David Cambron Jiménez The relationship between education and gender on employability with welfare programs in Mexico: A llogit analysis repositorio.lasalle.mx/handle/lasalle/ Vol. 7 No. 2, 2025, jul- dic, pp. 114-131 Revista Internacional de Salarios DignosISSNDOIISSNHandle repositorio.lasalle.mx/handle/lasalle/ENLACES repositorio.lasalle.mx/handle/lasalle/ORCID:Licencia de Atribución Creative Commons 4.0CC BY 4.0Este documento se puede descargar y compartir con otras personas, siempre que se reconozca su autoría.Repositorio Institucional REIMSRISAD REVISTA INTERNACIONAL DE SALARIOS DIGNOSThe Relationship Between Education and Gender on Employability with Welfare Programs in Mexico: A Logit Analysis David Cambron Jimenez1Received: 01/09/2025 Accepted: 19/11/2025 AbstractThis paper analyzes the relationship that exists between level of education, gender, and employability within the framework of the social welfare program /6venes Construyendo el Futuro (}CF), part of Mexico's Becas de/ Bienestar. A logit model was applied using date from the Centro de lntegraci6n Juvenil (Cl}) to estimate the probability of obtaining employment before the maximum twelve-month duration of the scholarship. The results highlight notable differences by sex and educational level: women, regardless of their education attainment, show negative slopes in their probability of employment; men with higher education also display a negative effect, though smaller in magnitude than the women, nevertheless, men with high school education obtain a positive effect. Overall, the study shows that gender and education are statistically significant variables, while the scholarship itself is marginally significant in the model. The global model is validated as significant against the null model, although the pseudo 𝑅2 shows low explanatory power in the model. The most relevant finding suggests that JCF scholarships fail to counteract structural inequalities in gender and education within labor market insertion, with women holding higher education degrees being the most disadvantaged by the scholarship, showing the lowest probability of employment in the model. Key Words Scholarships, Unemployement, Gender Introduction In Mexico, young people who are neither studying nor working pose a structural challenge to the country's economic and social development. The /6venes Construyendo el Futuro (JCF) scholarships are a Mexican public policy designed to mitigate this phenomenon, offering a monthly scholarship of 8,470.17 Mexican pesos in 2025 and training in workplaces to young people between the ages of 18 and 29 (Gobierno de Mexico, 2025). The program's overall objectives are: "to integrate young people into on-the-job training activities and thus equip them with the tools for a better life" and "to steer young people away from unemployment and antisocial behavior." (Secretaria del Trabajo y Prevision Social, 2023), which translates into a program that seeks to train young people for the labor market and thereby reduce unemployment. Mexican government data shows that 6 out of 10 young people who participate in the program can join a productive activity or find employment (Gobierno de Mexico, 2024).1 Graduate of the Bachelor's Degree in Economic and Financial Engineering, La Salle University, Mexico CityCambron Jiménez David”Cambron Jiménez Davideconomy that benefit only a small number of sectors and increase the social inequality gap. Yanez-Contreras and Cano-Hernandez (2011) analyzed that age, socioeconomic status, educational level, marital status, rural or urban area, belonging to any minority group, crime, low-quality employment, innovation, technological development, government reforms, and labor regulations may be factors that influence unemployment in Latin America. Another important