though good model fit indices of scalar model, the chi-square difference test between partial metric model and scalar model was significant; ΔX2 = 79.81, df = 10, p < .001. This means that scalar model was significantly worse than partial metric model. In order to improve model fit for the scalar model and establish partial scalar model, we allowed intercept parameters for item 1, item 5, item 9, item 10, and item 11 to vary across the two samples. The comparison of this partial scalar model and partial metric model indicated that chi-square difference test was no longer significant; ΔX2 = 6.80, df = 5, p = .235. Additionally, the model fit indices for the partial scalar model indicated an excellent fit.
Even though good model fit indices of strict model, the chi-square difference test between partial scalar model and strict model was significant; ΔX2 = 46.48, df = 12, p < .001. This means that strict model was significantly worse than partial scalar model. In order to improve model fit for the strict model and establish partial strict model, we allowed residual parameters for item 1, item 9, and item 11 to vary across the two samples. The comparison of this partial strict model and partial scalar model indicated that chi-square difference test was no longer significant; ΔX2 = 14.88, df = 9, p = .094. Besides, model fit indices for the partial strict model indicated absolute fit. Therefore, we can conclude that the Motivation Scale indicates partial strict invariance across the American and Mexican groups.
An independent samples t-test showed that there was no significant difference in the total motivational achievement scores for US (M = 39.87, SD = 6.82) and Mexicans (M = 39.95, SD = 8.06), t(462) = 0.126, p = .90. We also calculated Hedges’ g to assess the effect size of the difference in motivation scores between the Mexican and US samples. The effect size was g = 0.012, indicating a negligible difference between the two groups. The 95% confidence interval for Hedges’ g ranged from −0.171 to 0.194.
Measurement Invariance for General Mattering Scale
Table 3 shows model fit information across the different models for the general mattering scale. Table 4 shows chi-square model comparisons between the nested models given in Table 3. Configural model resulted in very good model fit indices and proved that both versions of GMS had the same structure. The results showed that the chi-square difference test between scalar and metric models was significant; ΔX2 = 14.47, df = 4, p = .006. Despite good model-fit indices found for the scalar model, the statistically significant chi-square difference suggests that scalar model is significantly worse than metric model. In other words, there is a lack of scalar invariance for the General Mattering Scale. Furthermore, modification indices did not suggest any potential intercept parameters to be freely estimated across the American and Mexican samples. Therefore, we could not establish partial scalar model, and metric model was the best trade-off between model fit and model complexity. An independent samples t-test showed that there was significant difference in the total general mattering scores for US (M = 14.90, SD = 2.84) and Mexicans (M = 15.52, SD = 2.83), t(496) = 2.47, p = .013. The effect size of Hedges’ g was 0.222, indicating a small difference between the two groups. The 95% confidence interval for Hedges’ g ranged from 0.046 to 0.399.
Model Fit Information for the General Mattering Scale Across the Different Models.
Chi-Square Model Comparison for the General Mattering Scale.
Note. See notes in Table 2.
Discussion
We aimed to evaluate the psychometric evidence for Achievement Motivation and General Mattering scores among Latinx college students in Mexico and the United States based on degree of measurement invariance. Our findings are promising for use of Achievement Motivation and General Mattering scores in measuring the impact of counseling or psychological interventions as well as in predictive research studies.
Findings from this study indicate that despite being normed on mostly White population, the Spanish version of the Achievement Motivation scale had partial strict invariance when used with Spanish-speaking Latinx college students from Mexico and Spanish-speaking college students from the United States. This finding suggests that representations of achievement motivation were measured similarly across these samples of Latinx students except for items 1 (In school work, I am (was) an achiever), 9 (At My place(s) of employment I am (was) ambitious, and 11 (At My place(s) of employment I am (was) thorough). When we allowed the intercept parameters for those survey items to vary across the groups, model fit became good. There are likely cultural differences that explain why those survey items did not function equally for both groups.
