Microsoft Word - 77-Manuscript-1049-1-11-20210324.docx Lamiño Jaramillo et al. Advancements in Agricultural Development Volume 2, Issue 1, 2021 agdevresearch.org 1. Pablo Lamiño Jaramillo, Graduate Assistant, Department of Agricultural Education & Communications, Texas Tech University, Box 42131, Lubbock, TX 79409-2131, pablo.lamino@ttu.edu, https://orcid.org/0000-0002-3941-4935 2. Amy Boren Alpízar, Associate Professor, Department of Agricultural Education & Communications, Texas Tech University, Box 42131, Lubbock, TX 79409-2131, amy.boren-alpizar@ttu.edu, https://orcid.org/0000-0001-9002-4855 3. Carla Millares-Forno, PhD, University of Nevada, Reno, 8050 Paradise Road, Las Vegas, NV, 89123, cmillares@unr.edu https://orcid.org/0000-0003-0482-9847 4. Rafael Quijada Landaverde, Gradate Associate, Department of Agricultural Communication, Education, and Leadership, 200 Fyffe road, Columbus, Ohio, 43210-6909, quijadalandaverde.1@buckeyemail.osu.edu https://orcid.org/0000- 0001-6489-0477 70 Agricultural Education and Migration: The Cases of El Salvador and Honduras P. Lamiño Jaramillo1, A. Alpízar2, C. Millares-Forno3, R. Quijada Landaverde4 Abstract Youth is defined as a period of transition from childhood to adulthood that ranges between the ages of 15 and 24, and it is the most mobile social group in the world. Youth migration in El Salvador and Honduras is a well-known problem; however, limited research has focused on the causes of migration and the impact of agricultural education programs on the decision to migrate. This study aims to identify and compare students' migration intentions in agricultural and non-agricultural programs of two rural communities of El Salvador and Honduras. For this quantitative study, a two- group model was used; the target population (N = 209) was composed of high school students with similar socioeconomic characteristics. Overall, the comparison between El Salvador and Honduras showed a significant difference between countries regarding their migration intentions (p < 0.05). Salvadorian students presented a stronger willingness to migrate. Moreover, students' intention of migration was evaluated based on their educational background. Youth who were not part of a formal agricultural program have a higher intention of migrating (p < 0.05). Finally, the main and interaction effects of intention to migrate, country of origin, and academic program based on the different migration drivers were analyzed. Keywords Youth migration, agricultural education programs, El Salvador, Honduras, rural Lamiño Jaramillo et al. Advancements in Agricultural Development https://doi.org/10.37433/aad.v2i1.77 71 Introduction and Problem Statement According to the United Nations (UN, 2019), youth is defined as a period of transition from the dependence of childhood to adulthood's independence that ranges from 15 and 24 years old. It is expected that by 2030, the number of youth will grow from 1.2 billion to nearly 1.3 billion globally. Moreover, this age group is the most mobile social group in the world (Global Migration Group, 2014). This growth and mobility add pressure to governments because of the need to provide youth with the necessary services (UN, 2018). For instance, approximately 47% of Salvadorans and 52% of Hondurans are under the age of 25, and most of them are unemployed with little opportunity for gainful employment, forcing them to seek other options to make a living, such as migration (Congressional Research Service, 2019). In 2015, the number of international migrants reached 244 million; this number is small compared to the noticeably higher number of migrants (740 million) who move inside their countries, mainly from rural to urban areas or from one rural area to another (Food and Agriculture Organization [FAO], 2016). The decision to stay in their home communities or migrate to other areas holds great economic, social, and emotional consequences, especially for rural students (Theodori & Theodori, 2014). The impact of youth's decision to migrate is important for rural communities because of the