phenomenon, but one not addressed in this research, is that unemployment as a cause of migration and as a consequence of poverty causes a reduction in remittances, impacting Mexico due to the insufficient supply of jobs for immigrants in the United States (Figueroa et al., 2012). Finally, Garda Vina in 2020 states: "There is clearly a clear connection between high youth unemployment rates, lack of education, and informal employment. If these people, who normally do not have completed compulsory education, manage to enter the labor market, the possibility that they will only find jobs in the informal economy and/or that the jobs they hold throughout their professional careers will be in this sector is extremely high." demonstrating that entering the labor market in Mexico is becoming increasingly complex. Discrepancies in employability by gender Pacheco and Parker (1996) mention that women continue to perform different jobs than men, as they continue to perform activities classified as "gender-specific." Although this trend has been reversed in the last decade, with women joining jobs that were previously considered "masculinized," such as entrepreneurship and employability, it is still necessary to recognize this group in order to achieve women's contribution to the productive process in Latin America (Mazuera Arias et al, 2017). This has led to the informal sector becoming an important destination for female employment (Lorfa et al., 2012, cited by Hernandez, 2020) and has resulted in greater job insecurity for women, such as part-time jobs, excessive working hours, low wages, and a lack of social benefits (Brown and Dominguez cited by Murayama, 2011). This phenomenon has resulted in 58% of the 10.8 million women in formal employment in Mexico not receiving a decent wage. Of this population, 37% earn poverty wages and 21% earn wages considered to be subsistence wages (Aquino, 2025). In addition, the report "La justicia econ6mica pendiente" (Pending Economic Justice) in 2025, led by Paulina Gutierrez, highlights that almost 2 out of 10 salaried women work excessive hours of more than 48 hours, and 47% of them do not have a stable contract that gives them job security (cited by Mendez, 2025). This document also argues that being indigenous increases the exclusion of women, since by 2025, more than 80% of indigenous women aged 15 and older had a secondary school education or less. The author shows that, added to the barriers of discrimination, this greatly affects their chances of finding decent work. Data Box 1. Data and Variables SymbolVariable Source𝑌𝑡Dichotomous variable of staying in Author's own elaboration based on the scholarship or leaving data from Centro de lntegraci6n"The Relationship Between Education and Gender on Employability with Welfare Programs in Mexico: A Logit Analysis”𝑋1𝑡Scholarship received monthly in the year it is awarded./uvenil (2024) Centro de lntegraci6n Juvenil (2024)Given the current context surrounding these /6venes Construyendo el Futuro welfare scholarships, this study aims to analyze whether these scholarships effectively promote integration into formal employment in the country or whether they contribute to unemployment, considering variables such as the amount received by the scholarship, educational level, and gender, adjusted and applied to the Mexican context in 2025. It should be noted that, although the model recognizes relevant variables, others such as region, level of informal employment, or social inequality index could not be included due to database limitations. To this end, the following conducted: 1. Adjustment of scholarships based on the start and end year, knowing the years when they entered (Gobierno de Mexico, 2018-2024). Considering that the years 2018 to 2021 are daily* and adjusted to 30 days, the others are monthly (see Table 1): Table 1. Index of monthly scholarship amounts per year.YearAmount in Mexican pesos2024848020237572202263102021$172.87 * 30 dias2020$141.70 * 30 dias2019$123.22 * 30 dias2018$88.36 * 30 diasSource: Prepared internally using data from the Mexican government. 