When applying psychometric instruments across cultural contexts, measurement invariance, which is the degree to which a measure assesses the same construct similarly across different groups, may become an issue. Several factors may influence whether scales achieve full invariance or only partial invariance. These factors include cultural norms, language differences, and varying social expectations that can affect how individuals interpret items on a scale (F. F. Chen, 2007). Researchers have found that individualism versus collectivism can play a role in how constructs like achievement motivation and general mattering are experienced and measured across different populations (Triandis, 1995).
Latinx students in the U.S., especially first- and second-generation immigrants, navigate bicultural identities, balancing the values of their heritage culture and the mainstream U.S. culture. For many U.S. Latinx students, achievement motivation may be tied to both personal success and the desire to uplift their families economically (Gonzalez, 2024; Santiago et al., 2014; Suarez-Orozco, 1993). The pressures of being first-generation college students or working part-time while studying might result in differing views on ambition and thoroughness in employment contexts. Specifically, Item 9 (ambition at work) and Item 11 (thoroughness at work) may reflect a more individualized notion of success in the U.S., where economic survival often requires balancing multiple responsibilities such as part-time jobs, familial duties, and academic commitments (Cervantes Arreola et al., 2018; Gonzalez, 2024). In addition, workplace ambition in the U.S. context may be influenced by the broader societal emphasis on individual achievement and upward mobility, as well as the opportunity for economic advancement through education and employment. Thus, U.S. Latinx students may interpret work-related items differently, seeing them as critical to achieving financial independence and success in a highly competitive labor market (Deckard et al., 2022; Gonzalez, 2024).
In Mexico, achievement motivation is often closely tied to family obligations and collectivist cultural values. Academic success is frequently viewed through the lens of family honor and the desire to contribute to the household (Aguirre-Zazueta & Ontiveros-Herrera, 2024). For Mexican students, ambition and thoroughness at work may not hold the same significance, as many students may not work while attending university or may view their academic success as the primary means of fulfilling their family obligations. Item 9 (ambition at work) and Item 11 (thoroughness at work) might be less salient in the Mexican context, where academic achievement is prioritized over work-related ambition during one’s time as a student.
In the United States, achievement motivation is often centered on personal success and self-fulfillment. In contrast, Mexico has a predominantly collectivist culture, where achievement motivation may be more closely linked to family obligations and community expectations (Aguirre-Zazueta & Ontiveros-Herrera, 2024; Suarez-Orozco, 1993). Furthermore, Castro et al. (2024) demonstrated that cross-country differences, such as between Mexico and other nations, could lead to partial measurement invariance due to the distinctive cultural values present in Mexico, particularly related to family dynamics and societal roles. This cultural difference may partially explain why the AMM and GMS achieved only partial invariance in this study, as participants in Mexico may interpret questions differently based on their social obligations and communal focus.
Findings from this study indicate that despite being normed on mostly White population, the Spanish version of the General Mattering scale had metric invariance when used with a Spanish-speaking Mexico population and a Spanish-speaking US population. This means that factor loadings and factor structure (i.e., the same items were measuring the same factors for each group- item 1 (How important do you feel you are to other people?) is loading on factor A for Latinx students from Mexico, and the same item is also loading on factor A for Latinx students from the United States) were equal across the groups of Spanish-speaking college students. This also shows that each survey item loaded onto the specified latent factor in a similar manner and with similar magnitude across the two groups of Spanish-speaking college students. Based on this information, we are confident that researchers and practitioners can use the General Mattering scale for both groups of Spanish-speaking Latinx college students. However, because intercepts and factor means are not equal, group comparisons cannot be made. Since the factor means are not equal, there is a risk that calculating total scores and running statistical analyses based on group means, such as t-tests or ANOVA. The psychometric properties of the scale itself could influence results. While our study did reveal significant mean differences, it is important to acknowledge that these contrasts may be affected by the underlying structure of the scale. However, a detailed examination of these psychometric influences extends beyond the scope of this paper. Our findings should be interpreted with caution, keeping in mind that some observed group differences may partly reflect scale properties rather than substantive differences.
