Helping Students Navigate the Path to College: What High Schools Can Do

IES PRACTICE GUIDE WHAT WORKS CLEARINGHOUSE

NCEE 2009-4066 U.S. DEPARTMENT OF EDUCATION The Institute of Education Sciences (IES) publishes practice guides in education to bring the best available evidence and expertise to bear on the types of challenges that cannot currently be addressed by single interventions or programs. Authors of practice guides seldom conduct the types of systematic literature searches that are the backbone of a meta-analysis, although they take advantage of such work when it is already published. Instead, authors use their expertise to identify the most important research with respect to their recommendations and conduct a search of recent publications to ensure that the research supporting the recommendations is up-to-date.

Unique to IES-sponsored practice guides is that they are subjected to rigorous external peer review through the same office that is responsible for independent review of other IES publications. A critical task for peer reviewers of a practice guide is to determine whether the evidence cited in support of particular recommendations is up-to-date and that studies of similar or better quality that point in a different direction have not been ignored. Because practice guides depend on the expertise of their authors and their group decisionmaking, the content of a practice guide is not and should not be viewed as a set of recommendations that in every case depends on and flows inevitably from scientific research.

The goal of this practice guide is to formulate specific and coherent evidence-based recommendations for use by educators addressing the challenge of increasing access to higher education. The guide provides practical, clear information on critical topics related to what schools can do to help students navigate the path to college and is based on the best available evidence as judged by the panel. Recommendations presented in this guide should not be construed to imply that no further research is warranted on the effectiveness of particular strategies for increasing access to postsecondary education. IES PRACTICE GUIDE

Helping Students Navigate the Path to College: What High Schools Can Do

September 2009

Panel William G. Tierney (Chair) University of Southern California

Thomas Bailey Columbia University

Jill Constantine Mathematica Policy Research

Neal Finkelstein WestEd

Nicole Farmer Hurd National College Advising Corps

Staff Jeffrey Max Christina Clark Tuttle Mathematica Policy Research

NCEE 2009-4066 U.S. DEPARTMENT OF EDUCATION This report was prepared for the National Center for Education Evaluation and Re- gional Assistance, Institute of Education Sciences under Contract ED-07-CO-0062 by the What Works Clearinghouse, a project of Mathematica Policy Research.

Disclaimer The opinions and positions expressed in this practice guide are those of the authors and do not necessarily represent the opinions and positions of the Institute of Education Sciences or the U.S. Department of Education. This practice guide should be reviewed and applied according to the specific needs of the educators and education agencies using it, and with full realization that it represents the judgments of the review panel regarding what constitutes sensible practice, based on the research that was available at the time of publication. This practice guide should be used as a tool to assist in decisionmaking rather than as a “cookbook.” Any references within the document to specific education products are illustrative and do not imply endorsement of these products to the exclusion of other products that are not referenced.

U.S. Department of Education Arne Duncan Secretary

Institute of Education Sciences John Q. Easton Director

National Center for Education Evaluation and Regional Assistance John Q. Easton Acting Commissioner

September 2009

This report is in the public domain. Although permission to reprint this publication is not necessary, the citation should be: Tierney, W. G., Bailey, T., Constantine, J., Finkelstein, N., & Hurd, N. F. (2009). Helping students navigate the path to college: What high schools can do: A practice guide (NCEE #2009-4066). Washington, DC: National Center for Education Evaluation and Regional Assistance, Institute of Education Sciences, U.S. Depart- ment of Education. Retrieved from http://ies.ed.gov/ncee/wwc/publications/ practiceguides/.

What Works Clearinghouse Practice Guide citations begin with the panel chair, followed by the names of the panelists listed in alphabetical order.

This report is available on the IES website at http://ies.ed.gov/ncee and http://ies. ed.gov/ncee/wwc/publications/practiceguides/.

Alternative Formats On request, this publication can be made available in alternate formats, such as Braille, large print, audiotape, or computer diskette. For more information, call the Alternative Format Center at (202) 205-8113. Helping Students Navigate the Path to College: What High Schools Can Do Contents

Introduction 1 The What Works Clearinghouse standards and their relevance to this guide 4

Overview 5

Scope of the practice guide 7

Status of the research 8

Summary of the recommendations 9

Checklist for carrying out the recommendations 11

Recommendation 1. Offer courses and curricula that prepare students for college-level work, and ensure that students understand what constitutes a college-ready curriculum by 9th grade 12

Recommendation 2. Utilize assessment measures throughout high school so that students are aware of how prepared they are for college, and assist them in overcoming deficiencies as they are identified 20

Recommendation 3. Surround students with adults and peers who build and support their college-going aspirations 26

Recommendation 4. Engage and assist students in completing critical steps for college entry 31

Recommendation 5. Increase families’ financial awareness, and help students apply for financial aid 38

Appendix A. Postscript from the Institute of Education Sciences 43

Appendix B. About the authors 46

Appendix C. Disclosure of potential conflicts of interest 48

Appendix D. Technical information on the studies 49

References 73

( iii ) HELPING STUDENTS NAVIGATE THE PATH TO COLLEGE: WHAT HIGH SCHOOLS CAN DO

List of tables

Table 1. Institute of Education Sciences levels of evidence for practice guides 2

Table 2. Recommendations and corresponding levels of evidence 6

Table 3. Examples of college preparatory course requirements 14

Table D1. Studies of college access programs that met WWC standards with or without reservations 54

Table D2. Studies of college access programs that potentially met WWC standards 56

List of exhibits

Exhibit 1. Example of course requirement mailing 16

Exhibit 2. Example of a personalized learning plan 17

Exhibit 3. Example of a college entrance exam schedule 33

Exhibit 4. Example of a college visit schedule 34

Exhibit 5. Example of a college admissions timeline 36

( iv ) Introduction document ranges from experimental evaluations of college access programs to ex- Access to higher education remains a chal- pert analyses of college access practices. lenge for many students who face barriers In looking for effective practices, the panel to college entry. Low-income students and paid particular attention to high-qual- students who are potentially the first in ity experimental and quasi-experimental their family to attend college have lower studies, such as those meeting the criteria college enrollment rates than other stu- of the WWC,4 and to patterns of practices dents.1 Although academic preparation ac- that are replicated across programs. counts for some of these differences, the disparities in college-going rates persist The research base for this guide was iden- for these groups of students even when tified through a comprehensive search for controlling for academic preparation.2 studies evaluating college access inter- College access outcomes have important ventions and practices. An initial search economic and social consequences: col- for this type of research conducted in the lege graduates earn more than those with United States in the past 20 years (1988– a high school degree and are more active 2008) yielded more than 500 studies. Of in their communities.3 these, 99 studies examined college access programs or related practices for high This guide is intended to help schools and school students and were eligible for fur- districts develop practices to increase ac- ther review because the study design in- cess to higher education. It can be use- cluded a comparison group. These studies ful for individuals who work in schools were reviewed by the WWC to determine and districts in planning and executing whether they were consistent with WWC strategies to improve preparation for, and standards. Of the 99 studies, 16 studies access to, higher education. A panel of met WWC standards with or without res- experts in college access programs and ervations. These 16 studies of 10 differ- strategies and in research methods devel- ent college access programs represent the oped the recommendations in this guide. strongest evidence of the effectiveness of The guide contains specific steps on how college access programs. to implement the recommendations that are targeted at school- and district-level To indicate the strength of evidence sup- administrators, teachers, counselors, and porting each recommendation, the panel related education staff. The guide also relied on the WWC standards for determin- indicates the level of research evidence ing levels of evidence, described below and demonstrating that each recommended in Table 1. It is important for the reader to practice is effective. remember that the level of evidence rating is not a judgment by the panel on how ef- As with all What Works Clearinghouse fective each of these recommended prac- (WWC) practice guide panels, this panel tices will be when implemented, nor are developed recommendations by consult- they a judgment of what prior research ing research evidence. The evidence that has to say about their effectiveness. The the panel considered in developing this level of evidence ratings reflect the panel’s judgment of the quality of the existing literature to support a causal claim that 1. Choy (2002); National Center for Education Statistics (2008). when these practices have been imple- 2. Ellwood and Kane (2000); Smith et al. (1997). mented in the past, positive effects on student academic outcomes were observed. 3. Baum and Ma (2007); Kane and Rouse (1995); National Conference on Citizenship (2006); U.S. Census Bureau (2002). 4. http://www.whatworks.ed.gov/

( 1 ) Introduction

Table 1. Institute of Education Sciences levels of evidence for practice guides

In general, characterization of the evidence for a recommendation as strong requires both studies with high internal validity (i.e., studies whose designs can support causal conclusions) and studies with high external validity (i.e., studies that in total include enough of the range of participants and settings on which the recommendation is focused to support the conclu- sion that the results can be generalized to those participants and settings). Strong evidence for this practice guide is operationalized as: • A systematic review of research that generally meets WWC standards (see http://ies. ed.gov/ncee/wwc/) and supports the effectiveness of a program, practice, or approach Strong with no contradictory evidence of similar quality; OR • Several well-designed, randomized controlled trials or well-designed quasi-experiments that generally meet WWC standards and support the effectiveness of a program, practice, or approach, with no contradictory evidence of similar quality; OR • One large, well-designed, randomized controlled, multisite trial that meets WWC standards and supports the effectiveness of a program, practice, or approach, with no contradictory evidence of similar quality; OR • For assessments, evidence of reliability and validity that meets the Standards for Educational and Psychological Testing.a

In general, characterization of the evidence for a recommendation as moderate requires studies with high internal validity but moderate external validity or studies with high external validity but moderate internal validity. In other words, moderate evidence is derived from studies that support strong causal conclusions but generalization is uncertain or studies that support the generality of a relationship but the causality is uncertain. Moderate evidence for this practice guide is operationalized as: • Experiments or quasi-experiments generally meeting WWC standards and supporting the effectiveness of a program, practice, or approach with small sample sizes and/ or other conditions of implementation or analysis that limit generalizability and no contrary evidence; OR Moderate • Comparison group studies that do not demonstrate equivalence of groups at pretest and, therefore, do not meet WWC standards but that (1) consistently show enhanced outcomes for participants experiencing a particular program, practice, or approach and (2) have no major flaws related to internal validity other than lack of demonstrated equivalence at pretest (e.g., only one teacher or one class per condition, unequal amounts of instructional time, or highly biased outcome measures); OR • Correlational research with strong statistical controls for selection bias and for dis- cerning influence of endogenous factors and no contrary evidence; OR • For assessments, evidence of reliability that meets the Standards for Educational and Psychological Testingb but with evidence of validity from samples not adequately representative of the population on which the recommendation is focused.