2. The 2,534 available CIJ data were used to classify them as Female or Male, 3. The highest level of education was classified in the CIJ database (using only upper secondary and higher education), 4. Create a threshold (dichotomous variable) with the time they actually spent on that scholarship, creating the dichotomous variable and work assumption, where: based on whether or not they managed to complete the program before 12 months, inferring that if they manage to do so before 12 months, it means that they did enter the labor market (assigning 1 before 12 months, 0 in all other cases). It should be noted that the construction of the threshold does not consider other reasons for leaving, such as dropout, dissatisfaction, or change in personal circumstances, due to the limitations of the database, 5. Construct education and gender variables by multiplying them by the dichotomous variable (see Appendix 1 for the database) and finally,Cambron Jiménez David𝑌𝑖 = 𝑎0 + 𝑎1𝑋1𝑖 + 𝑎2𝑋1𝑖𝐷1 + 𝑎3𝑋1𝑖𝐷2 + ⋯ + 𝑎𝑘𝑋𝑘𝑖𝐷𝑘 + 𝑈𝑖 𝑌𝑖 𝑋1, 𝑋2, … , 𝑋𝑘 𝐷1, 𝐷2, … , 𝐷𝑘 𝑘 𝑎𝑗𝑈𝑖𝑋1, 𝑋2, … , 𝑋𝑘 𝐷1, 𝐷2, … , 𝐷𝑘 𝑌𝑖 𝐻0: 𝑎𝑗 = 0 𝑣𝑠 𝐻𝑎: 𝑎𝑗 ≠ 0, 𝑝 𝑝𝑣𝑎𝑙𝑢𝑒 𝑝𝑣𝑎𝑙𝑢𝑒 < 𝑌𝑖𝐿𝑅 = −2[𝑙𝑛𝐿0 − 𝑙𝑛𝐿1]~𝜒2𝑝 < 0.05𝐷1 𝐷2𝐷1= { 1 𝑖𝑓 𝐹𝑒𝑚𝑎𝑙𝑒 (𝑀) 0 𝑖𝑓 𝑀𝑎𝑙𝑒 (𝐻)”𝐷2= { 1 𝑖𝑓 𝐻𝑖𝑔ℎ𝑒𝑟 𝐸𝑑𝑢𝑐𝑎𝑡𝑖𝑜𝑛 0 𝑖𝑓 𝑈𝑝𝑝𝑒𝑟 𝑆𝑒𝑐𝑜𝑛𝑑𝑎𝑟𝑦 𝐸𝑑𝑢𝑐𝑎𝑡𝑖𝑜𝑛 𝑋1𝑖 𝑌𝑖𝑌̂𝑖 = 𝐸(𝑌𝑖|𝑋1, 𝑋2, … … , 𝑋𝑘𝐷1𝐷2= 𝑎̂0 + (𝑎̂1 + 𝑎̂2 + 𝑎̂3 )𝑋1𝑖 + ⋯ + 𝑎̂𝑘𝑋𝑘𝑖𝐷𝑘𝑌̂𝑖 = 𝐸(𝑌𝑖|𝑋1, 𝑋2, … … , 𝑋𝑘𝐷1𝐷2= 𝑎̂0 + (𝑎̂1 + 𝑎̂3 )𝑋1𝑖 + ⋯ + 𝑎̂𝑘𝑋𝑘𝑖𝐷𝑘𝑌̂𝑖 = 𝐸(𝑌𝑖|𝑋1, 𝑋2, … … , 𝑋𝑘𝐷1𝐷2= 𝑎̂0 + (𝑎̂1 + 𝑎̂2 )𝑋1𝑖 + ⋯ + 𝑎̂𝑘𝑋𝑘𝑖𝐷𝑘𝑌̂𝑖 = 𝐸(𝑌𝑖|𝑋1, 𝑋2, … … , 𝑋𝑘𝐷1𝐷2= 𝑎̂0 + 𝑎̂1𝑋1𝑖 + ⋯ + 𝑎̂𝑘𝑋𝑘𝑖𝐷𝑘𝑎̂𝑘+1𝑌𝑖𝑌𝑖𝑌𝑖𝑋1, 𝑋2, … , 𝑋𝑘1 𝑖𝑓 𝑌𝑖 𝑚𝑒𝑒𝑡𝑠 𝑡ℎ𝑒 𝑐𝑜𝑛𝑑𝑖𝑡𝑖𝑜𝑛; 𝑝𝑟𝑜𝑏 = 𝑝 = { 0 𝑖𝑓 𝑌 𝑑𝑜𝑛′𝑡 𝑚𝑒𝑒𝑡 𝑡ℎ𝑒 𝑐𝑜𝑛𝑑𝑖𝑡𝑖𝑜𝑛; 𝑝𝑟𝑜𝑏 = 1 − 𝑝 𝑖𝑌1𝑋1𝑖 𝑌𝑖 = 𝑎0 + 𝑎1𝑋1𝑖𝑢𝑖𝑌̂ = 𝐸(𝑌|𝑋1𝑖)𝑌𝐸(𝑌) = ∑1𝑖𝑃(𝑌 = 𝑖) = 0 ∗ (1 − 𝑝) + 1 ∗ 𝑝 = 𝑝Cambron Jiménez David𝑌 𝑃𝑖 = 𝐸(𝑌 = 1|𝑋 , 𝑋 , … , 𝑋 ) = 1𝑌 𝑧𝑖 = 𝑒1 2 𝑘1+𝑒−𝑧𝑖1+𝑒𝑧𝑖𝑧𝑖 = 𝑎0 + 𝑎1𝑋1𝑖 + 𝑎2𝑋2𝑖 + ⋯ + 𝑎𝑘𝑋𝑘𝑖 lim 𝑃𝑖 = 0 lim 𝑃𝑖 = 1𝑧𝑖→∞    𝑃𝑡     = 𝑒𝑧𝑡 1−𝑃𝑡 𝑧𝑖→∞ 𝑋1, 𝑋2, … , 𝑋𝑘𝐿 = ln (    𝑃𝑡    ) 𝑧 = 𝑎 + 𝑎 𝑋 + 𝑎 𝑋 + ⋯ + 𝑎 𝑋𝑋1, … , 𝑋𝑘𝑡 1−𝑃𝑡 𝑡0 1 1𝑡2 2𝑡𝑘 𝑘𝑡𝐷1, 𝐷2, … , 𝐷𝑘𝐷1 𝐷2= { 1 𝑖𝑓 𝐹𝑒𝑚𝑎𝑙𝑒 (𝑀) 0 𝑖𝑓 𝑀𝑎𝑙𝑒 (𝐻) = { 1 𝑖𝑓 𝐻𝑖𝑔ℎ𝑒𝑟 𝐸𝑑𝑢𝑐𝑎𝑡𝑖𝑜𝑛 0 𝑖𝑓 𝑈𝑝𝑝𝑒𝑟 𝑆𝑒𝑐𝑜𝑛𝑑𝑎𝑟𝑦 𝐸𝑑𝑢𝑐𝑎𝑡𝑖𝑜𝑛𝑌𝑡 = 𝑎0 + 𝑎1𝑋1𝑡 + 𝑎2𝑋1𝑡𝐷1 + 𝑎3𝑋1𝑡𝐷2 + 𝑈𝑡 𝑌𝑡 = 𝑎0 + 𝑎1𝑋1𝑡 + 𝑎3𝑋1𝑡𝐷2 + 𝑈𝑡 𝑌𝑡 = 𝑎0 + 𝑎1𝑋1𝑡 + 𝑎2𝑋1𝑡𝐷1 + 𝑈𝑡 𝑌𝑡 = 𝑎0 + 𝑎1𝑋1𝑡 + 𝑈𝑡"The Relationship Between Education and Gender on Employability with Welfare Programs in Mexico: A Logit Analysis”finally, model (10) presents a man with upper secondary education in the sample of Centro de lntegraci6n Juvenil. Running the logit model gives us the following estimate:𝑌̂𝑡 = 0.4466 + 0.0722𝑋𝑡1 − 0.0366𝑋1𝑡𝐷1−0.0908𝑋1𝑡𝐷2 + 𝑈𝑡-11𝑃𝑣𝑎𝑙𝑢𝑒(0.072) (0.024) (0.000)The expression shows the gender variable (𝐷1 ) and the education variable (𝐷2 ) are significative (ver 𝑃𝑣𝑎𝑙𝑢𝑒 0.024 and 0.000), while (𝑋1 ) even that the 𝑃𝑣𝑎𝑙𝑢𝑒 is greater than 0.05, is marginally explanatory, as will be discussed further below. In order to avoid multicollinearity issues among the model regressors, Pearson's correlation matrix was estimated between the variables scholarship, education, and gender. The resulting matrix is presented below:Table 2. Pearson's correlation matrixScholarshipEducationGenderScholarship1.0000.2580.281Education0.2581.0000.096Gender0.2810.0961.000Source: Own compilation in R based on sample data. As can be seen, all cross-correlations are below 0.30. The correlations between the scholarship and its interactions with education (0.258) and gender (0.281) are low and consistent with the way the variables were constructed, while the correlation between