potential loss of labor force, community leaders, volunteers, and parents of future generations (Demi et al., 2009). Youth migration in El Salvador and Honduras is a recognized problem, and several reasons such as poverty, violence, insecurity, and family reunification in the cities have been identified as potential drivers of this phenomenon (Warren & Kerwin, 2017). Education is often a driver of rural-to-urban migration; rural people who want to pursue a higher education level are more willing to leave their communities to seek educational opportunities (Corbettm, 2007; Kodzycki, 2001). However, this is not the case with all the educational programs. Agricultural education programs teach students better agricultural practices and motivate them to replicate new techniques on their lands, which can reduce migration (Rhoda, 1983). Youth migration is a subject of significance in these countries; unfortunately, limited studies have been developed in Central America to determine the impact of agricultural education programs on the decision to migrate (Roth & Hartnett, 2018). Theoretical and Conceptual Framework The Theory of Planned Behavior (TPB) by Ajzen (1991) was used to guide this study. The TPB aims to predict behavioral intention based on three primary constructs: attitude toward the behavior, subjective norms, and perceived behavioral control. The first construct of the theory is the attitude toward the behavior, which is the favorable or unfavorable evaluation of a given behavior. The second construct is subjective norms that are social pressure to execute or not execute a given behavior. Perceived behavioral control is the third construct and refers to people's perception of the ease or difficulty of performing the behavior of interest. The combination of the three constructs allows predicting future intentions (Ajzen, 1991). Lamiño Jaramillo et al. Advancements in Agricultural Development https://doi.org/10.37433/aad.v2i1.77 72 Although the TPB has been recognized for its facility of predicting behavior, the theory continues to evolve, and other variables might increase the model's utility (Burton, 2004; Whitmarsh & O'Neill, 2010). Ajzen (1991, p.179) argued the following regarding adaptations "in principle, open to the inclusion of additional predictors if it can be shown that they capture a significant proportion of the variation in intention or behavior." Therefore, for this study, the migration drivers were selected using previous literature and research experience, and it has been designed to match with the TPB (Castelli, 2018; Foroughi et al., 2001; Ramos-Vidal et al., 2019; Yazdan-Panah et al., 2017; Yazdan-Panah & Zobeidi, 2017). Thus, subjective norms, the ideas that other community members have regarding migration, are composed of Subjective Expectations, Interpersonal Ties, Social Support, and Disputes. The perceived migration behavioral control, the perceived level of difficulty regarding migration, was composed of Access to Extension Activities and Residential Satisfaction. The attitude toward migration, the positive or negative idea of migration, comprises environmental impacts (Ajzen, 1985; Yazdan-Panah et al., 2017). Figure 1 shows how selected migration drivers were added to TPB variables. Figure 1 Theory of Planned Behavior Note. This figure shows the relationship that each component has with the predictable behavior. Adapted from the study (Ajzen, 1991) Attitude Toward Migration Subjective Norms Migration Behavior Perceived Migration Behavioral Control Actua l beha viora l Contr ol Social Support Interpersonal Ties Subjective Expectations Disputes Access to extension activities Residential Satisfaction Environmental Impacts Intention to Migrate Lamiño Jaramillo et al. Advancements in Agricultural Development https://doi.org/10.37433/aad.v2i1.77 73 Purpose The study aims to identify and compare the migration intentions of high school students in agricultural and non-agricultural programs from two rural communities, one from El Salvador and one from Honduras. The following objectives guided the study: 1. Compare students' intention to migrate by country of origin. 