1. Introduction

Social mobility implies changes in people's socioeconomic status related to education, employment or income (^{Grajales and Monroy-Gómez-Franco, 2018}). Mobility in terms of education is a good predictor of individual well-being and income in modern societies (^{Assaadet al., 2019}; ^{Cabrera, 2016}; ^{Psacharopoulos and Patrinos, 2018}). For instance, in Latin America, mobility is higher in countries with more investment in education (^{Behrmanet al., 2001}).

Children from households where the parents have low education levels face more obstacles to continuing their studies (^{Assaadet al., 2019}; ^{Székely, 2015}), while parents with higher education levels reduce dropout rates and promote university enrollment (^{Alcaraz, 2020}; ^{Cabrera, 2016}). The transmission of educational capital may vary by gender, with it mostly being men who impact their children's educational attainment, although there is no consensus in this respect (^{Corti and Scherer, 2022}; ^{Erát, 2021}; ^{Skopek and Leopold, 2020}).

Specialized literature generally analyzes this transmission by considering the education of the parent with the higher level of education (^{Erikson, 1984}). The correlation of parents’ education is analyzed from the perspective of assortative mating, differentiating between homogamy (similar education) and heterogamy (different education), including hypergamy and hypogamy, depending on who has the higher level of education, with hypergamy being when the father has the higher level of education and hypogamy when the mother has the higher level of education (^{Esteve and McCaa, 2007}; ^{Huo and Chen, 2022}). This approach delves deeper into the impact of the combination of the educational backgrounds of both parents on their children’s social mobility (^{Alcaraz, 2020}; ^{Mare, 1996}; ^{Schneebaumet al., 2015}; ^{Torche, 2010}).

Assortative mating addresses various questions about intergenerational or intragenerational mobility and, from an economic point of view, partner selection is not random and responds to factors such as educational level, ethnicity, religion, occupation and social class, which directly influence the human capital available in the home (^{Huo and Chen, 2022}; ^{Edwards and Roff, 2016}).

Intergenerational educational mobility explains the reproduction of inequality and how structural disadvantages limit human capital development, restrict opportunities and reinforce poverty cycles (^{De Hoyoset al., 2010}). In recent decades, educational mobility in Mexico has been limited and short-lived for lower socioeconomic strata (^{CEEY, 2019}). Progress has been made in primary and secondary education; however, gaps persist at the high school and higher education levels. Only 5% of young people whose parents who have no education go on to attend university, compared to 64% of those whose parents attended university (^{Fernández and Martínez, 2024}).

Mexico’s education spending as a percentage of gross domestic product (GDP) is below international recommendations, despite its importance in reducing inequalities and promoting early development (^{Llanos and Beltrán, 2023}). Education policies do not prioritize early childhood (^{Blanco, 2024}), ignoring its benefits in relation to academic performance, productivity, development of skills and reduction of inequalities (^{BID, 2019}; ^{Heckman, 2008}). Furthermore, spending on early childhood has accounted for only 0.03% of GDP since 2018 (^{Llanos and Beltrán, 2023}).

Thus, educational mobility in Mexico is limited and unequal given the evidence of unequal educational opportunities and the fact that economic inequalities continue to affect access to and continuity of education. Young people and children will comprise tomorrow's workforce so policies that focus on the most vulnerable are essential to achieving equal opportunities (^{Blanco, 2021} and ²⁰²⁴), avoiding a return to the poverty trap and mitigating the intergenerational persistence of inequality.

In Latin America, intergenerational educational mobility increased in lower socioeconomic strata, although it has remained stagnant among higher socioeconomic groups (^{Azevedo and Bouillon, 2011}). In Mexico, research on educational assortative mating and mobility among children is limited. In contrast to the region, there is evidence of educational mobility in homogamy (^{Muñoz, 2024}) and a decrease in hypergamy (^{Esteve and McCaa, 2007}).