In general, characterization of the evidence for a recommendation as low means that the recommendation is based on expert opinion derived from strong findings or theories in Low related areas and/or expert opinion buttressed by direct evidence that does not rise to the moderate or strong level. Low evidence is operationalized as evidence not meeting the standards for the moderate or strong level. a. American Educational Research Association, American Psychological Association, and National Council on Measurement in Education (1999). b. Ibid.

( 2 ) Introduction

They do not reflect judgments of the rela- recommended practices would be difficult tive strength of these positive effects or to study in a rigorous, experimental fash- the relative importance of the individual ion; in other cases, it means that research- recommendations. ers have not yet studied this practice, or that there is weak or conflicting evidence A strong rating refers to consistent and of effectiveness.6 generalizable evidence that an intervention strategy or program improves Three of the five recommendations made outcomes.5 by the panel received a low evidence rating. For example, recommendation 2, A moderate rating refers either to evidence which describes the use of assessments from studies that allow strong causal con- to measure college readiness, was deter- clusions but cannot be generalized with mined to have a low level of evidence (see assurance to the population on which a Table 2). This means that there are few ex- recommendation is focused (perhaps be- isting studies designed to test, in a discrete cause the findings have not been widely and valid manner, the causal relation be- replicated) or to evidence from studies that tween the utilization of assessment mea- are generalizable but have more causal sures and college going. Nevertheless, the ambiguity than that offered by experi- authors of this practice guide, based on mental designs (e.g., statistical models of expert judgment and knowledge of prac- correlational data or group comparison de- tice, consider the use of assessment to be a signs for which equivalence of the groups critical component of a well-implemented at pretest is uncertain). strategic plan for increasing access to college. Hence, although the level of evidence A low rating refers to evidence from stud- rating is low, the panel has included as- ies that do not meet the standards for sessment as one of the five recommended moderate or strong evidence and/or expert practices. opinion based on reasonable extrapola- tions from research and theory. Citations in the text refer to studies of programs that have implemented vari- A low level of evidence rating does not ous practices. Not all of these programs indicate that the recommendation is any contribute to the level of evidence rating: less important than other recommenda- although some of these programs have tions with a strong or moderate rating. had rigorous evaluations of their impacts, Rather, it suggests that the panel cannot others have not. Furthermore, some of the point to a body of research that demon- programs that have been rigorously evalu- strates its effect on student achievement. ated have found positive effects on college In some cases, this simply means that the access outcomes; others have not.

5. Following WWC guidelines, improved outcomes are indicated by either a positive, statistically significant effect or a positive, substantively 6. For more information, see the WWC Frequently important effect size (i.e., greater than 0.25). See Asked Questions page for practice guides, http:// the WWC guidelines at http://ies.ed.gov/ncee/ ies.ed.gov/ncee/wwc/references/idocviewer/ wwc/pdf/wwc_version1_standards.pdf. doc.aspx?docid=15&tocid=3.

( 3 ) Introduction

The What Works Clearinghouse • Potentially Meets Standards for studies standards and their relevance to that require additional information to this guide determine whether they meet evidence standards; typically refers to quasi-ex- In terms of the levels of evidence indi- perimental studies that do not provide cated in Table 1, the panel relied on WWC sufficient information to assess base- evidence standards to assess the quality line equivalence. of evidence supporting educational programs and practices. The WWC addresses Following the recommendations and sug- evidence for the causal validity of instruc- gestions for carrying out the recommen- tional programs and practices according to dations, Appendix D presents more in- WWC standards. Information about these formation on the research evidence that standards is available at http://ies.ed.gov/ supports each recommendation. ncee/wwc/references/idocviewer/doc. aspx?docid=19&tocid=1.7 The technical We appreciate the efforts of Jeffrey Max, quality of each study is rated and placed Christina Clark Tuttle, Kristin Hallgren, into one of four categories: Moira McCullough, and Sarah Wissel, Math- ematica Policy Research staff members • Meets Evidence Standards for random- who participated in the panel meetings, ized controlled trials and regression characterized the research findings, and discontinuity studies that provide the drafted the guide. We also appreciate the strongest evidence of causal validity. help of the many WWC reviewers who contributed their time and expertise to • Meets Evidence Standards with Res- the review process. We also thank Scott ervations for all quasi-experimental Cody, Shannon Monahan, and Neil Seftor studies with no design flaws and ran- for helpful feedback and reviews of earlier domized controlled trials that have versions of this guide. problems with randomization, attri- tion, or disruption. William G. Tierney Thomas Bailey • Does Not Meet Evidence Standards for Jill Constantine studies that do not provide strong evi- Neal Finkelstein dence of causal validity. Nicole Farmer Hurd

7. Reviews of studies for this practice guide applied version 1.0 WWC standards. Interested readers can access these standards at http://ies. ed.gov/ncee/wwc/references/iDocViewer/Doc. aspx?docId=20&tocId=1.

( 4 ) Helping Students grade, or earlier, and that they stay on that Navigate the Path path throughout high school. to College: What Second, many students do not take the nec- High Schools Can Do essary steps during high school to prepare for and enter college because they are not Overview aware of these steps or because they lack the guidance or support needed to com- A well-educated workforce is critical for main- plete them.11 In addition to the academic taining the economic competitiveness of the obstacles discussed earlier, students need United States. The strength of our economy to complete a number of discrete steps in hinges on the ability of our education sys- high school to enroll in college, such as tak- tem to meet the demand for highly educated ing college entrance exams, searching for workers. As a result, there are persistent calls colleges, applying for financial aid, submit- to improve access to higher education and to ting college applications, and selecting a encourage students and adults to continue college. In their senior year, students have their education beyond high school.8 How- to decide where to go; how to apply; and, ever, reaching college remains a challenge for most important, how to pay for college. many low-income and potentially first-gener- These issues should be considered, opti- ation students who (a) are not academically mally, in the earlier years of high school, prepared or (b) lack knowledge about how to but in the senior year students must make apply to, and pay for, college.9 College enroll- decisions. Students may lack adequate ad- ment rates for these students continue to lag vice, particularly if no one in their immedi- behind those of their peers despite overall ate families has completed a two- or four- improvements in college attendance. year degree. Students and their families need advice from knowledgeable school The challenge of improving the college-go- staff if they are to successfully navigate the ing rate can be traced to two key difficulties. college application processes. First, students must be academically prepared for college by 12th grade. The oppor- As a result, a large part of the obligation for tunities to academically prepare for college enabling students to gain the academic, so- narrow as students progress through high cial, and cultural skills to gain entrance to school. If students do not start taking college college falls upon our teachers, counselors, preparation courses in the 9th grade, they and school administrators. High schools will be less likely to enroll in college. In addi- play a critical role in preparing students tion, students who are not reading or doing academically for college and assisting stu- math at grade level will not be prepared for dents through the steps to college entry. college-level work.10 The problem is made They also can take steps to influence stu- more difficult if students and their families dents’ access to college-going peer groups are unaware that their performance is inad- and to encourage high academic expecta- equate. Schools need to ensure that students tions of students. The college-going culture are on the path to college beginning in 9th of a high school, or lack thereof, becomes important in college-going decisions. When students, teachers, and administrators 8. Pathways to College Network (2004). openly talk about preparing for and going 9. Avery and Kane (2004); Ikenberry and Hartle to college, the climate in the school can (1998); National Center for Education Statistics (2005); Roderick et al. (2008); U.S. General Ac- move toward college access. counting Office (1990). 10. Adelman (1999); Cabrera and La Nasa (2001); 11. Ikenberry and Hartle (1998); U.S. General Ac- Wimberly and Noeth (2005). counting Office (1990).

( 5 ) Overview

Table 2. Recommendations and corresponding levels of evidence

Recommendation Level of evidence

Academic preparation

1. Offer courses and curricula that prepare students for college-level work, and ensure that students understand what constitutes a college-ready Low curriculum by 9th grade

2. Utilize assessment measures throughout high school so that students are aware of how prepared they are for college, and assist them in Low overcoming deficiencies as they are identified

College aspirations and expectations

3. Surround students with adults and peers who build and support their Low college-going aspirations

Steps for college entry

4. Engage and assist students in completing critical steps for college entry Moderate

5. Increase families’ financial awareness, and help students apply for Moderate financial aid

Source: Authors’ compilation based on analysis described in text.