2. Compare students' intention to migrate by academic program. 3. Determine the main and interaction effects of intention to migrate, country of origin, and academic program based on the different migration drivers. 4. Calculate the results managing missing data with two techniques, pairwise deletion and mean imputation. Methods For this quantitative study, a two-group model was used. The target population (N = 209) was composed of high school students with similar economic and social characteristics from two communities, Chalatenango, El Salvador (n = 104), and Jesus de Otoro, Honduras (n = 105). Two groups from each community were recruited, one group participated in a formal Agricultural Education Program (AEP), and the other group was part of a Non-Agricultural Education Program (Non-AEP). Data were collected using a paper-pencil instrument that explored Salvadorian and Honduran youth's intention to migrate. The survey had five sections and 50 questions: demographic information, participants' academic and agricultural background, migration drivers, residential satisfaction, and intention to migrate. For the "Migration Drivers" section, 34 5-point Likert- type questions were created by the researchers, ranging from 1 (completely disagree), 2 (Disagree), 3 (Neither Agree nor Disagree Agree), 4 (Agree), 5 (completely agree). Scales containing multiple items were developed to measure the following migration drivers: social participation, social support, access to extension activities, environmental impacts, interpersonal ties, disputes, residential satisfaction, and subjective expectations (Ajzen, 1985 & Yazdan-Panah et al., 2017). The drivers were selected based on previous literature and under experts' criteria (Castelli, 2018; Foroughi et al., 2001; Ramos-Vidal et al., 2019; Yazdan-Panah et al., 2017). The instrument was pilot tested to determine its reliability and validity. A panel of experts confirmed the questionnaire's face validity and field-tested in a pilot study that included 20 young people from both countries. Cronbach's alpha was calculated to measure constructs' reliabilities. For social participation, the reliability was .74, social support .73, access to extension activities .89, environmental impact .75, interpersonal ties .74, disputes .77, residential satisfaction .73, and subjective expectations .73. Results ranged from .73 to .89, which means that they were acceptable (Rubin & Babbie, 2009). Lamiño Jaramillo et al. Advancements in Agricultural Development https://doi.org/10.37433/aad.v2i1.77 74 The surveys with more than 10% of missing values (six surveys) were deleted from the data analysis (Raaijmakers, 1999). Missing values were between 2 and 8% for each Likert-type per statement. Therefore, data analyses were conducted independently, first using the Pairwise Deletion Technique (PDT) and then using Mean Imputation Technique (MIT) for the missing data. The PDT is used in missing data that are independent and all the missing cases to be used in the analysis (Field, 2013, p.231; Shi et al., 2020). The MIT allows using the mean value of the variable in place of the missing data point. Field (2013, p.231) suggested replacing the missing score with the average score when there is a large sample and a small number of missing values. The data collected were analyzed based on the objectives. Descriptive statistics were used to understand the participants of the study. Different independent chi-square was conducted to compare migration intention depending on the country of origin and academic program. Finally, a 2 x 3 x 2 Factorial Multiple Analysis of Variance (MANOVA) was conducted to examine the main effects and interactions effects of independent variables of country of origin ("El Salvador" vs. "Honduras"), academic program ("AEP" vs. "Non-AEP"), and intention to migrate ("Yes" vs" I don't know" vs "No") on migration drivers, including Social Participation, Access to Extension Activities, Social Support, Interpersonal Ties, Disputes, Subjective