In this context, the state of Nuevo León is of particular interest. Despite its high level of schooling, educational gaps based on household socioeconomic level persist (^{Solís, 2022}) and are exacerbated by adverse experiences such as child labor or teenage pregnancy. Research has shown that intergenerational educational mobility in the state continues to depend on social origin (^{Solís, 2022}). Additionally, the Monterrey metropolitan area has historically attracted migration and educational expansion, however, it also exhibits high territorial inequalities and persistent gaps in access to higher education (^{Solíset al., 2007}). This situation of high educational attainment and internal structural barriers makes Nuevo León an interesting region in which to analyze how the educational mating of parents influences the educational opportunities of their children.

This study analyzes the educational attainment of children based on assortative mating, using data from the 2021 Social Mobility Survey in Nuevo León (^{CEEY, 2022}). The empirical analysis shows the effects of combining the education of both parents, rather than analyzing their isolated contribution using predictive models. The main interest lies in observing whether homogamy in basic education remains predominant in the state and to what extent these family structures limit educational mobility. The study also examines whether hypogamy and hypergamy are associated with an increased probability of children achieving higher levels of education. This approach enables us to identify differential patterns of human capital transmission based on the educational structure of the household.

To evaluate these hypotheses, an ordered logit model was used with educational assortative mating as the explanatory variable. The results indicate that children with at least one parent who has completed upper secondary or higher education are more likely to pursue professional studies than children with parents who only completed basic education, especially if the mother has a higher level of education. Additionally, conditions such as child labor or teenage motherhood negatively affect educational attainment.

The article is structured as follows: the second section presents a literature review; the third section presents the data and methodology; the fourth section presents the results; and, finally, the fifth section discusses the conclusions.

2. Literature review

Social mobility shows how a person's socioeconomic variables change in relation to their parents or throughout their life (^{Maralani, 2023}). Intergenerational mobility represents observable changes of an individual in relation to their specific origin, while intragenerational mobility reflects changes throughout an individual’s life (^{Vélezet al., 2012}).

Various studies indicate that differences in educational attainment between parents and children may be associated with factors such as occupation, family size, social status, inherited wealth and, more recently, educational mating (^{Ermischet al., 2006}; ^{Mare, 2000}). Thus, mobility is analyzed as a reflection of the structural conditions that each social group faces in order to access better opportunities (^{Torche, 2010}).

Most studies on intergenerational effects use longitudinal data, mostly with two-generation models, though some incorporate grandparents to detect effects not apparent in the first generational leap (^{Ferrieet al., 2021}; ^{Mare, 1996}; ^{Mare and Song, 2023}). The most commonly used variables in the analysis of social mobility are income, education, occupation and wealth. However, there is also literature that includes perceived or subjective mobility (^{Vélezet al., 2012}).

Education is a key factor in promoting social mobility, but access is not equal for all and depends on household conditions. Families with fewer resources face greater restrictions when it comes to investing in education, which can perpetuate their situation of poverty (^{Barhamet al., 1995}; ^{Berti, 2001}; ^{De Hoyoset al., 2010}). Therefore, poverty traps can be explained by low human capital accumulation from childhood.

While the literature focuses on individual returns to education (^{Card, 1999}), it also examines how parental education influences their children's educational path (^{Mare, 1996}; ^{Torche, 2010}). According to the principle of dominance, studies typically consider the educational level of the parent with the highest level of education (^{Erikson, 1984}), while some examine the influence of the father or mother separately and consider the sex of the child (^{Alcaraz, 2020}; ^{Schneebaumet al., 2015}). Intergenerational educational social mobility occurs when children achieve higher educational levels than their parents (^{Maralani, 2023}).

The increase in women's education has transformed generations, impacting birth rates, the number of children, children's health, the age at which women become mothers, household income, human capital and children's education (^{Becker and Lewis, 1973}; ^{Currie and Moretti, 2003}; ^{Maralani, 2023}; ^{Mare and Maralani, 2006}). The traditional formula of ^{Becker and Tomes (1979)} is challenged by female labor participation, educational expansion and reduced gender differences in education.