( 6 ) Scope of the impact college persistence is described practice guide when relevant. The panel believes that every student The purpose of this guide is to recommend should leave high school with the skills steps that educators, administrators, and required to attend a two- or four-year in- policymakers can take, beginning in the stitution. To that end, high schools must 9th grade, to increase access to higher edu- provide students with information to con- cation. The guide targets high schools and sider postsecondary training and to assess school districts and focuses on effective their readiness. Also, all high school cur- practices that prepare students academi- ricula should, at a minimum, prepare stu- cally for college, assist them in completing dents to begin taking college-level courses the steps to college entry, and improve their at a two-year institution without the need likelihood of enrolling in college. The rec- for remediation in any subject area upon ommendations address the discrete steps entering that institution. For students that students need to take throughout high who wish to enroll in a four-year institu- school and describe how high schools can tion, their high school curriculum should use mentors and peers to support students’ include options that prepare students for college aspirations. The panel recognizes the more rigorous academic requirements that simply providing students with infor- of four-year institutions. However, we do mation is insufficient, and, throughout, the not believe that all students should be guide recommends that high schools offer required to complete a high school cur- hands-on assistance and guidance in pre- riculum that prepares students for a four paring students for college. year college. Principal and district administrators should work to ensure that at all The recommended steps derive from the high schools, curricula alternatives exist characteristics of college access programs, appropriate for students who aspire to any school reforms, and policy interventions level of postsecondary training. that have shown promise in increasing access to college, particularly for low-in- We believe every high school should have come and first-generation students. The a college access strategy for students that panel focused on programs and practices incorporates our recommendations. Such a with evidence of their impact on academic strategy would address four areas: curricu- preparation for college (e.g., high school lum, assessment, aspirations, and hands- completion and course taking), completion on assistance with college entry activities. of the steps for college entry (e.g., submit- Some recommendations will include best ting college and financial aid applications), practices for all students; for example, or college enrollment and attendance. Al- recommendation 2 talks about identifying though the panel recognizes the impor- assessments of college readiness and mak- tance of college persistence for low-income ing students aware of their proficiency on and first-generation students who are less these assessments, and recommendation likely than other students to complete a 3 discusses building college aspirations. degree,12 the focus of the guide and the These recommendations are useful for stu- recommended practices is on how high dents still exploring their interest in attend- schools and districts can improve access ing college as well as providing feedback to higher education. However, evidence to students who have already determined on whether the recommended practices that they are interested in attending college. Other recommendations will be targeted at 12. National Center for Education Statistics students who have decided they want to (2004); Nunez and Cuccaro-Alamin (1998). attend a two- or four-year institution after

( 7 ) scope of the practice guide

high school. For example, recommendation four-year college are used to distinguish 4 focus on assisting students in their search between these two types of institutions. for identifying specific colleges, taking college entrance exams, and completing col- Status of the research lege applications. The recommendations reinforce that high schools need to be pre- Overall, the panel believes that the exist- pared to inform and support students to ing research on college access services and obtain their highest aspirations. programs is not at a level to provide con- clusive evidence of best practices. Studies Although this guide does not directly ad- of promoting college access generally look dress steps that college and universities can at specific programs that provide a bundle take to promote access, it does highlight the of services, and not at individual services, important role these institutions should making it difficult to isolate a specific ser- play and may be of benefit to higher educa- vice’s contribution to college readiness tion administrators. Much of the evidence and enrollment. In addition, the panel en- in the practice guide is based on programs countered varying impacts across college that were implemented by or in partnership access programs with ostensibly similar with postsecondary institutions. That said, services. The reasons for the varying im- the panel notes that other entities influence pacts are difficult to determine. In some access to college and may benefit from the cases, the programs are serving differ- recommendations in this guide as well. For ent populations of students. For example, example, parents and families have an es- some may target students who already sential role to play in helping their children have some interest in attending college, prepare for college. Elementary and middle whereas others may focus on students schools also can help set students on the who are unlikely to attend college due to path to college. In addition, community difficulty with achievement, attendance, or organizations often play a critical role in behavior in high school. In addition, col- providing the academic or social supports lege access programs have been studied to assist students in preparing for college. and tested in the real world, where a range Finally, states and the federal government of college access services is provided. impact college access through financial aid Thus, research on college access programs policies. Although the guide does not tar- is generally designed to ask whether a par- get these stakeholders, they may find the ticular college access program is more ef- recommendations relevant. fective than other services being provided. It is not designed to ask whether college Although the guide addresses ways to im- access services are effective compared to prove students’ and parents’ knowledge offering no services at all. of financial aid (see recommendation 5), it does not include recommendations on In offering these recommendations, the how to provide financial assistance to stu- panel is confident that it is important to dents. The panel focused on steps that offer college access services to ensure high schools and districts could take to that all students who want to attend col- improve college access, and federal and lege are prepared to do so. The guide in- state financial aid policy is beyond the cludes the set of recommendations that scope of the guide. we believe are a priority to implement. However, the nature of the research is Throughout the guide, the panel uses the such that we do not have a strong evi- term college to refer broadly to all types dence base for recommending specific of two- and four-year institutions. When practices over others; thus, the recom- necessary, the terms two-year college and mendations are all supported by low or

( 8 ) scope of the practice guide

moderate levels of evidence as described place for all students at every grade level. in the introduction. Recommendation 2 promotes a culture of evidence by encouraging schools and Summary of the recommendations districts to use assessments that determine whether students are on track aca- This practice guide includes five recom- demically for college and points out the mendations for how high schools and importance of early warning systems for school districts can improve access to students who are deficient in particular higher education. The first two recom- courses. The panel emphasizes here that mendations focus on preparing students assessment without action is virtually academically for college by offering a col- meaningless. Once deficiencies have been lege preparatory curriculum and assessing found, students and their families need to whether students are building the knowl- understand them, and they need to be as- edge and skills needed for college. These sisted in overcoming them. two recommendations reflect the panel’s belief that students are best served when Recommendation 3 describes how high schools develop a culture of achievement schools can help students build college- and a culture of evidence. The next recom- going networks by linking students to col- mendation describes how high schools lege-educated mentors, encouraging stu- can build and sustain college aspirations dents to form academically oriented peer by surrounding students with adults and groups, and allowing students to explore peers who support these aspirations. Rec- a variety of careers. These activities can ommendations 4 and 5 explain how high build a college-going identity and support schools can assist students in completing students’ aspirations. the critical steps to college entry, including college entrance exams and college and Recommendations 4 and 5 address steps financial aid applications. schools can take to assist students in completing the discrete tasks for college entry. Recommendation 1 advises schools and The panel considers it imperative that districts to ensure that every student has thinking about applying to college and the ability to be ready to take college- how to pay for college need to begin before level courses by beginning preparation in the 12th grade. Financial literacy about the 9th grade. Students and their families college affordability is an example of an need to understand what the requirements activity that could occur as early as 9th are for college, and what is needed to apply grade. At the same time, some activities to certain postsecondary institutions. In are specific to the senior year. This guide particular, the panel recommends that offers recommendations for the actions students complete Algebra I by the end of that will enhance the ability of students to 9th grade, and by graduation, complete complete the college application process coursework in core academic and elective successfully. areas that make them minimally proficient to attend community colleges without the The panel appreciates that schools and need for remediation. Such actions can districts may face challenges and road- strengthen the culture of achievement blocks in implementing all of the recom- within a school. mendations. Many of these recommendations could require additional staff or other If schools and districts are to monitor resources that are not easily accessible to student progress toward being academi- schools. Implementing these strategies may cally prepared for college, then they need require changing mind-sets and promot- to have adequate assessment measures in ing new behaviors, which may not happen

( 9 ) scope of the practice guide

immediately. Schools also have different will be achieved by a coordinated effort personnel responsible for different activi- in implementing these recommendations. ties, and the panel has avoided specifying The suggested practices need to be devel- one individual who must undertake a partic- oped systematically, monitored, evaluated, ular activity and instead has focused on key and modified, if necessary. The guide is actions that schools need to take to improve not meant as a resource that school prin- college access. To address these concerns, cipals can use to implement an individual each recommendation includes a series of recommendation and be successful. Al- roadblocks and suggested approaches that though an individual recommendation offer innovative solutions to some issues may succeed in improving college-going that schools may encounter when imple- rates, the panel discourages schools and menting the recommendations. districts from employing a piecemeal approach. Students will be best served by The panel believes that the greatest suc- a strategic plan for implementing all five cess in increasing student access to college recommendations in their schools.

( 10 ) Checklist for carrying out the  Facilitate student relationships with recommendations peers who plan to attend college through a structured program of extracurricular Recommendation 1. Offer courses and activities. curricula that prepare students for college-level work, and ensure that  Provide hands-on opportunities for students understand what constitutes a students to explore different careers, and college-ready curriculum by 9th grade assist them in aligning postsecondary plans with their career aspirations.  Implement a curriculum that prepares all students for college and includes op- Recommendation 4. Engage and assist portunities for college-level work for ad- students in completing critical steps vanced students. for college entry

 Ensure that students understand what  Ensure students prepare for, and constitutes a college-ready curriculum. take, the appropriate college entrance or admissions exam early.  Develop a four-year course trajectory with each 9th grader that leads to fulfilling  Assist students in their college search. a college-ready curriculum.  Coordinate college visits. Recommendation 2. Utilize assessment measures throughout  Assist students in completing college high school so that students are aware applications. of how prepared they are for college, and assist them in overcoming Recommendation 5. Increase families’ deficiencies as they are identified financial awareness, and help students apply for financial aid  Identify existing assessments, standards, and data available to provide an  Organize workshops for parents and estimate of college readiness. students to inform them prior to 12th grade about college affordability, schol-  Utilize performance data to identify arship and aid sources, and financial aid and inform students about their academic processes. proficiency and college readiness.  Help students and parents complete  Create an individualized plan for stu- financial aid forms prior to eligibility dents who are not on track. deadlines.

Recommendation 3. Surround students with adults and peers who build and support their college-going aspirations

 Provide mentoring for students by recent high school graduates who enrolled in college or other college-educated adults.