Expectations, Environmental Impacts, and Residential Satisfaction. Findings The study aimed to identify and compare the migration intentions of high school students in agricultural and non-agricultural programs from El Salvador and Honduras. The sample consisted of 104 youth in AEP and 105 in a Non-AEP. In El Salvador, more participants were from AEP (n = 54, 51.92%), than Non-AEP (n = 50, 48.08%) programs. On the other hand, in Honduras, more participants were in a Non-AEP (n = 55, 52.38%) than an AEP (n = 50, 47.62%) program. In total, 106 males and 103 females participated in this study. In El Salvador, the majority of participants were male (n = 65, 62.50%), while in Honduras, most of the participants were female (n = 64, 61.95%). One hundred and twenty-three participants considered the father as the family head, and 86 considered the mother as the family head. In El Salvador, the majority considered men as the family head (n = 71, 68.27%), whereas in Honduras, women were considered the family head (n = 53, 50.48%). Concerning students' access to land, Honduras had a greater number of students from families that own land (n = 78, 74.29%) than the participants from El Salvador (n = 39, 37.50%). Table 1 shows the general and specific distribution by academic program, gender, family head, and access to land. Lamiño Jaramillo et al. Advancements in Agricultural Development https://doi.org/10.37433/aad.v2i1.77 75 Table 1 Summary of Sociodemographic information Characteristics Total (N = 209) El Salvador (n = 104) Honduras (n = 105) f % f % f % Academic Program AEP 104 49.76 54 51.92 50 47.62 Non-AEP 105 50.24 50 48.08 55 52.38 Gender Male 106 50.72 65 62.50 41 39.05 Female 103 49.28 39 37.50 64 61.95 Family Head Male 123 58.85 71 68.27 52 49.52 Female 86 41.15 33 31.73 53 50.48 Access to Land Landowners 117 55.98 39 37.50 78 74.29 Non-Landowners 87 41.63 65 62.50 22 20.95 Prefer not to say 5 2.39 5 4.76 To answer objective one, a Chi-Square test of independence was calculated comparing the intention to migrate (yes = 1, I do not know = 2, and no = 3) and country of origin ("El Salvador," and "Honduras"). Table 2 shows the results of PDT and MIT methods. Results were the same and showed a significant relationship between intention to migrate and country of origin (c2 (2) = 17.77, p < .001). In total, 206 youth participated in the intention to migrate section. Overall, 56.73% of Salvadorian youth considered migration as an option, while 25.96% said No, and 17.30% were undecided. In Honduras, 36.27% of the youth have no intention to migrate, followed by 35.29% undecided, and 28.43% have the intention to migrate. Cramer's Value was used to determine the association's strength; results showed a value of .29, considered as medium effect size (Cohen, 1988). Table 2 Chi-Square Results for Intention to Migrate Based on Country of Origin (N = 206) Intention to migrate El Salvador (n =104) Honduras (n = 102) c2 (2) Cramer's V n % n % Yes 59 56.73 29 28.43 17.77* .29 I don't know 18 17.30 36 35.29 No 27 25.96 37 36.27 Note. *p < .05 For objective two, a Chi-Square test of independence was calculated comparing the intention to migrate (Yes = 1, I don't know = 2, and No = 3) and academic program ("AEP," and "Non-AEP"). Table 3 shows the results of the Chi-square using PDT and MIT. Results show a significant Lamiño Jaramillo et al. Advancements in Agricultural Development https://doi.org/10.37433/aad.v2i1.77 76 relationship between intention to migrate and academic program (c2 (2) = 27.67, p < .001). Overall, 39.81% of the youth in the agricultural program do not have the intention to migrate, while 30.10% said yes, and 30.10% were undecided; 55.34% of students in Non-AEP considered migration as a good option, followed by 23.33% undecided, and 23.33% do not have the intention to migrate. The Cramer's Value for the strength of the association was .37, considered a medium effect size (Cohen, 1988). Table 3 Chi-Square Results for Intention to Migrate Based on Academic Program (N = 206) Intention to migrate AEP (n =104) Non-AEP (n = 102) c2 (2) Cramer's