Based on the foregoing, the concept of assortative mating in education becomes relevant. This is defined as homogamy when the parents have similar levels of education. hypergamy when the father has a higher level of education and hypogamy if the mother has the higher level of education. While factors such as ethnic origin, cultural identity and location also influence couple formation (^{McPhersonet al., 2001}; ^{Robles, 2024}), studying them in relation to the transmission of human capital to children is also important (^{Kremer, 1997}; ^{Mare, 2000}; ^{Schwartz, 2010}).

Educational mating is not random. Couples tend to have similar characteristics, with education being one of the most stable factors (^{Huo and Chen, 2022}). In various contexts, it has been observed that women with higher levels of education have been observed to pair with equally or more educated men (^{Currie and Moretti, 2003}). These dynamics influence the opportunities passed on to children and can reinforce or weaken patterns of social inequality (^{Davia and Legazpe, 2017}; ^{Fernándezet al., 2005}).

The literature consulted also suggests that couples with homogamy at low levels of education tend to perpetuate limited human capital conditions, whereas those with at least one parent who has completed secondary or higher education can generate a more favorable environment for their children's educational development (^{Corti and Scherer, 2022}; ^{Erát, 2021}; ^{Esteveet al., 2016}). In contexts with higher levels of education among females, hypogamy can also open up opportunities for mobility.

In developed countries, homogamy is more prevalent among those with higher levels of education. In contrast, in Latin America it predominates among those with lower levels of education, which limits social mobility (^{De Hauwet al., 2017}; ^{Esteveet al., 2016}). While educational mobility has improved in families with low levels of education (^{Azevedo and Bouillon, 2011}), public policies do not always reduce inequalities. These inequalities are often maintained or reinforced by labor segmentation or social exclusion (^{Muñoz, 2024}; ^{Neidhöferet al., 2018} and ²⁰²¹). ^{Güellet al.(2015)} find that a high educational correlation between parents reduces social mobility< however, homogamy shows intergenerational mobility in Latin America.

Studies on assortative mating in Mexico focus on the characteristics of couples, without linking them to the educational attainment of their children (Mier and Terán, 2009; ^{Quilodrán and Sosa, 2004} and ²⁰¹¹). Mexico’s probability of positive educational mobility compared to other Latin American countries is 60%, placing it behind countries such as Chile, Uruguay, Costa Rica and Jamaica (^{Muñoz, 2024}). Mexico has greater educational homogamy and less educational hypergamy due to women's increased enrollment in school and their greater persistence in education (^{Esteve and McCaa, 2007}).

According to ^{Moreno (2017)} and ^{Muñoz (2024)}, educational mobility is higher in northern Mexico and Mexico City, while the center and south of Mexico show lower mobility. In the absence of national or regional studies on assortative mating in conjunction with intergenerational educational mobility in children, the literature provides only a partial perspective, by studying either the origin of mating or the educational attainment of children based on the principle of dominance.

In the state of Nuevo León, ^{Solíset al.(2007)} analyze variations in assortative mating based on couples’ education and migration, from a mobility perspective, without considering its effects on their children’s education. In the mid-20th century, educational homogamy in Monterrey shifted from occurring mostly at the lowest educational levels to occurring largely at the secondary level. Migration and increased job opportunities led to mating between individuals with different levels of education.

Meanwhile, ^{Solís (2022)} shows that children in Nuevo León achieve higher levels of education than their parents after taking into account socioeconomic status. However, this intergenerational increase contrasts with the fact that approximately two-thirds of children are unable to access higher education, highlighting limitations in educational mobility in the region.

This analysis provides a comprehensive approach by integrating the ideas of ^{Solíset al.(2007)} and ^{Solís (2022)}. The authors particularly emphasize the fact that the increase in homogamy is not solely due to educational causes but may also be linked to spatial isolation and social segregation resulting from a lack of public services and housing (^{Pozas, 1990}; ^{Solíset al., 2007}; ^{Solís, 2022}).

Intergenerational educational mobility is a phenomenon that helps us understand the mechanisms that reproduce inequality in contexts such as Mexico. Since children's educational attainment is influenced by their parents' level of education, structural disadvantages persist, limiting human capital development and restricting equal opportunity, while strengthening cycles of poverty (^{De Hoyoset al., 2010}). Studying these patterns is important for designing policies that can break this intergenerational transmission of inequality (^{Barhamet al., 1995}; ^{Berti, 2001}).