( 11 ) Recommendation 1. college. A four-year course trajectory can then help students plan and Offer courses and complete the coursework needed curricula that prepare to prepare for college. students for college- Level of evidence: Low level work, and ensure that students The panel judged the level of evidence for this recommendation to be low. None of understand what the studies examining the impact of offer- constitutes a college- ing a college-ready curriculum met WWC standards. The lack of evidence partly ready curriculum reflects the challenge of rigorously eval- by 9th grade uating the impact of high school course taking—students who choose to enroll in rigorous courses can differ in important The courses students take in high school ways from students who do not. have important consequences for their academic preparation and their ability The evidence for taking a college-ready to access college.13 Yet, low-income curriculum consists of six studies that po- and first-generation students are less tentially met standards.15 Two of the stud- likely than other students to complete ies provide mixed evidence on the effect a rigorous high school curriculum that of a rigorous high school curriculum,16 prepares them for college, either because and four studies show positive effects of it is not offered by their high school AP courses.17 The evidence for academic or they are not encouraged to enroll advising is stronger, with six relevant pro- in it.14 It is critical that high schools grams that had studies meeting standards, enable students to enroll in courses but the impact of academic advising could that will prepare them academically for not be isolated from other program com- college-level work. This process has two ponents.18 Despite the limited evidence for steps: offering the relevant courses and this recommendation, the panel believes advising students to take them. that offering the courses needed to prepare for college and informing students about those courses are critical steps for High schools should offer, as a default, improving college access. a college-ready curriculum that includes specific courses in key subjects. The panel defines a college-ready curriculum 15. Allensworth et al. (2008); Attewell and Domina as one that, when completed, will enable (2008); Dougherty, Mellor, and Jian (2006); Har- grove, Godin, and Dodd (2008); Jeong (2009); students to enroll in college without need Keng and Dodd (2008). for remediation. In addition, all schools 16. Allensworth et al. (2008); Attewell and should offer Advanced Placement (AP) Domina (2008). or other college-level opportunities. By 17. Dougherty, Mellor, and Jian (2006); Hargrove, 9th grade, students need to understand Godin, and Dodd (2008); Jeong (2009); Keng and the courses that comprise this curricular Dodd (2008). track and their importance for accessing 18. EXCEL—Bergin, Cooks, and Bergin (2007); Tal- ent Search—Constantine et al. (2006); Middle Col- lege High School—Dynarski et al. (1998); Sponsor- 13. Adelman (1999, 2006). a-Scholar—Johnson (1998); Upward Bound—Myers 14. Adelman (1999); Alexander (2002); Martinez et al. (2004); Quantum Opportunity Program and Klopott (2003); Wimberly and Noeth (2005). (QOP)—Schirm, Stuart, and McKie (2006).

( 12 ) Recommendation 1. Offer courses and curricula that prepare students for college-level work

Brief summary of evidence to However, a positive correlation does not support the recommendation mean that the coursework caused higher achievement. Two studies that potentially meet standards examined the effect of high school course Six programs with studies meeting stan- taking.19 One study found that taking a dards provide evidence on academic advis- more intense curriculum in high school has ing.26 Most of these programs offered indi- positive effects on high school performance vidual assistance to students in selecting and the likelihood of entering and complet- the classes needed to prepare for college. ing college.20 Curricular intensity is based Sponsor-a-Scholar also worked with school on the number of credits in core subject staff to ensure that students enrolled in a areas, the highest math course taken, the college preparatory curriculum, and EXCEL number of AP courses completed, and en- made completion of a college preparatory rollment in remedial math or English. How- curriculum a requirement for receiving a ever, a study of a school district that ended scholarship. Two of these programs27 had remedial classes and required college prep a positive impact on college enrollment, coursework for all students found no effect whereas the other four programs28 had on high school dropout rates or the likeli- no impact. hood of entering college.21 How to carry out this The panel identified four studies that po- recommendation tentially meet standards that examine the effect of AP course taking.22 These studies 1. Implement a curriculum that prepares all report positive effects of enrolling in an AP students for college and includes opportu- course or taking the AP course and exam nities for college-level work for advanced on high school completion, college entry, students. and college degree completion. The panel recommends that high schools Five correlational studies provide addi- and districts offer the courses and curri- tional evidence on the relationship be- cula needed to prepare students for college. tween course taking and achievement.23 This includes providing courses that are Two of these studies found a positive cor- required for entry into a two- or four-year relation between a rigorous high school college and providing rigorous academic curriculum and completion of a college coursework that prepares students for the degree,24 and three studies reported a pos- demands of college. Table 3 presents exam- itive effect on high school achievement.25 ples of college preparatory course requirements. Although there are slight differences in the requirements, all include four years 19. Allensworth et al. (2008); Attewell and of English, at least three years of mathemat- Domina (2008). ics, two to three years of science and social 20. Attewell and Domina (2008). studies, and one to two years of a foreign 21. Allensworth et al. (2008). 22. Dougherty, Mellor, and Jian (2006); Hargrove, 26. EXCEL—Bergin, Cooks, and Bergin (2007); Godin, and Dodd (2008); Jeong (2009); Keng and Talent Search—Constantine et al. (2006); Mid- Dodd (2008). dle College High School—Dynarski et al. (1998); 23. Adelman (1999, 2006); Gamoran and Hanni- Sponsor-a-Scholar—Johnson (1998); Upward gan (2000); Lee and Ready (2009); Lee, Croninger, Bound—Myers et al. (2004); QOP—Schirm, Stuart, and Smith (1997). and McKie (2006). 24. Adelman (1999, 2006). 27. Talent Search and Sponsor-a-Scholar. 25. Gamoran and Hannigan (2000); Lee and 28. EXCEL, Middle College High School, QOP, and Ready (2009); Lee, Croninger, and Smith (1997). Upward Bound.

( 13 ) Recommendation 1. Offer courses and curricula that prepare students for college-level work

Table 3. Examples of college preparatory course requirements

Program/ Additional Requirements English Mathematics Science Social Studies Courses High Schools Four years Four years: At least three Three or more At least one That Work Algebra I, years: biol- years computer course geometry, ogy, chemis- Algebra II, and try, physics or a fourth higher- applied phys- level mathematics ics, or anatomy/ course physiology State Scholars Four years Three years: Three years: Three and Two years of a Initiative (SSI) Algebra I, Algebra biology, chemis- a half years: language other II, and geometry try, and physics U.S. and world than English history, geography, economics, and government California’s A–G Four years Three years: Two years: Two years: Two years of a Requirements Elementary and biology, chemis- world history, language other advanced algebra try, physics, or cultures, and than English; one and geometry physical science geography; U.S. year of visual and history performing arts; one year of college preparatory elective Indiana “Core 40” Four years Three years: Three years: Three years: Three years of Curriculum Algebra I, Algebra biology, chem- U.S. history, world language, II, and geometry istry or physics, U.S. govern- fine arts, and/or and one addi- ment, economics, physical education tional course world history or geography Academic Four years Three years: Three years: Three years One year of a Competitiveness including Algebra biology, chemis- language other Grant Requirements I and a higher- try, and physics than English level class KnowHow2Goa Four years Three or more Three or more Three or more Possibly foreign years: including years years language, arts, Algebra I and a computer science higher-level class a. Source: www.knowhow2go.org

language. These requirements also specify with opportunities for prepared students that students should take algebra and other to take college or college-level courses. higher-level mathematics courses during This includes dual enrollment arrange- high school. The panel recommends that ments that allow students to take college at a minimum, all students should pass Al- courses for high school and college credit; gebra I by the end of their 9th-grade year. AP courses; and the International Bacca- Currently, 21 states and the District of Co- laureate (IB) program, which also can pre- lumbia have implemented a college-ready pare students for the academic demands curriculum for all students as a gradua- of college and facilitate some students’ tion requirement;29 districts in other states admission to more selective schools.30 could consider making a college-ready track A variety of resources are available to the default curriculum. help schools implement these types of programs, including state and federal AP The panel recommends that schools en- incentive programs or e-learning options hance their college-ready curriculum 30. Dougherty, Mellor, and Jian (2006); McCauley 29. Achieve, Inc. (2009). (2007); Perkins et al. (2004).

( 14 ) Recommendation 1. Offer courses and curricula that prepare students for college-level work

and partnerships with postsecondary in- schedule drop-in hours for students to stitutions to offer dual enrollment to quali- receive academic advising and assistance fied students.31 with selecting courses from a teacher, counselor, or other staff person.36 2. Ensure that students understand what constitutes a college-ready curriculum. Schools and districts also should provide continuing professional develop- There is substantial evidence that stu- ment or counseling for counselors, regis- dents do not understand the curricular trars, teachers, and other staff on college requirements for college entry and suc- prep course requirements, so that they cess, even those for community colleges.32 can serve as an informative resource for High schools should clearly communicate students.37 with students and families to ensure that they understand the courses needed for 3. Develop a four-year course trajectory college (and that students are on track to with each 9th grader that leads to fulfilling complete them, as discussed further in a college-ready curriculum. recommendation 2), before they enter high school. For example, students should know Beginning in 9th grade, high school coun- that in many states, they need to take the selors should work individually with each following types of classes beginning in student to ensure that he or she has a 9th grade: plan to complete the courses during high school. This could be structured as an in- • Geometry, algebra, trigonometry, dividualized education, learning, or gradu- advanced math ation plan that guides a student’s curricular choices throughout high school (see • American history, world history, civics Exhibit 2).38 Even though several states require that schools develop a plan that • Earth/physical science, biology, defines the courses a student will take in chemistry, physics high school,39 high schools should make sure that these plans are living documents This communication can come in the form that are referred to by teachers and coun- of a mailing (see Exhibit 1)33 or in gen- selors and provided to parents. eral advice provided by school or college access program staff by the end of 8th Providing students with information about grade.34 In later years, students still need the courses that are needed to prepare for one-on-one attention—from a counselor, college is only the first step. High schools a teacher, an administrator, or program need to ensure that students take the col- staff—to facilitate and encourage rigor- lege-ready curriculum throughout high ous course taking.35 A high school might school. The panel recommends that high schools first develop a general four-year 31. Hargrove, Godin, and Dodd (2008); Jackson (2009); Jeong (2009); Karp et al. (2007); Keng and Dodd (2008); Klopfenstein and Thomas (2009); 36. Calahan et al. (2004). Quint, Thompson, and Bald (2008); Siskin and Weinstein (2008). 37. Austin Independent School District, Office of Program Evaluation (2002); Perna et al. (2008). 32. Perna et al. (2008); Plank and Jordan (2001). 38. Christie and Zinth (2008); Robinson, Stempel, 33. Dounay (2008). and McCree (2005). 34. Constantine et al. (2006); Maxfield et al. 39. Education Commission of the States. Addi- (2003); Quigley (2003). tional High School Graduation Requirements and 35. Bergin, Cooks, and Bergin (2007); Gandara (2002, Options (http://mb2.ecs.org/reports/Report. 2004); Gandara et al. (1998); Johnson (1998). aspx?id=740, accessed June 2, 2009).

( 15 ) Recommendation 1. Offer courses and curricula that prepare students for college-level work

Exhibit 1. Example of course requirement mailing

Source: South Dakota Board of Regents website, www.sdbor.edu.