V n % n % Yes 26 25.00 62 60.80 27.67* .37 I don't know 38 36.50 16 15.70 No 40 38.50 24 23.50 Note. *p < .05 For objective three, a 2 x 3 x 2 Factorial MANOVA was conducted to examine the main effects and interactions effects of independent variables of the country of origin ("El Salvador" vs. "Honduras"), academic program ("AEP" vs. "Non-AEP"), and Intention to migrate ("Yes" vs." I don't know vs. "No") have on the migration drivers, including Social Participation, Access to Extension Activities, Social Support, Interpersonal Ties, Disputes, Subjective Expectations, Environmental Impacts, and Residential Satisfaction. The assumption of normality was met because skewness and kurtosis levels were in the range of +/- 2, and a histogram showed a normal distribution. According to Levene's test, the homogeneity of variance was no accomplished in the study. Therefore, Pillai's Trace was used to interpret the results to guarantee the robustness of the findings, despite the assumption violation. Finally, the homogeneity of variance was calculated using Box's test. Results show a Box's M value of 349.24 with a p-value of .03 for PDT and a Box’s M value of 469.56 with a p- value of 0.04 for MIT, which represents a violation of the assumption. However, Field (2013) argues that Box's test findings are sensitive to large sample sizes, and the significant value could be the result of the sample. Table 4 shows the results of the sources that were significant. For intention to migrate, and using PDT, there was a significant medium effect, V = .26, F(16,270) = 2.47, p = .02, partial η2 =.13; for MIT there was reported a significant small effect ,V = .19, F(16,354) = 2.29, p = .03, partial η2 = .09. Using PDT and MIT, a significant medium effect was found on country of origin, V = .28, F(8,134) = 6.44, p < .001, partial η2 = .28); V = .21, F(8,176) = 5.76, p < .001, partial η2 = .21). For Academic program, a significant large effect was found for both methods, PDT, V = .49, F(8,134) = 16.09, p < .001, partial η2 = .49, and MIT , V = .45, F(8,176) = 18.38, p < .001, partial η2 = .46. Finally, for the interaction country of origin and academic program a significant medium effect size was found, using PDT, V =.32, F(8,134) = 7.93, p < .001,partial η2 = .32, and MIT, V =.27, F(8,176) = 8.15, p < .001, partial η2 = .27 (Cohen, 1988). Lamiño Jaramillo et al. Advancements in Agricultural Development https://doi.org/10.37433/aad.v2i1.77 77 Table 4 Multivariate Analysis of Variance for Migration Intentions Method Source PDT MIT F p partial η2 F p partial η2 Academic Program (1) 16.09 < .001* .49 18.38 < .001* .46 Country of Origin (2) 6.44 < .001* .28 5.76 < .001* .21 Intention to Migrate (3) 2.47 .02* .13 2.29 .03* .09 Interaction 1 x 2 7.93 < .001* .32 8.15 < .001* .27 Interaction 1 x 3 1.35 .16 .07 1.05 .40 .05 Interaction 2 x 3 .85 .62 .48 1.02 .43 .04 Interaction 1 x 2 x 3 .89 .52 .50 1.26 .26 .05 Note. *p < .05; PDT = Pairwise Deletion Technique; MIT = Mean Imputation Technique As complementary tests for the significant MANOVA, univariate Analysis of Variance (ANOVAs) were implemented to understand the interaction effects of the independent variables on the eight migration drivers. Therefore, the ANOVA revealed a significant main effect for country of origin, PDT, F(1,141) = 9.97, p = .01, partial η2 = .06; MIT, F(1,183) = 9.34, p = .01, partial η2 = .05, on the migration driver, “Social Participation.” For the Access to Extension Activities construct, the ANOVAs revealed a significant effect, for country of origin, PDT, F(1,141) = 10.20, p = .01, partial η2 =.07; MIT, F(1,183) = 10.93, p = .01, partial η2 =.06, and academic program, PDT, F(1,141) = 89.82, p < .001, partial η2 = .39, and MIT, F(1,183) = 83.18, p < .001, partial η2 = .31. Besides, it was found a significant effect in the interaction of country of origin and academic program using PDT, F(1,141) = 4.67, p = .03, partial η2 = .03; MIT, F(1,183) = 3.21, p = .07, partial η2 = .01. For the Social Support construct, a significant effect was observed in country of origin, PDT, F(1,141) = 