Your Future Depends On Your Aspirations: Predictor Of Educational Attainment Outcome

Introduction

Post-secondary education is becoming more imperative to the success of young people across the United States. As a result, organizations like the Lumina Foundation are committing to increase the proportion of Americans with a post-secondary credential to 60% by 2025 (Lumina Foundation). Schools are focused on adjusting curriculum and providing resources to students and families to make education after high school attainable.

However, all of this work begins with perceptions of what is possible. Students, parents, and teachers all have perceptions about the educational attainment level of the students and these can be powerful indicators for educational attainment outcomes for those students. A study by Boser, Wilhelm, & Hanna (2014) showed that high school teachers “have lower expectations for students of color and students from disadvantaged backgrounds” (p. 2). Our study seeks to disaggregate these findings from Boser, Wilhelm, & Hanna (2014) by students who identify as White, Asian/Hawaiian/Pacific Islander, Hispanic, and Black using data from the Educational Longitudinal Study of 2002 (ELS). Investigating possible relationships between perceptions of educational attainment and attainment outcomes, especially in relation to race, has the possibility to provide up-to-date training for teacher education and educational leadership programs.

Background

Race

As the White population is outpaced by other racial populations and older White workers get ready to retire, the economy will become more reliant on Asian, Black and Latinx populations (Schak & Nichols, 2017). Although states have developed goals to increase the number of people obtaining a college credential, there remains an educational attainment gap for Black and Latinx students in the United States (Schak & Nichols, 2017). In 2016, the racial composition of those who completed at least a bachelor’s degree was 54 percent Asian, 35 percent White, 21 percent Black, and 15 percent Hispanic (The National Center for Education Statistics, 2016). Previous research has shown that educational perceptions vary by racial/ethnic groups (Lowman & Elliott, 2009). It is important to understand how students, parents, and teachers’ educational expectations vary by race to better provide support and resources.

Perceptions

Equally as critical to educational attainment outcomes are the expectations or, perceptions, that adults have of the students. Spera, Wentzel, & Matto, (2009) referenced research from Bronstein et al. (2005) and Wigfield (1993) when discussing the relationship between parents’ aspirations and student academic outcomes, “researchers have found that parental aspirations for their children’s educational attainment are significantly and positively related to their children’s setting of academic goals, persistence in school, course enrollment, intellectual accomplishments, and attendance of college” (p. 1140). One of the key factors that influence parents’ perceptions is their education level. Interestingly enough, there appear to be differences among racial groups for parents’ aspirations for their students’ educational attainment. Spera, Wentzel, & Matto (2009) again highlight Driessen et al. (2005) and Stevenson et al. (1990) when stating, “researchers have found that African American and Hispanic parents place a high value on education, are concerned with educational issues, and have aspirations for their children that equal those of non-minority parents” (p. 1141). Hitlin (2006) points out how there are also differences in how students internalize these aspirations in the form of values. Whitbeck & Gecas (1988) mention that “parents are more likely to attribute values similar to their own to daughters than to sons” (Hitlin, 2006, p. 28). One could make some assumptions about how those internalized values play out in the form of educational attainment outcomes. Lastly, parents’ expectations also serve to “buffer low teacher expectations” which Kirk, Lewis-Moss, Nilsen, & Colvin (2011) credit Benner and Mistry (2007) in their study (p. 89). To properly investigate the relationship between perceptions and expectations of educational attainment for students, the role parents play into these ideas cannot be undervalued.

In a similar breath, teachers’ expectations of students’ educational attainment are just as powerful influencing student’s perceptions of what they are capable of. Gershenson, Holt, & Papageorge (2016) called attention to a study conducted by Gregory & Huang, (2013) detailing that ‘teachers also likely affect students’ beliefs by directly imparting their expectations to students. For example, protection models hypothesize that teacher expectations ‘protect against,’ or counteract, negative expectations created by neighborhood effects or lack of access to educationally-successful role models” (Gershenson, Holt & Papageorge, 2016, p. 211). This works similarly to how parent expectations can buffer against low teacher expectations for students. Gershenson, Holt & Papageorge (2016) go further on to say, “we find that non-black teachers have significantly lower educational expectations for black students than black teachers do...our results provide insights into the mechanisms through which student-teacher demographic mismatch affects academic achievement and provide novel causal evidence that demographic mismatch affects teachers’ expectations for students’ long-run educational attainment” (p. 211). This continues to emphasize the point that there might be differences in expectations for students of different racial backgrounds that need to be considered when constructing these models.

Research Questions and Hypotheses

This research seeks to further analyze the study conducted by the Center for American Progress (CAP), The Power of the Pygmalion Effect Teachers Expectations Strongly Predict College Completion (Boser, Wilhelm, and Hanna, 2014). The research conducted by CAP analyzes the educational attainment expectations by students, teachers, and parents. We will be disaggregating the educational perceptions and outcomes by race and controlling for other variables.

This research seeks to understand 1. The educational attainment expectations differences between 10th grader students, teachers, and parents and 2. How does the educational attainment expectations contrast between White, Asian, Black, and Latinx students? To analyze these two questions, the following questions, and hypothesis will be answered:

Are there differences between educational attainment between 10th grader students, teachers, and parents?