( 16 ) RecoECOmmendationEnDATIOn 1. OOfffferER coCOURSESurses anAnDd cuCURRICUrriculalA thTHATat prPREPAREepare stSTUDEudentsnTS forOR coCOllegellEgE--llevEvElel woWORrk

Exhibit 2: Example of a personalized learning plan

NAME Student ID # Meeting Dates: Grade: 9 10 11 12 GRADUATION REQUIREMENTS: 220 Credits English 4 years (40 credits) CAHSEE Passed: ELA Math 1st Sem Grade 2nd Sem Grade Remember that High School Graduation Requirements English I English I are not the same as the College Admission Require- English 2 English 2 ments (A–G). Know the difference! Visit the school English 3 English 3 website for detailed college admissions information. English 4 English 4 Social Studies 3 Years (30 Credits) HIGH SCHOOL CEEB CODE xxxxxx 1st Sem Grade 2nd Sem Grade WH WH A–G Requirements (NO “D” grades Accepted) CALIFORNIA STATE UNIVERSITY (CSU) www.csumentor.edu US Hist US Hist UNIVERSITY OF CALIFORNIA (UC) www.ucop.edu/pathways US Gov’t/ Econ/US Econ Gov’t Mathematics 2 Years (20 credits) – Subject Area Years Required Must include HS Algebra A. Social Science Two years. One year US History or ½ year US History and ½ year of st nd 1 Sem Grade 2 Sem Grade US Government, and one year of Algebra Algebra World History. ______B. English Four years of college-preparatory Math Math English composition and literature. ______C. Mathematics Three years required. Four years Science 2 Years (20 Credits) – Algebra 1 recommended. Algebra 2/Trigonometry Physical and Life/Biological Geometry 1st Sem Grade 2nd Sem Grade Advanced Math Phy Sci Phy Sci D. Laboratory Science Two years required. Three years Life Life recommended. (Two of the following: Biology, Chemistry, Physical Education 2 Years (20 Credits) Physics, or other approved 1st Sem Grade 2nd Sem Grade college preparatory science.) E. Language other than English Two years required. Three years PE 9/Sport PE 9/Sport recommended—at least two PE/Sport PE/Sport years of the same language. Fine Arts OR Foreign Language 1 Year (10 Credits) F. Visual Performing Arts One year is required. 1st Sem Grade 2nd Sem Grade G. College Preparatory Elective One year in most of the above A–F areas, or approved elective. FA/FL FA/FL NCAA (www.ncaaclearinghouse.org) Health (5 Credits) and 16 Core Courses beginning with the class of 2008 College & Career Planning (5 Credits) Health CCP Career Goal:

Electives (75 Credits) 1st 2nd 1st 2nd Plans after High School: Sem Sem Sem Sem Grade Grade Grade Grade

Student Signature

Parent Signature Additional Graduation Requirement:

(continued)

( 17 ) RECOmmENDATIONEnDATIOn 1. OOffffER COCOURSESURSES ANAnDD CURRICUCURRICULAlA THATTHAT PREPAREPREPARE STUDESTUDENTSnTS fOR COllLLEGEEgE-l-LEVEvElEL WORWORk

Exhibit 2: Example of a personalized learning plan (continued)

“Four-Year Plan” Worksheet

Student Name Date

Career Goal (check one) Four–Year University Community College (Transfer Program) Trade/Tech/Art School Other

GRADE 9 (Student must take 6 classes—60 credits total for year) GRADE 10 (Student must take 6 classes—60 credits total for year) Fall Semester Spring Semester Fall Semester Spring Semester English I English I English II English II Math Math Math Math Science Science Science Science PE PE World History World History Reading or Elective Health PE PE Elective Elective Elective Elective Elective (Optional) Elective (Optional) Elective (Optional) Elective (Optional) GRADE 11 (Student must take 5 classes—50 credits total for year) GRADE 12 (Student must take 5 classes—50 credits total for year) Fall Semester Spring Semester Fall Semester Spring Semester English III English III English IV English IV U.S. History U.S. History American Gov’t Economics Elective Elective Elective Elective Elective Elective Elective Elective Elective Elective Elective Elective Elective (optional) Elective (optional) Elective (optional) Elective (optional) Elective (optional) Elective (optional) Elective (optional) Elective (optional)

Four-Year University Bound Student (Example) Meeting UC / CSU A–G Course Sequence

Career Goal (check one) Four–Year University Community College (Transfer Program)

GRADE 9 (Student must take 6 classes—60 credits total for year) GRADE 10 (Student must take 6 classes—60 credits total for year) Fall Semester Spring Semester Fall Semester Spring Semester English I or English I Acc English I or English I Acc English II or English II Acc English II or English II Acc Algebra I or higher math Algebra I or higher math Geometry or higher math Geometry or higher math Biology Biology Chemistry or Conceptual Physics Chemistry or Conceptual Physics Health College Career Planning World History or AP European World History or AP European PE Activities 9 or Sport PE Activities 9 or Sport PE or Sport PE or Sport World Language 1 World Language 1 World Language II World Language II Elective (Optional) Elective (Optional) Elective (Optional) Elective (Optional) GRADE 11 (Student must take 5 classes—50 credits total for year) GRADE 12 (Student must take 5 classes—50 credits total for year) Fall Semester Spring Semester Fall Semester Spring Semester English III or AP Language English III or AP Language English IV or AP Literature English IV or AP Literature U.S. History or AP US History U.S. History or AP US History American Gov’t or AP Gov’t Economics or AP Economics Algebra II/Trig or higher math Algebra II/trig or higher math Advanced college prep math Advanced college prep math Advanced College Prep Science Advanced College Prep Science Visual Performing Art Visual Performing Art World Language III World Language III Elective Elective Elective (optional) Elective (optional) Elective (optional) Elective (optional)

Source: Adapted from materials created by a National College Advising Corps program site.

( 18 ) Recommendation 1. Offer courses and curricula that prepare students for college-level work

course trajectory that defines the poten- in their efforts. Developing a culture of tial timing and sequence of college-ready achievement among the faculty is a key classes for students. This may include op- strategy to supporting students in their ef- tions for the core courses students should forts so that the teachers are interested in take each year in order to prepare for col- and willing to help students who are chal- lege. This approach offers a curriculum lenging themselves. Schools can provide path that students can use to inform their academic support for students who take specific selection of classes each year of the most rigorous course load available high school. by setting up peer-tutoring opportunities so that stronger students can work with Potential roadblocks and solutions those students who may be struggling in honors, AP, or IB classes. Teachers can Roadblock 1.1. Teachers may not be be encouraged to set up this sort of peer trained to teach advanced courses. system among students in their individual classes. Suggested Approach. Not all teachers must be trained to teach advanced courses, Roadblock 1.3. Our high school has lim- but teachers should have access to pro- ited information on entering 9th-grade stu- fessional development opportunities that dents to assist them in planning their high help them sharpen their skills so that the school coursework. curriculum they teach is as rigorous and engaging as possible. Helping teachers un- Suggested Approach. High schools need derstand how their classes fit with a college academic information on incoming stu- preparatory sequence begins with asking dents to help them plan a four-year course them to participate in the planning of the trajectory that will prepare students for articulation of the curriculum. Schools college. High schools can coordinate with also can reach out to institutions of higher middle schools to obtain transcripts, aca- education to implement dual enrollment demic records, and other resources that opportunities or have community college help high school staff better understand professors teach courses on campus. the needs of incoming students. This information can be used to assist students in Roadblock 1.2. Enrolling students who selecting appropriate high school courses are not prepared for the academic rigor in and to help high schools create an appro- college prep or college-level classes is seen priate four-year plan with students. High as counterproductive. schools also can coordinate with middle schools to inform students about how Suggested Approach. It is critical that all their middle school performance affects students have the option to participate in the courses they will take in high school these types of classes and are supported and their ability to access college.

( 19 ) Recommendation 2. the panel expects that research on their Utilize assessment use also will expand. measures throughout Although four programs with studies meet- high school so that ing standards included practices related to data use and additional instruction to as- students are aware of sist students, these practices were neither how prepared they are isolated in the evaluation nor necessarily a major component of the program.41 for college, and assist Studies of two additional programs42 that them in overcoming potentially meet standards suggest that the use of data to identify and notify stu- deficiencies as they dents of their academic progress during are identified high school had an impact on college outcomes. There is also suggestive evidence that district- or statewide use of assess- Completing the courses needed to ments associated with college readiness graduate from high school and meet (such as PLAN and ACT43) is associated college entry requirements does with improved college outcomes, but this not guarantee that students have correlation does not mean that requiring the knowledge and skills needed to students to take those tests caused im- succeed in college. Many high schools proved access to college.44 produce students who may pass state exit exams and meet graduation Brief summary of evidence to standards but still are academically support the recommendation unprepared for college, as evidenced by the nearly 60 percent of students Two programs with studies that potentially who are required to take remedial met standards assess students when they courses as a condition of enrollment.40 are high school juniors to determine their High schools must assess student readiness for college-level work. California’s progress to identify, notify, and assist Early Assessment Program (EAP) uses as- students who are not adequately sessment results to inform students about prepared as early as possible in their whether they need additional preparation academic career. to become college ready and includes supplemental programming for students who Level of evidence: Low do not meet expectations.45 The College

The panel determined that the level of evi- 41. Talent Search—Constantine et al. (2006); dence supporting this recommendation is Sponsor-a-Scholar—Johnson (1998); QOP— low. In this case, the rating is not neces- Schirm, Stuart, and McKie (2006); Upward sarily a result of limited or poor research: Bound—Seftor, Mamun, and Schirm (2009). the ability to implement related data and 42. College Now—Crook (1990); California Early assessment systems is a fairly recent de- Assessment Program (EAP)—Howell, Kurlaender, and Grodsky (2009). velopment. Advances in both capabilities and resources devoted to state longitu- 43. Although ACT was originally an acronym for American College Testing, the official name dinal datasets have been promising, but is now ACT. PLAN is the name of an assessment these data do not yet exist for many juris- administered by ACT. dictions. As they become more prevalent, 44. ACT (2008a, 2008b, 2009a, 2009b). 45. California State University (2005); Howell, 40. Bailey (2009). Kurlaender, and Grodsky (2009).