education and gender (0.096) is practically nil. These results indicate that there is no strong linear relationship between the regressors in the model, so no multicollinearity problem is identified that could skew the coefficient estimates. Therefore, these variables do provide information to explain the probability of being employed. Returning to equation (11) gives rise to the four cases described above: Case 7: Woman with higher education It is obtained by doing 𝐷1 = 1 y 𝐷2 = 1𝑌̂𝑡 = 0.4466 − 0.0552𝑋1𝑡 + 𝑈𝑡 Case 2: Man with higher education It is obtained by doing 𝐷1 = 0 y 𝐷2 = 1 𝑌̂𝑡 = 0.4466 − 0.0186𝑋1𝑡 + 𝑈𝑡 Case 3: Woman with upper secondary education It is obtained by doing 𝐷1 = 1 y 𝐷2 = 0(12) (13)Cambron Jiménez David𝑌̂𝑡 = 0.4466 − 0.0356𝑋1𝑡 + 𝑈𝑡 𝐷1 = 0 𝐷2 = 0 𝑌̂𝑡 = 0.4466 + 0.0722𝑋𝑡1 + 𝑈𝑡𝑃𝑣𝑎𝑙𝑢𝑒𝑠𝑅2 𝑐ℎ𝑖2 𝑝 > 𝜒2𝑝̂𝑝̂𝑝̂𝑝̂"The Relationship Between Education and Gender on Employability with Welfare Programs in Mexico: A Logit Analysis”Table 3. Probability matrix based on the logit model (𝑝̂) in percentage The columns of the matrix show gender. The rows of the matrix show level of education. The subscripts indicate which case they belong to.(𝑝̂)WomenMenHi�her Education53.62% 157.28% 2Upper Secondary Education55.84% 359.50% 4Source: Own elaboration based on the logit model Running significance tests, the overall fit of the model is indeed significant compared to the null model. As defined by UCLA and OARC Stats in 2006, the likelihood ratio test compares the "null model" (with a single intercept) with the "full model" (with all explanatory variables), evaluating whether the set improves the fit, since the statistic is based on the difference between the likelihood logarithms of both models and the chi-square distribution (𝜒2 ). This means that at least one of the explanatory variables provides information for predicting the dichotomous variable. The pseudo 𝑅2 (0.0092) corresponds to the McFadden index, which should not be interpreted as a 𝑅2 linear, but rather as a relative measure of improvement of the model with respect to the null model. Therefore, even though it is low (common due to the structural variables not considered and the social contexts with high unexplained variability), the global likelihood test validates that the model does have a predictive capacity statistically different from zero. This is because it showed a value of LR=31.66 with 3 degrees of freedom, which is statistically significantdue to (𝑝 < 0.001). Based in the next hypotheses:𝐻0: All coefficients of the explanatory variables are equal to zero, meaning that the complete model does not improve on the null model, 𝐻1: At least one of the coefficients is different from zero, meaning that the complete model does improve on the null model, Thus, rejecting the null hypothesis that all explanatory variable coefficients are equal to zero, accepting the alternative that at least one of the variables, whether education, gender, or scholarship, provides relevant information to explain the probability of employment. In addition, two predictors, both the education variable and the gender variable, are significant, being below 𝑝 < 0.05 and, although the scholarship is not significant at 5%, it is marginally significant, implying that not all of the model is noise, since at least two predictors show consistent effects due to the Wald test, which has the following hypothesis: 𝐻0: The coefficients are equal to zero and the variables are not important for the model, 𝐻1: The coefficients are different from zero and the variables are important for the model. Thus, giving results where gender and education reject 𝐻0 as long as the scholarship does not reject 𝐻0 (See Box 2). That is why this model, rather than providing perfect predictions, serves to detect close statistical relationships