15.57, p < .001,partial η2 = .10; MIT, F(1,183) = 16.91, p < .001,partial η2 = .09. Interpersonal Ties construct had a significant effect for country of origin, using the PDT, F(1,141) = 6.65, p = .01, partial η2 = .05; MIT, F(1,183) = 6.25, p = .01, partial η2 = .03. Moreover, it was found a significant effect for intention to migrate, using MIT, F(2,183) = 3.40, p = .03, partial η2 = .04; but not found significant effect for PDT, F(2,141) = 2.38, p = .09, partial η2 = .03. For the Dispute construct, a significant effect was found for academic program, PDT, F(1,141) = 42.85, p < .001, partial η2 = .23; MIT, F(1,183) = 34.64, p < .001, partial η2 = .16, and the interaction between of country of origin and academic program, PDF, F(1,141) = 37.26, p < .001, partial η2 = .21; MIT, F(1,183) = 35.30, p < .001, partial η2 = .16. Country of origin was significant only using the PDT method, F(1,141) = 11.67, p = .01, partial η2 = .08, MIT, F(1,183) = 3.50, p = .06, partial η2 = .02. Finally, for Environmental Impacts construct, ANOVAs revealed a significant effect on intention to migrate, PDT, F(2,141) = 7.35, p = .01, partial η2 = .09, and MIT, Lamiño Jaramillo et al. Advancements in Agricultural Development https://doi.org/10.37433/aad.v2i1.77 78 F(2,183) = 8.16, p < .001, partial η2 = .08. Table 5 summarizes the ANOVAs that are significant, depending on each migration driver, and using the two processes to manage missing data. Table 5 Univariate Analysis of Variance for Social Participation, Access to Extension Activities, Social Support, Interpersonal Ties, Disputes, and Environmental Impacts as a Function of Main and Interaction Effects of the Intention to Migrate, Country of Origin, Academic Program Source Method PDT MIT F p partial η2 F p partial η2 Social Participation Country of Origin (2) 9.97 .01* .06 9.34 .01* 0.5 Access to Extension Activities Country of Origin 10.20 .01* .07 10.93 .01* .06 Academic Program (1) 89.82 < .001* .39 83.18 < .001* .31 Interaction 1 x 2 4.67 .03* .03 3.21 .07 .01 Social Support Country of Origin 15.57 < .001* .10 16.91 < .001* .09 Interpersonal Ties Country of origin 6.65 .01* .05 6.25 .01* .03 Intention to migrate (3) 2.38 .09 .03 3.40 .03* .04 Disputes Academic Program 42.85 < .001* .23 34.64 < .001* .16 Country of Origin 11.67 .01* .08 3.50 .06 .02 Interaction 1 x 2 37.26 < .001* .21 35.30 < .001* .16 Environmental Impact Intention to migrate 7.35 .01* .09 8.16 < .001* .08 Note. *p < .05; PDT = Pairwise Detention Technique; MIT = Mean Imputation Technique; the variables Subjective Expectations and Residential Satisfaction was not significant for any source. Post-hoc test was conducted for the intention to migrate variable because it was the only independent variable that had more than two levels. The researchers assumed that the sample sizes per condition were equal; therefore, the Bonferroni was implemented; this test is well- known for being conservative and has strong statistical power (Field, 2011, p.374). Table 6 shows the Post-hoc test used as a follow-up test to determine the differences between the means of intention to migrate. Findings from the PDT shows that for the construct of Environmental Impacts, the migration option “I don’t know” (M = 3.21, SD = .77) received a higher score than the migration options “No” (M = 2.69, SD = .77) and “Yes” (M = 2.60, SD = 0.61). For the MIT, the results were similar for the construct, the migration option “I don’t know” (M = 3.23, SD = .84) received a higher score than the migration options “Yes” (M = 2.72, SD = 0.70) and “No” (M = 2.70, SD = .78). ANOVA result using MIT, showed significance on the Ties construct; however, further analysis did not confirm the previous results. Lamiño Jaramillo et al. Advancements in Agricultural Development https://doi.org/10.37433/aad.v2i1.77 79 Conclusions, Discussion, and Recommendations Results of this research study are supported by previous studies that established an increased willingness to migrate from Salvadorian youth than Honduran youth (Congressional Research Service, 2019; Rodriguez et al., 