( 20 ) Recommendation 2. Utilize assessment measures throughout high school

Now program in New York City uses assess- Correlational studies that examined ACT’s ment data to determine whether students College Readiness System provide sugges- are eligible for dual enrollment courses or tive evidence on identifying and notifying need developmental classes to prepare for students who are not college ready.50 The college-level coursework.46 Studies of both College Readiness System includes the EX- programs found that they reduced the need PLORE and PLAN assessments in 8th and for remediation in college, and College Now 10th grades that are precursors to the ACT increased the number of college credits that (one of two national college admissions students earned. tests) and the COMputer-adapted Placement Assessment and Support Services (COM- Two additional programs with studies PASS), a college placement test adminis- that potentially met standards include tered by ACT. The panel also relied on other elements of academic support. Project descriptive and qualitative studies.51 GRAD analyzed data to understand and track students’ progress toward meeting How to carry out this graduation requirements. GEAR UP sites recommendation provide individualized academic support for students with academic problems and 1. Identify existing assessments, standards, those who do not perform well on stan- and data available to provide an estimate of dardized assessments.47 Studies of both college readiness. programs examined middle school outcomes and did not report high school or Assessments can play a key role in alert- college outcomes.48 ing students, parents, and teachers about whether students are “on track” for col- Four college access programs that provided lege matriculation when they graduate academic assistance to improve students’ from high school. Currently, no single academic proficiency had studies that met college-readiness assessment is commonly WWC standards.49 Talent Search, Sponsor-a- available or used by schools and districts Scholar, and the Quantum Opportunity Pro- (although there is progress in that direc- gram (QOP) offered fairly low-intensity aca- tion52). Recognizing the limited time and demic assistance through tutoring services resources schools have to develop a new or homework help after school. Although assessment, the panel recommends that Talent Search and Sponsor-a-Scholar had a high schools consider several existing as- positive impact on college enrollment, aca- sessments that can provide an early indi- demic services formed a minor component cation of students’ academic preparation of all three. Upward Bound offered addi- for college: tional academic coursework throughout the school year and during a six-week summer • College or community college place- session but did not have an impact on col- ment exam. Schools and districts can lege enrollment or degree attainment. use whole assessments or a subset of items from existing college or community college placement exams as a diagnostic measure. Although many placement exams are school specific, 46. Crook (1990); Karp et al. (2007, 2008). 47. Standing et al. (2008). 48. Opuni (1999); Standing et al. (2008). 50. ACT (2008a, 2008b, 2009a, 2009b). 49. Talent Search—Constantine et al. (2006); 51. Achieve, Inc. (2009); Austin Independent Sponsor-a-Scholar—Johnson (1998); QOP— School District, Office of Program Evaluation Schirm, Stuart, and McKie (2006); Upward (2002); Quint, Thompson, and Bald (2008). Bound—Seftor, Mamun, and Schirm (2009). 52. See, for example, Achieve, Inc. (2009).

( 21 ) Recommendation 2. Utilize assessment measures throughout high school

some common assessments can be ad- In some cases, schools may be able to ob- opted by a high school (e.g., COMPASS tain financial support for implementing and ACCUPLACER, an assessment de- one of these assessments.56 veloped by the College Board and used to help determine course selection for The information gathered from these as- students). sessments should be combined with other indicators of academic progress to deter- • College admissions exams. High mine if students are on track for college schools can have students take one as defined by coursework progression of the college admissions exams de- and academic proficiency. High schools signed for students in early high school can connect assessments in each of these grades (e.g., PSAT, EXPLORE, PLAN).53 areas into a cohesive set of information These assessments can gauge early ac- that can be used in guidance and planning. ademic preparation in math and read- High schools can assess coursework pro- ing as well as reasoning and critical gression against the college preparatory thinking. Later in high school, states tracks described in recommendation 1. can have all students take the college Academic proficiency information is con- admission exams (e.g., SAT, ACT) to tained in existing state assessments, and gauge their college readiness. postsecondary aspirations can be assessed in a brief student survey. • Statewide college and career readiness assessments. Schools in states To gauge whether they are successfully that already conduct a college or career preparing students for college, high assessment should take advantage of schools should gather information on these assessments and use them as an postsecondary enrollment for past stu- indicator of college preparedness.54 dents. In some states, high schools can gather this information from a state data- • Local assessments. In districts, base that tracks students from kindergar- schools can use existing benchmark ten through college. In other states, high assessments on a regular basis to mea- schools and districts can track their grad- sure students’ progress against stan- uates through the National Student Clear- dards tied to academic proficiency.55 inghouse (http://www.studentclearing- house.org), a comprehensive student-level repository of data from 3,300 postsecondary institutions attended by 92 percent 53. Achieve, Inc. (2009); Dounay (2006); Howell, of college students in the United States. Kurlaender, and Grodsky (2009). For example, Buffalo, New York, administers the PSAT to first- Alternatively, high schools may be able to year high school students; Chicago, Illinois, uses partner with local and regional postsec- the ACT’s EPAS system, administering EXPLORE ondary institutions to gather information to its 8th and 9th graders and PLAN to its 10th on the enrollment of their graduates. The and 11th graders. panel recommends that high schools use 54. Ten states administer college and career- these data to understand the enrollment readiness assessments to all students: four are state specific, as is the Early Assessment Pro- rate, persistence, and degree attainment gram, and can replace placement tests; one ad- of graduates in order to better understand ministers the SAT; and five administer the ACT the impact of current practices. statewide (Achieve, Inc., 2009). 55. Quint, Thompson, and Bald (2008). For example, school districts in Richmond, Virginia, and Fresno, California, use benchmark assessments 56. For example, Florida and South Carolina pro- every nine weeks to measure students’ progress vide funding to districts that want to administer against standards tied to academic proficiency. the PSAT or PLAN assessment to their students.

( 22 ) Recommendation 2. Utilize assessment measures throughout high school

2. Utilize performance data to identify and about the progress they are making and inform students about their academic profi- the hurdles they need to overcome in ciency and college readiness. becoming college ready.58 Students and families should receive the results of The information schools collect on aca- the data collected by the school, pos- demic performance and college readiness sibly in the form of a data report or a (step 1) should be used to identify students letter. For example, a data report might who are falling behind and to inform all stu- include information on course grades, dents of their progress in becoming college college-readiness assessment results, ready. This applies to both the courses stu- and high school course completion.59 dents need to be qualified for college entry Students identified as below grade level and the skills they acquire in those courses or not on track for college should have to avoid remediation once they matriculate. an individual meeting with someone The use of performance data should occur at the school to discuss the results as early as 9th grade to ensure that students and their implications for accessing can take the necessary steps to get back college. Students who are not making on track. The panel recommends using the progress toward completing graduation data in the following ways: or college preparatory requirements should be notified of possible interven- • Identify students with college expec- tions that can help them get back on tations who are performing below track (e.g., summer school, remediation grade level and who are not on a col- programs).60 lege-ready track. Schools should identify students who are not meeting grade- 3. Create an individualized plan for students level standards and who are not on track who are not on track. for college but have college aspirations. Although state assessments can be used Students who are not on track to complete a to identify students performing below typical academic course sequence often have grade level, course grades, grade point trouble catching up and meeting college- average (GPA), course completion, and readiness objectives.61 The earlier in high college-readiness assessments can be school a student can catch up to a standard used to identify students who are not on course sequence, the greater the likelihood track for college. For example, a school of meeting college entrance requirements at can flag students who are performing the time of high school graduation. below a certain GPA, or students who have not completed courses on the col- High schools should work with students lege preparatory track. High schools who are not on track to develop a plan should obtain and use middle school that will assist them in “catching up.”62 The transcripts of their incoming students plan should specify the steps students will to support course placement and flag take to get back on track academically and entering 9th graders with academic defi- the additional instruction they will receive ciencies before those students step foot to support their academic proficiency. on campus.57 High schools, colleges, and a variety of

• Inform all students about their per- 58. Dounay (2006). formance and its implications for ac- 59. Gewertz (2009). cessing college. Discussions with stu- 60. Christie and Zinth (2008). dents should be held at least annually 61. Wimberly and Noeth (2005). 62. Quint, Thompson, and Bald (2008); Robinson, 57. Gewertz (2009). Stempel, and McCree (2005).

( 23 ) Recommendation 2. Utilize assessment measures throughout high school

student academic support programs pro- can use double blocking to enable first- vide a range of options for students who year students needing extra help to take are behind but eager to make progress. “catch-up” classes for two periods each Teachers, counselors, and college advisors day during the first semester and then the can play a pivotal role in helping students regular academic or college prep classes make the best choices for supplemental in the second semester.66 instruction and in connecting students back into more typical instructional pro- Tutorials should be provided whenever grams at the appropriate time. The panel possible to accommodate all students—on encourages high schools to choose these Saturdays, before school, after school, or programs carefully, paying particular at- during lunch. One potentially effective ap- tention to the “fit” between a particular proach is to develop and publicize a matrix student’s demonstrated need and the pro- that shows when tutorials are available and gram’s intent. “Reteaching” a student with who will be providing them. In this way, similar instructional strategies, in simi- students needing additional help need not lar instructional settings, may not be as be limited to working with the teacher in useful as a more customized approach whose class they are struggling.67 to matching deficiencies with deliberate progress objectives. Potential roadblocks and solutions

Specifically, high schools and districts can Roadblock 2.1. Obtaining new data on collaborate with postsecondary institu- students can be expensive, schools do not tions or existing college access programs have the capacity or resources to generate to offer additional instruction during out- student-level reports, and students already of-class time. One example is tutoring and take enough tests. homework assistance by college students, program staff, or teachers, and in a vari- Suggested Approach. It is certainly true ety of formats: both in small groups and that few schools have the capacity to im- one-on-one.63 More formally, schools can plement this recommendation at the build- implement “recovery” programs for math ing level, but it is feasible for many dis- courses, in which students who fail a unit tricts. Consider replacing any district-level are immediately required to attend after- assessment that does not measure college school instruction for that unit and as an readiness with one that does, such as the incentive for attendance may be given PSAT or equivalent. Alternatively, exam- the opportunity to improve their grade ine how well the district-level assessment for that unit.64 Another option is summer already correlates with college going and programs for academic enrichment or re- success. The panel recommends taking medial skills when necessary.65 advantage of all of the reports and/or data systems that are generated and main- An alternate strategy is additional or re- tained by the state. structured courses to address academic deficiencies during school time. Schools Roadblock 2.2. Some school staff—teachers and/or counselors—do not have the time or the training to collect or analyze 63. Austin Independent School District, Office of data and may even view the use of data Program Evaluation (2002); Calahan et al. (2004); Johnson (1998); Maxfield et al. (2003). as destructive (a mechanism for criticism) 64. Robinson, Stempel, and McCree (2005). 65. Austin Independent School District, Office of 66. Kemple, Herlihy, and Smith (2005); Quint, Program Evaluation (2002); Kallison and Stader Thompson, and Bald (2008). (2008); Snipes et al. (2006). 67. Robinson, Stempel, and McCree (2005).