between education, gender, and scholarships. The low explanatory power is reflected in the fact that more relevant variables are missing in the effectCambron Jiménez Davidof being employed through a /6venes Construyendo el Futuro scholarship, as described in the literature, but this does not invalidate the findings in the model. When evaluating the model's discriminatory capacity, the ROC curve was estimated using the predicted probabilities. The area under the curve was AUC: 0.5598, indicating low predictive ability and barely above chance (AUC=0.50). The ROC curve was very close to the diagonal (see image 1), showing that the model has limited ability to distinguish between young people who found employment and those who did not. Image 1. ROC curve of the logit model and AUC (Title in Spanish) Curva ROC del Modelo Logit =ci 1.2 1 .0 0.8 0.6 0.4 0.2 0.0 -0.2 Specificity Area Under the Curve (AUC)= 0.5598 Source: Prepared internally based on the logit model. Graph created in R. Discussion Two variables included in the analysis, education and gender, were found to be significantly accurate. Dichotomous variables were incorporated into this analysis in order to distinguish and analyze two factors related to how a specific welfare grant, namely the /6venes Construyendo el Futuro (Youth Building the Future) grant, can be used. The results showed that, of the four cases, women with higher education benefited the least. This can be contrasted with Flores' 2024 research cited by Ramirez Carrera in the same year, which showed that there are precarious working conditions for women with high academic qualifications, many of them with postgraduate degrees, aged between 25 and 40. This shows that, although there has been an increase in women's access to higher education and the labor market, this does not guarantee decent conditions. On the other hand, we see that the slope in both cases for women is negative, which shows that, in terms of employability in Mexico, being a woman can make it much more difficult than being a man, as evidenced by the model. According to INEGI, as of July 2020, the employment rate in Mexico was 34.7% for women, compared to 71.8% for men. Returning to the approach of Levy and Szekely (2016) and Garcia Vina (2020) described in the literature review, there is a reality in Mexico regarding a variable that the model does not study”Cambron Jiménez Davidpower of the model and contribute to the low pseudo value. In future extensions of the work, the incorporation of these socioeconomic controls would allow for the estimation of a more complete model with greater predictive power. The low discriminatory capacity observed in the ROC curve (AUC=0.5598) is consistent withpseudo 𝑅2 reduced logit model. Although the model is statistically significant and showsconsistent relationships between the variables included (scholarship and mainly education and gender), its predictive power is low. This suggests that the phenomenon of employability depends on additional factors not found in the database, thus limiting the model's ability to adequately separate cases of job success and failure. Conclusions A logit analysis was applied with the main objective of determining the probability of being employed through a /6venes Construyendo el Futuro scholarship and leaving the program before 12 months, which is the maximum period for using this scholarship. These scholarships are characterized as a social welfare program in Mexico for young Mexican men and women who are neither studying nor working, in exchange for which they receive a scholarship and are trained for a job so that they can enter the formal labor market. The variables used for this model were the amount of the scholarship, their level of education (higher or upper secondary), and their