2019). The comparison of youth from AEP and Non-AEP showed that investing in agricultural education programs reduces youth's intention to migrate. These results match with findings from Rhoda (1983), whose research indicated that agricultural program interventions might reduce migration intentions. The 17 % of Salvadorian and 35 % of Honduras youth are indecisive about migrating. Since this group is still undecided, the TPB suggests that this group's members can be influenced in one direction or the other. As such, it would be beneficial for programs to increase youth's involvement in community activities to increase their sense of belonging and social participation. The perception of the environment was significantly different in the intention to migrate variable and Bonferroni post-hoc confirmed the results. Consequently, future studies should explore the effect that the perception of the environment has on youth intentions to migrate and explore if other variables can affect this interaction. Furthermore, future studies could examine young people's perspectives in agricultural and other technical education programs in rural areas in other countries to create country profiles for rural agricultural education. Understanding the young people's ideas of their opportunities at home in agriculture and other areas will be important in rural development initiatives. Future research exploring rural youth's intention to migrate should examine the youth migration issue in more depth. Ethnographic qualitative studies would shed much needed light on the migration intentions of youth. Additional studies could be developed evaluating youth in other countries and the impact that other variables have on migration intention, such as access to inputs. Replicating this study in other regions besides El Salvador and Honduras could also strengthen the results and provide more robust data. It is important to evaluate the pros and cons of using different missing data techniques. For this study, the use of PDT and MIT displayed the same result for the Chi-square; however, ANOVA results, were different based on the missing data technique used demonstrating that results can be influenced, and as a result, the interpretation of the analysis can be influenced. Therefore, it is important to report the method used to manage missing data. Lamiño Jaramillo et al. Advancements in Agricultural Development https://doi.org/10.37433/aad.v2i1.77 80 References Ajzen, I. (1985). From intentions to actions: A theory of planned behavior. In Kuhl J. & Beckmann J. (eds) Action Control ( pp. 11-39). SSSP Springer Series in Social Psychology. Springer. https://doi.org/10.1007/978-3-642-69746-3_2. Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179–211. Burton, R. J. (2004). Reconceptualising the 'behavioural approach' in agricultural studies: A socio-psychological perspective. Journal of Rural Studies, 20(3), 359–371. http://dx.doi.org/10.1016/j.jrurstud.2003.12.001 Castelli, F. (2018). Drivers of migration: Why do people move? Journal of Travel Medicine, 25(1). https://doi.org/10.1093/jtm/tay040 Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Lawrence Erlbaum Associates Congressional Research Service. (2019). Central American migration: Root causes and U.S. policy. CRS. Corbett, M. (2007). Learning to leave: The irony of schooling in a coastal community. Fernwood Books. Demi, M. A., McLaughlin, D. K., & Snyder, A. R. (2009). Rural youth residential preferences: understanding the youth development-community development nexus. Community Development, 40(4), 311–330. https://doi.org/10.1080/15575330903279606 Field, A. (2013). Discovering statistics using IBM SPSS statistics. Sage publications. Food and Agriculture Organization. (2016). Addresing rural youth migration at its root causes: A conceptual framwork. FAO. Foroughi, E., Misajon, R., & Cummins, R. A. (2001). The relationships between migration, social support, and social integration on quality of life. Behaviour Change, 18(3), 156–167. https://doi.org/10.1375/bech.18.3.156 Global Migration Group. (2014). Migration