Contexts of Educational Aspirations and School Grades of Rural Students

PERSISTENT CHALLENGES IN RURAL COMMUNITIES AND SCHOOLS

Rural communities have higher poverty rates that affect not only students and their education, but also adults and families more likely to live in multigenerational poverty (Thiede et al. 2018; Lichter and Schafft 2016). Even when educational attainment has increased in rural areas, the differences between rural and urban communities remain; for example, 33 percent of the U.S. urban population have a bachelor’s degree, but only 19 percent do in rural areas (Marré 2017). We see similar patterns regarding earnings. There is a gap between urban and rural regions where workers with similar education are offered higher earnings and more employment opportunities in urban areas (Marré 2017). Living in concentrated poverty is associated with poor housing, poor health conditions, higher school dropout rates, and employment dislocations (Economic Research Service 2020). Disparities persist between rural and urban communities in educational, economic, and social conditions.

The distribution of income and race is also a topic of recent interest because it provides an important lens for creating greater equity in opportunities for youth and families. Through a joint-distribution analysis of race and income, Sean Reardon, Joseph Townsend, and Lindsay Fox (2017) identify a number of trends based on data from the American Community Survey. For example, black and Latino individuals and families of any income level tend to live in neighborhoods that are substantially poorer than those of white or Asian residents. They find that many low-income white households are in rural areas, and because of this, most poor white residents live in majority white neighborhoods. Much of this work addresses the trends and conditions underneath a larger trend toward segregation.

Evidence indicates that urban-rural and poverty indicators are not sufficient to understand the role of location and poverty in child and youth development. Portia Miller, Elizabeth Votruba-Drzal, and Rebekah Levine Coley (2019) argued that place, poverty, and community resources and stressors provide a richer perspective. They used data from the Early Childhood Longitudinal Study, Kindergarten class of 2010, which contained data from children’s experiences and contexts while in kindergarten through second grade. Generally, they described the complex nature of ecosystems that interact. The positive role of rich cultural assets in urban communities was diminished by greater violence and disadvantage. Location had limited associations with parenting style and child achievement; nurturing supportive parenting had small positive associations with achievement. The core of their model suggested that urbanicity (location) had direct effects on cultural and community resources as well as community violence and disadvantage, which directly affected achievement and indirectly affected achievement through cognitive stimulation and parental nurturing. The associations between urbanicity and community contexts and conditions were substantial, but their effects on achievement, both direct and indirect, were small.

This complex interaction of place, poverty, and community resources and stressors also was relevant to the experiences and opportunities of individuals in different racial and ethnic communities. With respect to youth achievement, community and cultural capital have been core components of the ecology of youth development, and these characteristics have differed across ethnic groups (see, for example, Rodriguez and Morrobel 2006, who built a case for the consideration of unique family and community characteristics relevant to Latino youth development).

SCHOOL GRADES AND EDUCATIONAL ASPIRATIONS

Educational aspirations and achievement are associated with short-term outcomes such as access to higher education, and long-term outcomes such as better occupational attainment, employment, and health status (Bohon, Johnson, and Gorman 2006). Further, aspirations are also associated with positive transition into adulthood, given that youth with clear goals and healthy behaviors tend to adapt better to transitions (Bandura et al. 2001; Eccles, Brown, and Templeton 2008).

For students from rural schools, Macarena Hernández and Dagmar Raczynski (2016) find in Chile that students want to achieve higher education levels and obtain good jobs. At the same time, they also identify major obstacles, including limited economic resources, school supports, and information. On a similar note, parents say that their main concern about their children’s education was not having the economic resources to help them (Guerrero 2014). It is possible that developing college aspirations in rural communities might be more challenging because of the high rates of poverty relative to urban areas.

Although promoting college aspirations is important for students, it must be accompanied by social and emotional support and access to information, especially for disadvantaged students such as those from rural schools, given that they are more likely to struggle with stress, financial limitations, and the uncertainties of moving away from their communities (Cottom 2017; Hektner 1995; Silva and Snellman 2018). In one of the most comprehensive studies of rural student achievement and aspirations, based on a 2007 national sample of more than six thousand students from sixty-four rural high schools, Matthew Irvin and his colleagues (2011) explore elements of school contexts and their associations with educational achievement and aspirations among rural youth.

The focus of this effort (Irvin et al. 2011) was on the student experience in high-poverty rural schools (n = 21) relative to low-poverty rural schools (n = 43). Although high- and low-poverty communities did not differ regarding educational attainment, as expected, some student characteristics (such as disability status, grade retention, parent expectations) were associated with lower educational attainment as well as school characteristics (such as the presence of a college preparation program and postsecondary activities, as well as school-level academic self-concept). Regarding academic achievement, no reliable differences were found between high and low-poverty communities, but again student-level and school-level characteristics were associated with school grades (achievement).