gender. What stood out about the model was that in all four cases among men and women, women with both higher and upper secondary education had a negative slope, while men with higher education also had a negative slope, but men with upper secondary education had a positive slope. This shows that men with this scholarship are more likely to find employment in Mexico than women. The results of the analysis also show that it is not particularly beneficial for women with higher education to enter this scholarship program, as they are less likely to leave the program to take up employment. Conversely, men with upper secondary education are more likely to find employment and leave the program (below the threshold of the dichotomous variable defined in the article). The factors found complement the literature (see literature review), which has identified not only gender and education as causes, but also the degree of labor informality, region, age, and many other variables in Mexico that are among the main causes of difficult labor market integration within that country. On the other hand, the analysis has limitations, considering that the model presents a low pseudo 𝑅2, The overall plausibility test shows that the model is better than the null model and, finally, that two of the three predictors are significant and consistent for the dichotomous probability through the Wald test (gender and education). The sample is limited, as only 2,534 data points available from the Mexican government's Youth Integration Center are used, and this model may not be representative of the entire population of the Youth Building the Future program. Furthermore, the model's low explanatory power, evidenced by the AUC of 0.5598, suggests that the program operates in a context where unaccounted structural factors limit the individual prediction of labor market insertion. Finally, the analysis seeks to be an approximation that explains how narrow and explanatory these variables can be, but not sufficient to provide a conclusive conclusion about the reality of"The Relationship Between Education and Gender on Employability with Welfare Programs in Mexico: A Logit Analysis”the program at the national level. Therefore, the ultimate goal of this model is more focused on detecting significant relationships that could lead to further investigation of this phenomenon, such as the labor sector, informality, region, or age, which are and could be key variables for a more robust model. Greetings Special thanks to the entire community at Universidad La Salle Mexico and Johannes Kepler Universitat, Linz, Austria. For believing in me by opening the doors of your institutions, training me, and developing me as a great professional. References Agresti, A. (2021). Categorical data analisis. (4th ed.). Wiley. Aquino, E. (2025, March 7). Desigua/dad econ6mica de las mujeres: enfrentan desempleo par cuidar, informalidad /abora/ y empleos con salarios de pobreza. Animal Politico. https:/ /animaIpoIitico.com /sociedad/ desigua ldad-econom ica-m ujeres-desem pleo- informaIidad-sa lari os Bruin, J. (2006). Newtest: command to compute new test. UCLA: Statistical Consulting Group. https:/ /stats.oa rc.ucla.edu/stata/ado/analysis/ Consejo Nacional de Evaluaci6n de la Polftica de Desarrollo Social. (2020). Ficha de monitoreo de programas y acciones federates 2020: J6venes Construyendo el Futuro. https://www.coneval.org.mx/EvaIuacion/IEPSM/Docu ments/Fi chas/2020/Ficha_Jovenes_ Construyendo _el_Futuro _2020.pdf Evans, R. (2025). Detecting and Dealing with Multicollinearity: An Introduction to Econometrics. econometricstutor.co.uk. https://www.econometricstutor.co.uk/multico11i nearity-detecting-a nd-deaIi ng-with-m ultico11i nearity Figueroa Hernandez, E., Ramfrez Abarca, 0., Gonzalez Elf