Another reason could be the bicultural identities of Latinx students in the U.S. and collectivist cultural values of Mexican students. Latinx students in the U.S. may grapple with the experience of marginalization and discrimination, which can affect their sense of mattering within both academic and social settings (Safa et al., 2018). Given these pressures, the way U.S. Latinx students interpret their importance to others might be more closely tied to their individual achievements and how they contribute to their family’s financial well-being or social mobility. On the other hand, general mattering is typically more rooted in collectivist cultural values in Mexico (Cienfuegos-Martínez et al., 2016), where individuals perceive their worth in relation to their roles within the family and community (Díaz Gaspar et al., 2024). Hence, mattering is less about personal achievement and more about fulfilling familial obligations and contributing to the well-being of others. Latinx students in Mexico may place a stronger emphasis on how their actions benefit their family or community, rather than how they are perceived by others in a broader societal context. This could explain why differences in factor means were observed between U.S. and Mexican students.
Implications for Practice
There are several implications for practice. First, in a clinical setting, these findings imply that practitioners working with Latinx clients should be aware that the cultural context of their clients may significantly influence how they perceive their importance to others (i.e., general mattering). For instance, Latinx individuals in the U.S. may experience mattering through a more individualistic lens, where personal achievements and contributions to family or societal success are paramount. In contrast, Latinx clients from Mexico might interpret mattering as more closely tied to family obligations and communal roles. School-based mental health professionals and researchers can use the Achievement Motivation Inventory and General Mattering Scale with a few caveats for implementation.
Second, researchers found that positive psychological interventions improved Latinx students’ mental health, hope, resilience, and life satisfaction (Lenz et al., 2020; Vela et al., 2019). Culturally adapted positive psychology interventions, which can focus on gratitude, happiness, and hope (Savage, 2011; Suldo, 2016) might improve achievement motivation and general mattering among Latinx college students. Therefore, college counselors and other mental health professionals who design interventions and programs to improve Latinx college students’ positive mental health can use the Achievement Motivation Scale and General Mattering Scale to identify baseline scores and subsequently identify progress over time. Using the Achievement Motivation scale and General Mattering Scale along with other measures of positive mental health (e.g., life satisfaction) will allow college counselors to document the impact of their evidence-based interventions on Latinx college students’ positive mental health (Lenz et al., 2023).
Limitations and Directions for Future Research
First, we used convenience sampling methods to select college students from a university in northern Mexico as well as a university in the southern region of the United States. Researchers should use random sampling to recruit a diverse range of Latinx students that will allow for additional statistical analyses and powered comparisons (Lenz et al., 2023). Second, all data collected in the current study came from a non-clinical sample of Latinx students. Researchers evaluating validity evidence and measurement invariance of the Spanish versions of the Achievement Motivation Inventory and General Mattering Scale with Latinx students in clinical distress may provide important information for their mental health. There are also opportunities to test measurement invariance between treatment and control groups in counseling intervention studies as well as between pre- and post-tests (C. C. Chen et al., 2020).
There are suggestions for future research. First, researchers can use predictive models to understand how individual factors, interpersonal, and institutional factors influence Latinx students’ achievement motivation or mattering. Second, researchers and practitioners will also generate additional psychometric evidence regarding instruments that measure what they are supposed to measure across different cultural groups including Latinx college students in the United States and Mexico. Finally, not all instruments used in counseling and specifically with Latinx populations have evidence of measurement invariance (C. C. Chen et al., 2020). Future researchers should (1) investigate why counseling researchers do not conduct these types of research studies and (2) how counselor education doctoral training programs could integrate additional training and coursework about psychometric and measurement invariance.
This study highlighted the need of using accurate, trustworthy, and culturally sensitive measures to comprehend the psychological experiences of Latinx college students. Both measures show promise applicability across the various cultural settings of Mexico and the United States, despite several items on the AMM not demonstrating strict invariance. Nevertheless, we recommend being aware of this finding when making cross-group comparisons.
Footnotes
Data Availability
Data available on request due to privacy/ethical restrictions.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