( 24 ) R meCOm eNDATION 2. Utilize assessment measures throughout high school

rather than as constructive (a strategy for available, make them available in their school improvement). classrooms or offices when possible, and model their effective use. When feasible, Suggested Approach. The panel recom- consider implementing common planning mends providing technical assistance to time or creating a “school improvement staff in interpreting and acting on findings data team” so that staff can share knowl- from student assessments. Make sure staff edge and ideas. are aware of the data that are currently

( 25 ) Recommendation 3. Although the panel identified evidence that supports mentoring—the first step of Surround students this recommendation—there is limited evi- with adults and dence on the other two steps. Six programs peers who build and with studies meeting or potentially meeting standards are aligned with one or more support their college- steps in this recommendation.71 Three of going aspirations these programs had a positive impact on college enrollment,72 and three did not impact enrollment.73 However, isolating Although 79 percent of students the impact of the recommended practices express college aspirations early in is difficult because these programs in- high school,68 college plans can falter cluded a variety of other strategies. A few if students do not take the necessary correlational studies reported a relation- steps to prepare for and enter ship between having friends with college college.69 By 12th grade, low-income plans and college enrollment, but they do and first-generation students are less not provide evidence on the causal effect likely than other students to expect to of college-going peers.74 Despite the low earn a bachelor’s degree or higher.70 evidence rating, the panel believes that High schools should build and support linking students with college-going adults students’ aspirations by developing and peers is important for building aspira- social networks that encourage college tions and supporting college entry. attendance and assist students in preparing for college. College students Brief summary of evidence to and college-educated adults can serve support the recommendation as mentors for students, providing guidance and support throughout The panel separately reviewed studies for the college preparation process. each of the steps that make up this recom- Extracurricular activities and college mendation. Four programs had studies access programs can encourage the that met or potentially met standards and formation of college-going peer groups provided mentoring services.75 Studies that share an interest in pursuing of Sponsor-a-Scholar, Career Beginnings, college. High schools can use career and Puente reported a positive impact exploration activities to develop students’ career interests and link those 71. Career Beginnings—Cave and Quint (1990); interests to postsecondary plans. Sponsor-a-Scholar—Johnson (1998); Puente— Gandara (2002); Career Academies—Kemple Level of evidence: Low (2004); QOP—Schirm, Stuart, and McKie (2006); Upward Bound—Seftor, Mamun, and Schirm (2009). The panel defined the level of evidence 72. Career Beginnings—Cave and Quint (1990); for this recommendation as low because Puente—Gandara (2002); Sponsor-a-Scholar— of limited evidence that the recommended Johnson (1998). practices improve college enrollment rates. 73. Career Academies—Kemple (2004); QOP— Schirm, Stuart, and McKie (2006); Upward Bound—Seftor, Mamun, and Schirm (2009). 68. National Center for Education Statistics 74. Horn and Chen (1998); Hossler, Schmit, and (2004). Vesper (1999); Sokatch (2006). 69. Gandara (2002); Kao and Tienda (1998); Rod- 75. Career Beginnings—Cave and Quint (1990); erick et al. (2008). Puente—Gandara (2002); Sponsor-a-Scholar— 70. National Center for Education Statistics Johnson (1998); QOP—Schirm, Stuart, and McKie (2006a); Nunez and Cuccaro-Alamin (1998). (2006).

( 26 ) Recommendation 3. Surround Students with College-Going Adults and Peers

on college enrollment,76 whereas a study career exploration activities, although nei- of QOP found no impact on college en- ther had an impact on college enrollment.81 rollment.77 Mentoring in these programs The Career Academies and Upward Bound consisted of a one-on-one relationship programs implemented a variety of career between a college-educated adult or case exploration activities, including speakers manager and a high school student. Men- from the business community, visits to tors were expected to meet regularly with employer sites, career planning assistance, students and, in most cases, provided as- career fairs, and job shadowing.82 The Ca- sistance or guidance with the college prep- reer Academies program also organized aration process. For example, mentors in students into learning communities that the Sponsor-a-Scholar program monitored focus on a career theme and provide work- students’ academic progress and assisted based learning experiences. them with college applications. The mentoring relationships lasted between two to How to carry out this four years depending on the program. recommendation

Three programs with studies that met or 1. Provide mentoring for students by recent potentially met standards focused on the high school graduates who enrolled in col- role of peers. A study of the Puente pro- lege or other college-educated adults. gram showed a positive impact on college enrollment,78 a study of the Career The panel recommends linking students to Academies program found no impact,79 adults who can serve as college-going role and a study of the Advancement Via In- models and build students’ interest in col- dividual Determination (AVID) program lege. High schools can recruit college-edu- did not measure college enrollment.80 All cated professionals to serve as volunteer three programs organized students into mentors by reaching out to local businesses groups that facilitated academically ori- interested in partnering with schools in the ented friendships. The Career Academies community. 83 High schools also can iden- program was structured as small learning tify volunteer mentors by recruiting local communities of students who take acad- college students—particularly graduates of emy classes together throughout high the high school—or partnering with a col- school. The Puente program mixed high- lege that has service-learning opportunities and low-achieving Latino students in an for college students willing to work with English class for two years, and students high school students.84 Individuals who in AVID, a program that placed promising share the same background as students, students in college preparatory course- such as high school alumni or profession- work, took a year-long course together als from the local community, may under- that supported their coursework and em- stand the types of challenges students face phasized a college-going identity. in reaching college.85

The panel identified two programs with studies meeting standards that offered 81. Career Academies—Kemple (2004); Upward Bound—Seftor, Mamun, and Schirm (2009). 82. Career Academies—Kemple, Poglinco, and 76. Career Beginnings—Cave and Quint (1990); Snipes (1999); Upward Bound—Myers et al. Puente—Gandara (2002); Sponsor-a-Scholar— (2004). Johnson (1998). 83. Cave and Quint (1990); Gandara (2004); Pell 77. QOP—Schirm, Stuart, and McKie (2006). Institute for the Study of Opportunity in Higher 78. Puente—Gandara (2004). Education (2006). 79. Career Academies—Kemple (2004). 84. Ladd (1992). 80. AVID—Watt et al. (2006). 85. Calahan et al. (2004); Gandara (2004).

( 27 ) Recommendation 3. Surround Students with College-Going Adults and Peers

Mentors can take on a variety of roles for panel recommends that mentors commu- students: nicate or meet at least monthly with first- year and sophomore students, and at least • Serve as college-going role models. weekly with juniors and seniors who are Mentors can serve as examples of col- engaged in the college application and selec- lege-going adults from the community tion processes. High schools also can sched- and share their experiences in prepar- ule social events or recreational activities ing for college, completing a college that bring together mentors and students. degree, and pursuing a career.86 An initial mentor training can prepare • Assist with the college entry process. mentors for their role. Providing examples The one-on-one relationship mentors of activities for mentors and students to have with students allows them to pro- complete together can support the mentor- vide individualized assistance with the ing relationship. In addition, high school college application and selection pro- staff should monitor mentor relationships cess for students interested in pursu- by checking in with students and mentors ing a four-year degree.87 This might to ensure that mentoring relationships are include helping with a college applica- supporting students.91 tion, reading an application essay, assisting with a financial aid application, 2. Facilitate student relationships with peers or researching college options (see rec- who plan to attend college through a struc- ommendation 5). tured program of extracurricular activities.

• Monitor academic progress. Men- The panel recommends that high schools tors can monitor students’ academic use extracurricular offerings to promote the progress by reviewing report cards formation of college-going peer groups. The and discussing students’ high school goal of these activities is to provide students coursework.88 Mentors can advocate with an opportunity to develop friendships for students who are struggling aca- with peers who plan to attend college. demically to receive tutoring or additional help. College access programs can bring a group of students together on a regular basis • Listen and advise. A mentor can sim- throughout the school year to focus on ply serve as a caring adult who listens preparing for college. These programs de- to the student, discusses his or her is- velop college-going peer groups by provid- sues or concerns, and offers advice as ing opportunities for students to work to- needed.89 gether toward a common goal of reaching college. Activities that encourage students To fulfill these roles, mentors need to com- to interact and collaborate can encourage municate regularly with students.90 The new relationships, and these programs can be used to promote a college-going identity. For example, a program might create vis- 86. Gandara (2004). ible markers of group participation, such 87. Cave and Quint (1990); Johnson (1998); Ladd as designating a group name and meeting (1992); Pell Institute for the Study of Opportunity 92 in Higher Education (2006). space or developing a group newsletter. 88. Johnson (1998). 89. Calahan et al. (2004); Schirm, Stuart, and McKie (2006). 91. Johnson (1998). 90. Johnson (1998); Schirm, Stuart, and McKie 92. Gandara (2004); Guthrie and Guthrie (2002); (2006). Mehan (1996).