as, J.M., Perez Soto, F., & Espinosa Torres, L. E. (2012). Analisis de/ desempleo, la migraci6n y la pobreza en Mexico. Revista Mexicana de Agronegocios, 30(enero-junio), 835-847. Sociedad Mexicana de Administraci6n Agropecuaria A.C. http://www.redalyc.org/articulo.oa?id=14123097006 Garda, A. (2025, May 13). Desempleo en Mexico repunta con 23 mil/ones de personas en exclusion /abora/. Grupo Milenio. https://www.milenio.com/negocios/desempleo-mexico-repunta-23-mii lones-personas Garda, A. K. (2018, december 26). Educaci6n en Mexico: insuficiente, desigualy la calidad es diffcil de medir. El Economista. https://www.eleconomista.com.mx/politica/Educacion-en-Mexico-i nsuficiente-desiguaI-y-la-ca Iidad-es-difi ciI-de-m edir-20181225-0028.htm I Garda Vina, J. (2020). La situaci6n de/ empleo de las j6venes y una propuesta de polfticas publicas para afrontar el desempleo juvenil. Revista Latinoamericana de Derecho Social, 30, 65-94. https:/ / doi .org/10.22201/i ij.24487899e.2020.30.14072 Greene, W.H. (2002) Econometric Analysis. 5th Edition, Prentice Hall, Upper Saddle River, 802. Hernandez Perez, J. (2020). Desempleo en Mexico par caracterfsticas sociodemograficas, 2005- 2018. Economfa UNAM, 17(50), 166-181.Cambron Jiménez DavidHosmer, D. W., Lemeshow, S., & Sturdivant, R. X. (2013). Applied logistic regression (3rd ed.). Wiley. INEGI. (2023). Encuesta Nacional de Ocupaci6n y Empleo (ENOE), cuarto trimestre 2023. https://www.inegi.org.mx/progra mas/ enoe/15ymas/ lnstituto Mexicano para la Competitividad. (2022). /6venes Construyendo el Futuro: i_aprenden las j6venes? https:/ /imco.arg.mx/jovenes-con struyendo-eI-futuro-aprenden-1os-jovenes/ lnstituto Nacional de Estadfstica y Geograffa (n.d.). Caracterfsticas educativas de la poblaci6n. https://www.inegi.org.mx/temas/ educacion/ Juarez, V. H. (2021, february). Fraude y corrupci6n, desvfan dinero de J6venes Construyendo el Futuro. El Sol de Puebla. https://www.elsoldepuebla.com.mx/local/fraude-y-corrupcion-desvia n-dinero-de-joven es-construyendo-el-futuro-6299984.htmI Levy, S., & Szekely, M. (2016). cMas escolaridad, menos informalidad? Un analisis de cohortes para Mexico y America Latina. El Trimestre Economico, 83(332), 499-548. https:/ /doi .org/10.20430/ete.v83i332.232 Lorfa, Eduardo; Libreros, Catalina; Salas, Emmanuel (2012). La fey de Okun en Mexico: una mirada de genera, 2000.2-2077.7. lnvestigacion Economica, 71(280), 121-140. McFadden, D. (1974). Conditional logit analysis of qualitative choice behavior. En P. Zarembka (Ed.), Frontiers in econometrics (pp. 105-142). Academic Press. Martf nez, F. (2002). Nueva visita al pafs de la desigualdad. La distribucion de la escolaridad en Mexico, 1970-2000. Revista mexicana de investigacion educativa, 7(16). Mazuera Arias, Rina & Carda, Marisela & Quintero, Marfa & Santafe Rojas, Akever Karina & Carreno Paredes, Myriam Teresa & Peraza, Miguel & Martf nez-Rozo, Camila-Andrea & Vera, Miguel & Tinoco-Guerra, Antonio-Enrique & Ruscheinsky, Aloisia & Danubia, Cara-Treis & Benites-Lopes, Silva & Aldrete-Glez, Ana-Paola & Pocovi-Garzon, Patricia & Robles, Josefina & Morales-Carrasco, Lilian-Victoria & Lascano, Nelson & Roberto, Edison & Albornoz-Garzon, Marfa-Augusta. (2018). Mujer, emprendimiento y empleabilidad: una mirada interdisciplinaria. Ediciones Universidad Simon Bolfvar. https://doi.org/[si existe] ISBN: 978-958-5430-90-7 Mendez, R. (2025). 