In 2016, based on a national sample of rural high schools, Matthew Irvin and his colleagues explored school characteristics and their associations with aspirations among African American, Latino, and American Indian rural youth. They find that educational aspirations across racial-ethnic groups were similar, and that school experiences such as school valuing, academic self-concept, and teachers’ educational expectations shaped students’ educational aspirations. In addition, Latino and American Indian students attended rural schools with more race segregation and less access to resources than white rural students did. Unfortunately, the 2016 and 2011 studies had similar limitations: the authors did not properly account for the nested nature of the data, leading to misestimated standard errors, potentially affecting inferences.

POSITIVE YOUTH DEVELOPMENT

Youth development researchers have focused on risk factors such as violence, poverty, substance use, and dropouts, namely, deficit-oriented models. As a consequence, research, policy, and practice have been misguided when the target population has been vulnerable adolescents. In contrast, the positive youth development approach addresses social and psychological factors that improve child and youth outcomes, focusing on developmental assets and positive qualities of youth (Lerner 2003; Benson 2003). These not only promote positive outcomes and thriving, but also reduce risky behaviors. Therefore, with a purpose of taking a different perspective and creating a basis for building youth and community assets, as a way to counteract or mediate the persistent challenges facing urban and rural communities, we rely on principles of positive youth development (Damon 2004). Doing so allows us to take a positive orientation in thinking about possible futures, rather than the typical deficit-oriented medical-model approach to youth development.

At least six propositions underlie positive youth development approaches with broad consensus (Benson 2006):

youth have the inherent capacity for positive development;

positive development is enabled when youth are embedded in relationships, contexts, and environments that nurture development;

positive development is further enhanced when youth participate in multiple meaningful relationships, contexts, and environments;

all youth benefit from these opportunities, the benefits of which generalize across gender, race, ethnicity, and family income;

community is a critical delivery system for positive youth development; and

youth themselves are major actors in their own development, serving as a central resource for creating the kinds of relationships, contexts, environments (ecologies), and communities that facilitate optimal development.

The ecologies and developmental contexts where youth are located interact with the inherent capacity of youth to grow and thrive; their developmental strengths, skills, competencies, values, and dispositions (Bronfenbrenner 1979); and two related aspects of developmental success, the reduction of high-risk behaviors and the promotion of healthy well-being or thriving (Benson 2006). Researchers are increasingly exploring many aspects of context that might influence positive youth development, including success in school and beyond—providing useful information for strong program and policy development.

SOCIAL AND EMOTIONAL LEARNING

A body of evidence and consensus (Eccles and Gootman 2002; Jones and Kahn 2017) also exists among youth development researchers and experts about the integral role of social and emotional learning. Two related definitions for social and emotional learning come from the Aspen Institute and CASEL. Aspen’s definition is that “Social and emotional development comprises specific skills and competencies that people need in order to set goals, manage behavior, build relationships, and process and remember information.” CASEL’s is that “Social and emotional learning (SEL) is an integral part of education and human development. SEL is the process through which all young people and adults acquire and apply the knowledge, skills, and attitudes to develop healthy identities, manage emotions and achieve personal and collective goals, feel and show empathy for others, establish and maintain supportive relationships, and make responsible and caring decisions.”¹

The Aspen Institute (2018, 2) argues that social, emotional, and academic development can be integrated in ways to reinforce educational equity, which it defines as a system in which “every student has access to the resources and educational rigor they need at the right moment in their education, irrespective of race, ethnicity, gender, sexual orientation, language, disability, family background, family income, citizenship, or tribal status.” This suggests that SEL may play an important role in students’ development, both in urban and rural settings.

RESEARCH QUESTIONS

Our goal was to investigate the educational experiences and contexts of rural youth, their school grades, and educational aspirations. Using the Minnesota Student Survey from 2013, 2016, and 2019, we addressed three research questions:

Are school grades and educational aspirations different between students from rural and urban school districts?

Are developmental skills, supports, and challenges associated with school grades and educational aspirations for rural and urban students?

Do school grades and educational aspirations differ for American Indian and Latino students in rural and urban communities (as these are the two larger nonwhite communities in rural Minnesota)?

First, we sought to provide insights about educational outcomes in underserved youth groups in the United States, including rural, American Indian, and Latino communities. Second, we investigated the potential role of SEL skills and supports and developmental challenges that could be addressed by school personnel to promote rural students’ educational aspirations and potentially improve school grades. Finally, we examined risk factors that could undermine educational outcomes of these students.