3. Data and methodology

This study uses the database from the 2021 ESRU Social Mobility Survey in Nuevo León, referred to in the document as 2021 ESRU-EMOVI Nuevo León (^{CEEY, 2022}). The survey was conducted in 2021 and is representative of men and women between the ages of 25 and 64 in the region, including the metropolitan area, suburbs and other areas. The survey collects data on the education of the respondents and their parents, which allows for the analysis of intergenerational educational mobility. The survey also contains information about the sociodemographic characteristics of the current household and the household in which the respondents lived at age 14.

The original 2021 ESRU-EMOVI Nuevo León sample included 3,767 individuals. For this study, a subsample of 2,668 individuals was used. This subsample included data on the educational level of both parents, which was necessary to create the educational mating categories. While the reduction itself does not imply less representativeness; however, as it is a selection based on available information, the subsample may differ from the original sample in certain aspects. It is assumed that the subsample retains the demographic and educational characteristics of the target population, at least in general terms. Although non-response bias cannot be formally assessed, the results should be interpreted with caution.

4. Results

Table 4 shows the marginal effects of the explanatory variables based on the category of assortative mating of parents. Categories with similar characteristics include primary homogamy, secondary education homogamy and high school/professional education homogamy. Categories with different characteristics (heterogamy) include a group of parents with secondary hypogamy, high school/professional hypogamy, secondary hypergamy and high school/professional hypergamy. Primary hypogamy is used as a reference in the model estimates.

When analyzing the probability of an individual finishing at primary school level or having no education, it is generally found that for any combination of parental education, the probability of finishing at this level is lower than that of the reference group. Specifically, children whose parents both have a secondary education are 7.3 percentage points (pp) less likely to finish at primary school level or have no education than children whose parents finished primary school or have no education. This probability increases if both parents have a high school or professional education, in which case the probability is 13.2 pp lower.

Regarding children with parents in educational heterogamy, the magnitude of the difference with respect to the base group is greater when the mother has a higher level of education than the father. For instance, if the mother has a secondary education and the father has a primary education or no education, the probability is 13.7 pp lower than that of respondents whose parents have the lowest level of education. Meanwhile, when the father has a secondary education and the mother has a lower level of education, the probability is 5.1 percentage points lower. The same trend applies to the secondary education level.

For those with high school as their highest level of education, all combinations of parental education have a marginal positive effect compared to the base group (primary homogamy) with increases ranging from 6.7 to 17.9 percentage points depending on the type of mating (see Table 4).

At the professional level, when the mother has a higher level of education than the father (hypomgyamy), the probability of reaching this level increases by 15 pp if she attains a secondary education and by 12 pp if she attains a high school or professional education. In the case of hypergamy (the father has a higher education), the effects are 5.6 pp and 10.5 pp, respectively. When both parents have the same level of education, the probability increases by 8.0 pp with a secondary education and by 14.5 pp with a high school/professional homogamy. These results suggest that the higher the parents' educational level, especially when they both have the same level, the greater the chances their children will reach the professional level.

Regarding control variables, no significant effects were found with respect to gender or number of siblings. However, certain adverse conditions show negative effects. Having their first child as a teenager reduces the probability of pursuing professional studies by 7.1 percentage points, while having participated in child labor reduces it by 4.3 percentage points.

Controls were included for age group, mother employed as a teenager and household resources, as well as fixed effects by geographical area, to control for contextual factors that could influence educational attainment.

5. Discussion and conclusions

This research contributes to the literature on social mobility by analyzing intergenerational educational mobility. Although this topic has been widely studied, few studies have considered the interaction between the educational levels of both parents. Typically, the parent with the higher level of education is used as a reference or each parent is analyzed separately, under the assumption of independent effects. However, the combination of both parents' educational capital significantly influences their children's outcomes (^{Edwards and Roff, 2016}). Thus, this study employs the concepts of homogamy, hypergamy and hypogamy to categorize educational mating.