8M 2025: ''lusticia econ6mica para mujeres" es el gran pendiente de Mexico, seg(m organizaci6n. MVS Noticias. https://mvsnoticias.com/economia/2025/3/5/8m-2025-justicia-economica-pa ra-mujeres-es-el-gran-pendiente-de-m exico-segu n- organizacion-681468.htm I Miquel Hernandez, M. (2022). Una vision def funcionamiento def programa ''l6venes Construyendo el Futuro". Palimpsesto. Revista Cientffica de Estudios Sociales lberoamericanos, 12(20), 111-125. https://doi.org/10.35588/pa.vl 2i20.5248 Morales-Ramos, E. (2011). Los rendimientos de la educaci6n en Mexico, Working Papers, No. 2011-07, Banco de Mexico, Ciudad de Mexico Munoz Chavez, R. L., Munoz Chavez, J. P., Valle Cruz, D., & Barrios Quiroz, H. (2023). J6venes Construyendo el Futuro: Analisis y recomendaciones sabre el disefio de la polftica publica."The Relationship Between Education and Gender on Employability with Welfare Programs in Mexico: A Logit Analysis”RECAI. Revista de Estudios en Contaduria, Administraci6n e Informatica, 12(33), 1-19. https://doi.org/10.36677/recai.v12i33.19330 Murayama, C. (2011). Resefia de Mexico: desigualdad econ6mica y genera, de Flor Brown y Lilia Dominguez (Coords.). Revista de Economia Mundial, (29), 327-330. Sociedad de Economia Mundial. Disponible en: http://www.redalyc.org/articulo.oa?id=86622169013 Organisation for Economic Co-operation and Development. (2021). Perspectivas def empleo 2027: Reforzando la resiliencia def mercado laboral tras la crisis def C0VID-79. https://www.oecd.org/employment-outlook/ Centro de Estudios Espinosa Yglesias. (2021). c,Y despues que? lnserci6n laboral de j6venes egresados del programa J6venes Construyendo el Futuro. https:/ /ceey.org.mx/i nsercion-laboraI-jovenes-jcf / Organisation for Economic Co-operation and Development. (2022). Panorama de la juventud en Mexico 2022. https://www.oecd.org/mexico/panorama-de-la-juventud-en-mexico-2022.htm Pacheco, E., & Parker, S. (1996). Participaci6n econ6micamente activa femenina en el Mexico urhano. Un breve recuento y algunos hallazgos recientes. Problemas Del Desarrollo, 27(106), 21-33. http://www.jstor.org/stable/45480800 Ramirez Carrera, S. (2024, 17 de junio). Mujeres profesionistas y precarizaci6n laboral: El C!EC continua ciclo de conferencias en colaboraci6n con la Universidad de la Corufia. Gaceta Digital UNAM. Recuperado de https://www.gaceta.unam.mx/mujeres-profesionistas-y-precarizacion-la boraI/ Reyes, J. A. (1975). El desempleo en Mexico. Universidad Nacional Aut6noma de Mexico, Facultad de Economia. Secretaria del Trabajo y Prevision Social. (2023). lnforme Anual def Programa J6venes Construyendo el Futuro 2023. Gobierno de Mexico. https:/ /jovenesconstruyendoelfuturo.stps.gob. mx Torres Gaitan, R. (1974). Desempleo y subdesempleo en Mexico. Problemas del desarrollo, 5(18), 13-18. World Bank Open Data. (n.d.). lndicador CINI. https:/ / datos.ba ncomundia1.org/i nd icador/SI.POV.GIN l?locati ons=MX Yanez-Contreras, M., y Cano-Hernandez, K. del C. (2011). Determinantes def Desempleo: Una rev1s10n de la literatura. Panorama Econ6mico, 19(19), 135-148. https://doi.org/10.32997/2463-0470-vol.19-num.19-2011-352 Zenteno-Hidalgo, A. C., y Silva, C. A. D. (2016). Factoresy practicas de alto desempefio que influyen en el clima laboral: analisis de un caso. lnnovar: Revista de Ciencias Administrativas y Sociales, 26(59), 119-136.Cambron Jiménez David. logit dico becmil becmiledu becmilgenIteration 0: log likelihood = -1726.9208 Iteration 1: log likelihood = -1711.1179 Iteration 2: log likelihood = -1711.0924 Iteration 3: log likelihood = -1711.0924Logistic regressionNumber of obs=2534LR chi2(3)=31.66Prob > chi2=0.0000Log likelihood = -1711.0924Pseudo R2=0.0092dicoCoef.Std. Err.zP>|z|[95% Conf.Interval]becmil0.07219760.04007671.800.072-0.00635130.1507465becmiledu-0.09083670.0181023-5.020.000-0.1263166-0.0553567becmilgen-0.03663720.016272-2.250.024-0.0685298-0.0047447_cons0.44676160.20484322.180.0290.04527640.8482469. mfxMarginal effects after logit y = Pr(dico) (predict) = .57739797variabledy/dxStd. Err.zP>|z|[ 95%C.I. ]Xbecmil0.01761690.009781.800.072-0.0015460.036785.42947becmil~u-0.0221650.00441-5.020.000-0.030814-0.0135164.23678becmil~n-0.00893980.00397-2.250.024-0.016721-0.0011593.8704