Loading wiki pages...

Wiki Version:
<p>Problem: </p> <p>In order to make the college completion agenda a reality, the United States needs to enroll more students, especially low-income students, into higher education, and, more importantly, ensure that they receive the support they need to remain enrolled and graduate.</p> <p>See "Rationale" document for more detail description of the college access and completion. </p> <p>Background: </p> <p>College Forward is a non-profit college access and success organization. College Forward’s mission is to coach motivated, underserved students to achieve the benefits of higher education and a college degree. Today, College Forward serves more than 4,300 students in Central and Gulf Coast Texas. Please see College Forward's website to learn more about their programs (<a href="http://collegeforward.org" rel="nofollow">collegeforward.org</a>).</p> <p>Hypotheses:</p> <ul> <li>Participation in College Forward will lead to increased postsecondary completion relative to students who do not participate in the program.</li> <li>Participation in College Forward (CoFo) will have differential effects based on a variety of student characteristics: </li> <li><strong><em>First-generation</em></strong> students, who are the first in their family to attend college, will benefit from CoFo more than continuing generation students. </li> <li>Students from <strong><em>lower-income</em></strong> family backgrounds will benefit more from CoFo than students from moderate-income backgrounds. </li> <li>Students who do not actively <strong><em>seek help</em></strong> will show greater gains in college completion as a result of CoFo than students who proactively seek help.</li> <li>Students with less <strong><em>grit</em></strong> will benefit more from CoFo than grittier students. </li> <li>Students who experience <strong><em>present bias</em></strong> (or procrastination) will experience higher college graduation rates as a result of CoFo than students who have no trouble with procrastination. </li> <li>Students who are likely to experience <strong><em>information overload</em></strong> will show greater effects of the CoFo program than students who do not expereicne information overload. </li> </ul> <p>Sample:</p> <ul> <li>High school juniors from multiple locations from College Forward's applicant pool in the fall of 2015 and the fall of 2016 will be randomly assigned either to receive the offer of participation in the program or to a control group that does not receive College Forward services. </li> <li>The full time span of this study is 9 years, covering two student cohorts, beginning when students enroll in College Forward’s program during their junior year of high school and continuing for up to 6 years in postsecondary education.</li> </ul> <p>Outcomes: </p> <ul> <li>FAFSA completion and renewal;</li> <li>College enrollment and persistence;</li> <li>College credit accumulation per semester;</li> <li>6-year bachelor's degree; associate's degree, or certificate completion;</li> <li>Workforce earnings.</li> </ul> <p>Method of Analysis: </p> <ul> <li>Ordinary Least Squares (OLS) regression will be used to evaluate the impact of College Forward on students’ postsecondary outcomes.</li> <li>Our core analytic model will be specified as follows:</li> </ul> <p>π‘Œπ‘–π‘—π‘‘ = 𝛽0 + 𝛽1𝑇𝑅𝐸𝐴𝑇𝑀𝐸𝑁𝑇𝑖𝑗𝑑 + 𝛾𝑋𝑖𝑗𝑑 + 𝛿𝑗𝑑 + πœ€π‘–π‘—π‘‘ - where π‘Œπ‘–π‘—π‘‘ is a vector of different outcomes we will incorporate in our analysis, such as whether students earn a degree within six years of high school or students’ earnings 6 – 8 years after high school, for student <em>i</em> in high school <em>j</em> in cohort <em>t</em>; - 𝑇𝑅𝐸𝐴𝑇𝑀𝐸𝑁𝑇𝑖𝑗𝑑 is a binary indicator for whether students have been randomly assigned to participate in the College Forward advising program or to the control group; - 𝑋𝑖𝑗𝑑 is a vector of student-level demographic, socioeconomic, and academic baseline covariates we will collect from the College Forward application, for both treatment and control students; - 𝛿𝑗𝑑 is a set of fixed effects for high school*cohort, to account for the level at which we will conduct the randomization (within cohort, within school, at the student level); - and πœ€π‘–π‘—π‘‘ is a residual error term. </p> <p>In this model 𝛽1 will provide an unbiased causal estimate of the impact of the offer to participate in College Forward on students’ college outcomes. </p> <p>We will also include statistical interactions to investigate whether assignment to College Forward had different effects for particular sub-groups of students.</p> <p>See "Study Design" document for additional information. </p>
This website relies on cookies to help provide a better user experience. By clicking Accept or continuing to use the site, you agree. For more information, see our Privacy Policy and information on cookie use.

Start managing your projects on the OSF today.

Free and easy to use, the Open Science Framework supports the entire research lifecycle: planning, execution, reporting, archiving, and discovery.

Create an Account Learn More Hide this message