Understanding the educational level at which mating occurs is essential for grasping intergenerational mobility since the impact varies depending on which parent has the higher level of education. The results reinforce the importance of considering the combined educational level and highlight significant differences between types of mating. A pattern consistent with the literature on the transmission of educational capital (^{Maralani, 2023}) was found, which identifies a positive effect of assortative mating on educational mobility in Nuevo León, as also suggested by ^{Kremer (1997)}.

Children whose parents have a higher level of education have an advantage over their peers whose parents are less educated (^{Assaadet al., 2019}). Nevertheless, when analyzing the results in terms of hypogamy and hypergamy, these differ from the findings of ^{Corti and Scherer (2022)}. Here, we found that, at any level, hypogamy has a greater influence than hypergamy. In contrast, ^{Corti and Scherer (2022)} demonstrate that there is a reward associated with hypergamy.

Based on the descriptive analysis, homogamy behavior in Nuevo León focuses on low levels of education, which is consistent with the findings of ^{Esteveet al.(2016)} for undeveloped countries. This result supports the first reasoned hypothesis. Meanwhile, hypergamy is more prevalent at higher levels of education, with a predominance of couples in which the man has a higher level of education than the woman.

Regardless of the overall educational level of the parents, it is important to note that the probability of the child remaining at the primary school level or below is negative; i.e., it is increasingly unlikely that new generations will have low levels of education. Regarding the mother's participation, in agreement with ^{Maralani (2023)} and ^{Mare and Maralani (2006)}, hypogamy increases the likelihood of higher educational attainment in children more than hypergamy does. This result supports the second hypothesis regarding hypogamy, as it indicates greater intergenerational educational mobility. However, data on hypergamy in Nuevo León does not corroborate this relationship.

Probability results in terms of educational attainment show short-range upward mobility in secondary education levels, as defined by ^{Solís (2022)} and ^{CEEY (2019)}, i.e., children are more likely to achieve the same educational level as their parents than a higher one. Therefore, to reduce the intergenerational persistence of inequalities in higher education, ^{Solis (2022)} suggests reorganizing the education budget, improving education management and strengthening the teaching profession.

Due to the close link between education and various government policies, budgetary suggestions and changes to the education system should be considered. For example, a 2 pp increase in the GDP in the national education budget, which is distributed more equally between educational levels, greater investment in and implementation of comprehensive early childhood care systems, early warning projects to encourage students to stay in school and projects to address shortcomings in school infrastructure, among other improvements specific to the Mexican education system (^{CEEY, 2024}; ^{Solís, 2022}).

However, if the results show a positive incentive for the inclusion of women in their children's educational attainment, the focus should not shift to a gender-focused policy, but rather to how to strengthen the future of the state's human capital to alleviate poverty traps (^{Barhamet al., 1995}; ^{Berti, 2001}; ^{De Hoyoset al., 2010}). In this regard, the importance of focusing on early childhood is paramount, as this is when the first learning inequalities arise (^{Heckman, 2008}). Furthermore, it is essential to support the liberation of women from caregiving tasks by rethinking the daycare program based on the direct transfer program (^{Solís, 2022}).

Special attention should also be given to individuals whose education is at risk due to circumstances such as having a child while still a teenager or being involved in child labor. Public policies should focus on helping these vulnerable groups to increase their chances of educational success, enabling them to achieve a better quality of life and break the pattern of inequality so it is not passed on from generation to generation.

It should be noted that the findings should be interpreted with caution since the analyzed population is from the state of Nuevo León, Mexico. Nuevo León is one of the most developed states and has a higher average level of education than other states in the country. However, some individuals face difficulties achieving high levels of education, which suggests that this type of survey should be conducted in regions with below average schooling in order to identify relevant factors and eliminate regional differences throughout the country. Additionally, it is important to verify which policies have been implemented and worked, as well as follow up on unsuccessful policies to guide them toward a satisfactory final result.

Finally, one of the limitations of the article was that the results show a direct relationship between the education of parents and that of their children, but the effect is not necessarily causal. According to ^{Björklund and Salvanes (2011)}, there may be other confusing factors, such as genetic or prenatal effects, which may influence the cross-sectional relationship between parents and children.