METHODS

The data were from the Minnesota Student Survey (MSS), administered triennially and anonymously to students in grades five, eight, nine, and eleven statewide. Designed by the Minnesota Departments of Education, Health, Human Services, and Public Safety, the survey has been used to monitor trends in students’ behaviors and experiences regarding education, health, risky behaviors, beliefs, social and emotional skills, and more. Data from public school districts in the 2013, 2016, and 2019 administrations were used in this study. The MSS is voluntary, that is, districts, schools, parents, and students can choose to opt out. Nevertheless, in 2013, 84 percent of public school districts participated, 85 percent in 2016, and 81 percent in 2019 (Minnesota Department of Education 2020). Additional data were taken from the Census Bureau’s American Community Survey, including information about the communities in which school districts were located. The most recent data available were collected in 2018. The data, aggregated for Minnesota school districts, were obtained from the Metropolitan Council (2019), a Minnesota regional policymaking body and planning agency.

DEFINING RURAL AND URBAN SCHOOL DISTRICTS

A rural location indicator was created for school districts outside the Twin Cities metropolitan area and with community populations of less than fifty thousand. School districts within the metropolitan area were included as urban school districts. In addition, five school districts outside the metropolitan area had community populations larger than fifty thousand (from sixty-five thousand to one hundred twenty thousand, consistent with the Census Bureau’s definition of urbanized areas). We identified forty-eight urban districts (representing 63 percent of the state population) and 253 rural districts (37 percent of the state population). Rural districts had an average high school district enrollment (grades eight to twelve) of 478 students (SD = 466) and urban districts had an average of 3,889 (SD = 3,355) students.

ANALYSES

To appropriately describe the differences in school district communities located in rural or urban settings, we provided a series of statistical summaries on a number of characteristics, including student and community characteristics. In each case, we also reported a standardized effect size to estimate the magnitude of differences in these for rural and urban school districts. For continuous data, we estimate the standardized mean difference as the difference between the two means divided by the pooled standard deviation (Cohen’s d). For differences in proportions, assuming an underlying normal distribution, d_probit was used as an unbiased estimator of the population standardized proportion difference; this estimator behaved well under controlled simulation in comparison to six other estimators (Sánchez-Meca, Marín-Martínez, and Chacón-Moscoso 2003). The standardized proportion difference effect size (d_probit) was computed using the practical meta-analysis effect size calculator.²

We used hierarchical linear modeling (HLM) to take into account the nested structure of the data (Raudenbush and Bryk 2002), where students (level 1) were nested within school districts (level 2). The variation in grades and educational aspirations were less than 4 percent between school districts. Nevertheless, HLM models were used to correctly estimate the standard errors of the school district effects because districts were far fewer than students (this approach used the correct degrees of freedom to account for more than 150,000 students from 253 rural districts and 218,000 students from forty-eight urban districts when estimating student and districts effects). All analyses were conducted using HLM 8 (Raudenbush et al. 2019). The HLM equations were presented in appendix 1. We estimated both models for rural and urban districts separately to support clearer interpretation, but did test a combined model using the rural indicator in interaction terms to test the significance of the rural effect on each student-level characteristic in a single model, rather than a series of post hoc tests of differences in coefficients from the separate rural and urban models (appendix 2). We standardized (M=0, SD=1) the SEL and Bullied measures and all district-level continuous variables to support clearer interpretation.

RESULTS

We first examined the distribution of study variables by school district location (table 1). Summary statistics and effect sizes for grades and college goals are presented in table 2. Overall, students from rural districts had lower grades and were less likely to have college goals than students from urban districts; however, the effect size for college goals was more than three times (0.36) the effect size for school grades (0.10), showing that the college aspiration gap was bigger. This was similar to what Jessica Drescher and her colleagues (2022, this issue) find for school grades—not a large difference in standardized test scores between students from U.S. rural and nonrural schools.

Only 61 percent of students from rural districts aspired to go to a four-year college, versus 74 percent from urban districts. Even though the effect size was small according to Cohen’s interpretation (Cohen 1988), the numbers were substantial.³ If we consider these values as good estimates of the population, then given the population of the 301 districts with 307,000 students, nearly forty-seven thousand rural and forty-eight thousand urban students did not have college goals.

The correlation between school grades and college aspirations was 0.35 in urban districts and 0.38 in rural districts. Student-level measures are presented in table 3. On average, students reported scores above 10 for developmental skills and supports (students felt the characteristics were more like them) and scores below 10 for Bullied (students felt the characteristic was less like them).

Students from rural districts had lower scores in developmental skills (0.05 to 0.09 SD lower) and supports (0.04 to 0.07 SD lower) than students from urban districts, and higher Bullied scores (0.10 SD higher). We find small differences on each measure; perhaps most notably, rural students reported lower commitment to learning (0.09 SD) and higher levels of being bullied (0.10 SD).

Descriptive statistics for the district measures are presented in table 4. The proportion of the population with incomes below 185 percent the poverty threshold was higher in rural districts (0.25 compared to 0.18); and the proportion of adults with bachelor’s degree or higher in rural communities (0.21) was half the proportion of urban areas (0.42).