Hypothesis: Differences exist between the educational attainment between the expectations between students, teachers, and parents.

Which group has a better prediction of students’ educational attainment?

Hypothesis: Teachers’ expectations will provide a better prediction of students’ educational attainment.

Do expectations differ between White, Asian, Black, and Latinx students? If so, how?

Hypothesis: There are differences in the expectations between White, Asian, Black and Latinx students. We predict that there are differences in expectations of White students and Asian students and that these will be higher than the expectations of Black and Latinx students.

Data

This research paper utilized data from the Education Longitudinal Study (ESL) from 2002, sponsored by the National Center for Education Statistics. ESL is a part of three other longitudinal studies conducted in 1972, 1980, and 1988. The study is a multilevel survey that includes several respondents including 10th-grade students and their teachers, parents, counselors, and principals (Lauff & Ingels, 2015). The study followed the 10th graders through a twelve-year span in 2002, 2006, 2012, and 2014. The study was also administered to one of the students’ parents to understand the aspirations they have for their children, home background, and educational history of the child. Among those who participated include 752 schools, 15,362 students, 13, 488 teachers, and 7,135 teachers. The data was collected through a two-stage stratified probability sample. Schools were first selected and then students were randomly selected within each school (Lauff & Ingels, 2015). The schools were stratified by U.S. Census division and locations.

Other data was collected through the American Council on Education, the U.S. Department of Education Central Processing System, SAT/ACT, and transcripts (Lauff & Ingels, 2014). The survey collected a variety of information including students’ backgrounds, out of school experiences, language, educational aspirations, technology, school climate, and career plans. The data represents a nationally representative sampling of students in the United States, Asian, Hispanic, and catholic schools were oversampled to account for the smaller data, therefore, statistically sound inferences can be made. Transcript data from respondents who did not complete high school is not included since the report utilizes post-high school information.

For our research, we utilized the first round of assessment conducted in 2002 that asked students “how far student thinks he/she will get in school”, teachers “how far teacher expects student to get”, and to parents “how far parent expects student to get”. We also utilized the parents level of education, parents’ income, students’ senior GPA, gender, and motivation. We disaggregated the data race by White, Asian, Hispanic, and Black. Lastly, we will look at the educational attainment that was collected in 2012 to measure the actual outcomes of students.

Measures

Dependent Variables

All variables in this study we coded missing, survey component legitimate skip, nonrespondent were coded as “.” for missing data. In order to examine how perceptions of educational attainment of students affect the actual outcomes, we have selected two dependent variables: Highest level of education earned as of the third follow-up (F3ATTAINMENT) recorded as an ordinal level measurement and if they earned a post-secondary credential (GRADCRED) as dichotomous level measurement. Informed by the Lumina Foundation’s focus on post-secondary credentials, the second dependent variable was created with undergraduate credential and above coded as a “1” for “yes” and everything below coded as a “0” for “no”.

Independent Variables

Race: As displayed in our review of the literature, race is an essential variable to consider. As a nominal level of measurement (byrace), we have to make some discussions around dropping certain groups and collapsing one in particular. We decided to drop American Indian/Alaska Native, non-Hispanic and More than one race, non-Hispanic because their totals were less than 9% of the total sample. In addition, we collapsed the two classifications for Hispanics into one group bringing the total to over 13%: Hispanic, no race specified & Hispanics, race specified.

Perceptions/Expectations: Additionally, this study relies on the variables centered on the perceptions or expectations of student’s educational attainment. All three of the variables selected [How far in school student thinks will get (bystexp), How far in school parents want 10th grader to go (byparasp), How far teacher expects student to get in school (BYTE20)] are ordinal levels of measurement that range from Don’t know to Obtain PhD, MD, other advanced degree.

Motivation: Another critical factor to consider is the motivation of the student, specifically motivation from external sources. Thinking about parents’ and teachers’ expectations for student educational attainment and how impactful they are on the student, they serve as motivation that might contribute to educational outcomes for the student. Instrumental motivation (utility interest) scale (byinstmo) is an interval level measurement that ranges from -1.994 to 1.579.

Parent level of education: The literature suggested that level education for the parent helps to inform the expectations parents have for their students’ educational attainment. We included the variables Mother’s highest level of education (bymothed) and Father’s highest level of education (byfathed). The variables are also ordinal levels of measurement with both ranging from Did not finish high school to Completed PhD, MD, other advanced degrees.

Family Income: Equally as useful to the educational level of the parents is the household income of the family of the students. The variable Total family income from all sources 2001 (byincome) is an interval level measurement that can operate as a ratio level measurement that ranges from $0 to $200,001 or more.

Gender: As discussed in the literature review, gender might help to explain some variation in outcomes for educational attainment due to how parents’ expectations are internalized by girls. Sex (bysex) is a dichotomous level measurement where “1” is coded as “Male” and “0” is coded as “Female”.

GPA: Similar to motivation, GPA could be influential for educational attainment outcomes in the sample. GPA taken for all courses in 9th - 12th grade (F1RGPP2) is also an interval level measurement that can operate as a ratio level measurement that ranges from 0.00 to 4.00.