and youth: Challenges and opportunities. UNICEF. https://globalmigrationgroup.org/system/files/23._Key_Messages_and_Policy_Recomm endations.pdf Lamiño Jaramillo et al. Advancements in Agricultural Development https://doi.org/10.37433/aad.v2i1.77 81 Kodrzycki, Y. K. (2001). Migration of recent college graduates: Evidence from the national longitudinal survey of youth. New England Economic Review, 1(January/Feburary), 13- 34. https://www.bostonfed.org/home/publications/new-england-economic- review/2001-issues/issue-number-1-januaryfebruary-2001/migration-of-recent-college- graduates-evidence-from-the-national-longitudinal-survey-of-youth.aspx Raaijmakers, Q. A. W. (1999). Effectiveness of different missing data treatments in surveys with likert-type data: Introducing the relative mean substitution pproach. Educational and Psychological Measurement, 59(5), 725–748. https://doi.org/10.1177/0013164499595001 Ramos-Vidal, I., Villamil, I., & Uribe, A. (2019). Underlying dimensions of social cohesion in a rural community affected by wartime violence in Colombia. International Journal of Environmental Research and Public Health, 16(2), 195. https://doi.org/10.3390/ijerph16020195 Rhoda, R. (1983). Rural development and urban migration: Can we keep them down on the farm? International Migration Review, 17(1), 34–64. https://doi.org/10.1177/019791838301700102 Rodriguez, N., Urrutia-Rojas, X., & Gonzalez, L. R. (2019). Unaccompanied minors from the Northern Central American countries in the migrant stream: Social differentials and institutional contexts. Journal of Ethnic and Migration Studies, 45(2), 218–234. https://doi.org/10.1080/1369183X.2017.1404257 Roth, B. J., & Hartnett, C. S. (2018). Creating reasons to stay? Unaccompanied youth migration, community-based programs, and the power of "push" factors in El Salvador. Children and Youth Services Review, 92(September), 48–55. https://doi.org/10.1016/j.childyouth.2018.01.026 Rubin, A., & Babbie, E. (2009). Essential research methods for social work. Brooks. Shi, D., Lee, T., Fairchild, A. J., & Maydeu-Olivares, A. (2020). Fitting ordinal factor analysis models with missing data: A comparison between pairwise deletion and multiple Imputation. Educational and Psychological Measurement, 80(1), 41–66. https://doi.org/10.1177/0013164419845039 Theodori, A. E., & Theodori, G. L. (2014). Perceptions of community and place and the migration intentions of at-risk youth in rural areas. Journal of Rural Social Sciences, 29(1), 103-121. https://egrove.olemiss.edu/jrss/vol29/iss1/5 United Nations. (2018). Youth and the 2030 Agenda for Sustainable Development. Department of Economic and Social Affairs Youth. https://www.un.org/development/desa/youth/world-youth-report/wyr2018.html Lamiño Jaramillo et al. Advancements in Agricultural Development https://doi.org/10.37433/aad.v2i1.77 82 United Nations. (2019, September 23). Youth. Who are Youth?: https://www.un.org/esa/socdev/documents/youth/fact-sheets/youth-definition.pdf Warren, R., & Kerwin, D. (2017). A statistical and demographic profile of the US temporary protected status populations from El Salvador, Honduras, and Haiti. Journal on Migration and Human Security, 5(3), 577–592. https://doi.org/10.1177/233150241700500302 Whitmarsh, L., & O'Neill, S. (2010). Green identity, green living? The role of pro-environmental self-identity in determining consistency across diverse pro-environmental behaviours. Journal of Environmental Psychology, 30(3), 305–314. https://doi.org/10.1016/j.jenvp.2010.01.003 Yazdan-Panah, M., & Zobeidi, T. (2017). Why do they want to migrate from rural areas? A psychological perspective from Iran. International Journal of Agricultural Management and Development, 7(3), 283–292. https://doi.org/10.22004/AG.ECON.292492 Yazdan-Panah, M., Zobeidi, T., & Bakhtiari, Z. (2017). Determinate of factors influencing intention to stay and reside in rural among rural students. Journal of Research and Rural Planning, 6(3). https://doi.org/10.22067/jrrp.v6i3.60560 © 2021 by authors. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/4.0/).