DISCUSSION AND PATHWAYS FORWARD

Rural communities tend to have troubling well-being indicators such as high poverty rates, low educational attainment, employment rates, and earnings relative to urban communities, yet researchers tend to focus on urban or national populations. In this study, we explored the educational experiences of rural youth. We used the positive youth development framework to highlight the social and emotional factors that are positively associated with youth outcomes as opposed to a deficit-oriented framework focused on risk factors. We also used an ecological approach and considered school and community supports and characteristics to more fully understand student educational outcomes, including school grades and college aspirations. This allowed us to explore some malleable factors that are available to school leaders and policymakers. We used a sample of nearly 152,000 rural and 218,000 urban secondary school students in 301 Minnesota school districts (253 rural and forty-eight urban) and used multilevel modeling to properly account for the nested structure of the data and differences in degrees of freedom for estimating the effects of student and district characteristics.

The first research question addresses the difference in school grades and educational aspirations between students from rural and urban districts. We find that students from rural district schools have lower grades than urban schools; however, this difference is small, about 0.1 on a 4-point scale (table 2). In contrast, for educational aspirations, the difference between students from rural and urban district schools is not negligible, effect size of 0.36 (table 2), even after controlling for students and school districts characteristics. This could be because cultural factors beyond school characteristics shape educational aspirations of rural students. For instance, rural students typically have to move away from home to go to college and face a conflict between staying in their community or continuing their education, which is not as common for students from urban areas, who often have more choices (Hektner 1995). As Nichole Bernsen and her colleagues (2022, this issue) find, rural students who have positive views of their community reported lower educational aspirations because they feel more attached to their communities. Likewise, social mobility in rural areas is associated with emotional and social costs that are different than in urban areas, as found by Ryan Parsons (2022, this issue). Further, traditional employment opportunities in rural areas such as agriculture and farming might not require college education, unlike those found in urban areas (Rojewski 1999). These factors might result in larger educational aspiration gaps between groups.

The results also show that male students, and those receiving special education services or participating in free or reduced-price lunch, have lower school grades and educational aspirations. This is consistent with what Matthew Irvin and colleagues (2011) find: students with disabilities tend to have lower grades and educational attainment than students with no disabilities in rural districts.

Our second question addresses the role of developmental skills, supports, and challenges. Having experienced trauma or being a victim of bullying can have detrimental effects on educational outcomes, for rural and urban students. Youth with higher levels of Commitment to Learning and Positive Identity and Outlook have higher grades and educational aspirations, and students with more Family and Community Support have higher educational aspirations for rural and urban students, suggesting that promoting SEL skills and supports can improve educational outcomes for all students.

Finally, our third question addresses the unique contexts of the two larger nonwhite communities in rural Minnesota. We find that Latino and American Indian students tend to have worse outcomes than their peers, and these disparities were larger for students attending districts with higher enrollments of Latino or American Indian students. These results are in rural and urban school districts, showing that these groups face more challenges in both settings. Likewise, the difference in educational aspirations between Latino students and their peers was larger in urban schools, suggesting that Latino students have less of a negative effect in rural schools.

Some limitations of the study include the fact that because the questionnaire was anonymous, we were not able to link students across years and use the MSS as a longitudinal dataset, or link survey responses to administrative data. And, as always from cross-sectional studies, we must be careful to not infer causation from these models. Last, the MSS is a self-reported anonymous survey; we therefore cannot corroborate the responses. Nevertheless, these results provide important insights about educational outcomes for underserved and more isolated groups in the United States, whereas school districts differ considerably between rural and urban locations on a number of community context characteristics. We find that student populations across settings do not differ much in developmental skills, supports, and challenges (table 3), including measures of SEL. It appears that schools in both urban and rural areas have students with relatively similar SEL skills, indicating students in both areas on average have some of the skills and supports needed to be successful in school. The conditioned associations between SEL measures and school grades and educational aspirations were remarkably similar for rural and urban school districts: SEL is relevant in both contexts.

We recognize that promoting college aspirations in rural settings is particularly challenging because, as mentioned, college attendance often involves students moving away from their rural communities and families, which may be less of a concern in urban settings (Hektner 1995; Elder 1996). College aspirations in rural communities are also challenging given the higher rates of poverty and lower family incomes. However, promoting SEL skills and supports is one promising approach available to schools and communities that could be addressed by both rural and urban schools. This should be of particular interest to rural school districts. Last, these results illuminate the additional challenges that communities of color face, particularly Latino and American Indian youth who have lower school grades and educational aspirations in urban and rural school districts. We ascribe to the principles of targeted universalism; universal goals may be achieved by targeting services and supports to those facing unique contexts, such as Latino and American Indian youth. Indeed, addressing unique community contexts is the pathway to greater educational equity.