( 28 ) Recommendation 3. Surround Students with College-Going Adults and Peers

High schools also can develop student students’ postsecondary plans.97 Students groups that encourage academically ori- may have limited understanding of the ented friendships, such as a debate club level or type of education required for an or an honor society.93 Schools can infuse occupation or career field. High schools these extracurricular activities with a col- should help students learn about the skills, lege-going message. For example, a debate knowledge, and postsecondary education club might visit a college to meet with the needed for their area of interest and pro- college debate team, or a community ser- vide examples of local colleges that offer vice club might collaborate with a student a degree in their area.98 This information organization from a local college. can be used to develop a long-term education plan for students that can be updated 3. Provide hands-on opportunities for stu- and revised over time. dents to explore different careers, and assist them in aligning postsecondary plans with Potential roadblocks and solutions their career aspirations. Roadblock 3.1. Mentoring relationships The panel recommends that high schools between students and mentors do not last, engage students in career exploration ac- and the availability of mentors changes tivities that provide hands-on experiences over time. with a career or occupation. A high school can design a sequence of career explora- Suggested Approach. High schools can tion activities that identify students’ career partner with local colleges that offer col- interests and provide a variety of activities lege students a chance to earn academic that inform and build on these interests.94 credits for volunteer work. College stu- For example, career or interest inventories dents who earn academic credit for their can be used to help students identify the mentoring role may be more likely to type of work or career that interests them. maintain their relationship with students High schools can use this information to throughout the school year. Local colleges invite local professionals from these ca- often have service-learning opportunities reer fields to speak about their education or campus organizations that can facilitate and career paths.95 Students can then be regular meetings with mentors. matched to job-shadowing opportunities that allow them to follow an adult through- Roadblock 3.2. Ninth-grade students are out the day and experience the day-to-day not interested in discussing careers or their work of a profession that matches their career interests. area of interest.96 By developing relationships with local employers, high schools Suggested Approach. Students may re- can link students to job-shadowing activi- late to the experience and career path of ties and help interested students obtain high school alumni who share the same short-term internships. background and perspective as they do. High schools should invite alumni to speak High schools can use career interests as with students about their occupations and a starting point for discussions about career paths. Listening to the outlook and advice of a role model who shares their 93. Gandara (2004); Mehan (1996). background and perspective may build 94. Kemple, Poglinco, and Snipes (1999). students’ interest in a career. 95. Kemple (2004); Mehan (1996). 96. Haimson and Deke (2003); Hershey et al. (1999); 97. Calahan et al. (2004); Roads to Success (2008). Kemple, Poglinco, and Snipes (1999); MacAllum 98. Austin Independent School District, Office of et al. (2002). Program Evaluation, (2002).

( 29 ) R meCOm eNDATION 3. Surround Students with College-Going Adults and Peers

Roadblock 3.3. Our school already offers Roadblock 3.4. Our school has insufficient many extracurricular activities and we can- resources to offer college access programs not add another. or other activities that bring together college-going peers. Suggested Approach. Given limited amounts of school resources and staff Suggested Approach. Schools can nomi- time, the panel recommends that high nate students for summer bridge programs schools think critically about the range at local colleges that allow students to inter- of extracurricular activities they offer. act with college-going peers, take additional High schools should consider the value academic coursework, and gain exposure to added by each activity and consider how a college campus. Although these programs activities contribute to the school’s college often target 12th-grade students who were goals. A strategic plan can help schools admitted to a college, summer programs think through the goals and expected out- also are available for students in grades 9 comes of each activity, as well as how to through 11. These programs can help stu- efficiently allocate available resources for dents learn about the social and cultural extracurricular activities. A high school challenges of attending college. For example, also could require completion of certain a program might teach students about time college-going milestones, such as submis- management and study skills or provide op- sion of a financial aid application, as a re- portunities to explore the college campus. quirement for attending an extracurricular High schools can inform students about the activity or social event. availability of these programs and help them understand the benefits of participating.

( 30 ) Recommendation 4. Level of evidence: Moderate

Engage and assist The panel judged the level of evidence for students in completing this recommendation to be moderate. Six pro- critical steps for grams with studies meeting standards with or without reservations assisted students college entry with the college entry process.101 Although these programs consisted of additional strategies to prepare students for college, all of Low-income and first-generation them focused on helping students com- students often face challenges in plete the steps to college entry. Three of completing the steps to college entry, the programs had a positive impact on col- such as taking college admissions lege enrollment,102 and three had no impact tests, searching for colleges, submitting on enrollment.103 The evidence supporting college applications, and selecting a this recommendation primarily consists of college.99 Students may not be aware programs that target low-income and first- of these steps, may lack information on generation students with average academic how to complete them, and may not achievement. Only one program, which did receive sufficient support and advice not have an impact on college enrollment, from those around them.100 targeted students at risk for dropping out of high school.104 Since programs that assisted students with the college entry process did High schools should engage students not consistently lead to positive impacts, the in the college entry process, providing panel defined the level of evidence for this hands-on assistance for each step. recommendation as moderate. Students who want to attend a two- or four-year institution should receive Brief summary of evidence to guidance in preparing for and taking support the recommendation the college admissions tests and searching for a college that matches Six programs with studies meeting stan- their qualifications, interests, and dards provided students assistance with goals. High schools should coordinate the college entry process.105 Five of these college visits to expose students to the college environment and to help 101. EXCEL—Bergin, Cooks, and Bergin (2007); them select a college. By providing Career Beginnings—Cave and Quint (1990); Tal- one-on-one assistance with college ent Search—Constantine et al. (2006); Sponsor- applications, schools can ensure that a-Scholar—Johnson (1998); QOP—Schirm, Stu- students submit applications that are art, and McKie (2006); Upward Bound—Seftor, Mamun, and Schirm (2009). complete, on time, and of sufficient 102. Career Beginnings—Cave and Quint (1990); quality. This recommendation is Talent Search—Constantine et al. (2006); Spon- targeted at assisting students who are sor-a-Scholar—Johnson (1998). interested in attending a two- or four- 103. EXCEL—Bergin, Cooks, and Bergin (2007); year institution. QOP—Schirm, Stuart, and McKie (2006); Upward Bound—Seftor, Mamun, and Schirm (2009). 104. QOP—Schirm, Stuart, and McKie (2006). 99. Cabrera and La Nasa (2000); Roderick et al. 105. EXCEL—Bergin, Cooks, and Bergin (2007); (2008). Career Beginnings—Cave and Quint (1990); Tal- 100. Choy et al. (2000); Cunningham, Erisman, ent Search—Constantine et al. (2006); Sponsor- and Looney (2007); Horn and Chen (1998); Ishitani a-Scholar—Johnson (1998); QOP—Schirm, Stu- and Snider (2004); Kao and Tienda (1998); Plank art, and McKie (2006); Upward Bound—Seftor, and Jordan (2001); Venezia and Kirst (2005). Mamun, and Schirm (2009).

( 31 ) Recommendation 4. Engage and assist students in completing critical steps

programs helped students with college and take the practice exams by 11th grade, entrance exams by offering exam prepara- and the actual exam before 12th grade. tion classes, providing waivers for exam Students who wait until their senior year fees, or encouraging students to take the to take the actual exam could miss a col- exams.106 All but one of the programs of- lege application deadline or not have an fered students the opportunity to visit opportunity to retake the test. college campuses.107 Although there was limited information to assess the nature of To ensure that students take the exams these visits, they were a common feature at the appropriate times, the panel rec- of these programs. Most of the programs ommends that schools clearly communi- offered hands-on assistance with the col- cate the timeline for the testing schedule, lege application process. For example, Tal- including registration deadlines and test ent Search provided individual counseling dates, to all students (see Exhibit 3 for an and advice on the college application pro- example of an entrance exam schedule). cess and helped orient students to the col- For example, schools could communicate lege entry steps. Some Career Beginnings information about key deadlines through sites offered classes on how to complete email or phone blasts that reach all stu- college applications. QOP case managers dents, in-person visits to English classes and Sponsor-a-Scholar mentors assisted by high school staff knowledgeable about students with college applications, and the entrance tests, or by setting up infor- Sponsor-a-Scholar held workshops for stu- mation tables at athletic events. School dents on the application process. staff can assist students with registering for the test, remind them about the testing How to carry out this schedule or other deadlines, or offer as- recommendation sistance with fee waivers.108 For example, teachers in homeroom classes or study 1. Ensure students prepare for, and take, the periods could encourage students to reg- appropriate college entrance or admissions ister for entrance exams and discuss how exam early. to prepare for the exams, such as getting a good night’s sleep and arriving to the College entrance exams, both the practice exam early. exams and actual exams, represent a potential barrier for students interested in a Schools should ensure that students are four-year college. However, students may prepared for college entrance exams by not know about the exams or may not offering exam preparation classes or know how to prepare for them, and they workshops, either directly through the may not follow through in scheduling or school or by partnering with college ac- taking the exams. High schools should cess programs or other organizations.109 make sure that students interested in at- Exam preparation classes can be offered tending a four-year institution prepare for on Saturdays or during other non-school hours,110 and test preparation can include providing direct tutoring, offering prac- 106. Career Beginnings—Cave and Quint (1990); Talent Search—Constantine et al. (2006); Spon- tice tests, or using training software for sor-a-Scholar—Johnson (1998); QOP—Schirm, Stuart, and McKie (2006); Upward Bound—Seftor, Mamun, and Schirm (2009). 108. Calahan et al. (2004); Maxfield et al. (2003). 107. EXCEL—Bergin, Cooks, and Bergin (2007); Talent Search—Constantine et al. (2006); Spon- 109. Calahan et al. (2004); Cave and Quint (1990); sor-a-Scholar—Johnson (1998); QOP—Schirm, Johnson (1998); Maxfield et al. (2003); Myers et Stuart, and McKie (2006); Upward Bound—Seftor, al. (2004). Mamun, and Schirm (2009). 110. Cave and Quint (1990); Maxfield et al. (2003).

( 32 ) Recommendation 4. Engage and assist students in completing critical steps

Exhibit 3. Example of a college entrance exam schedule SAT Test Dates for 2008/09 Nearby Test Test Date Test Main Deadline Late Fee Deadline Locations October 4, 2008 SAT I & II September 9, 2008 September 16, 2008 High Schools A & B November 1, 2008 SAT I & II September 26, 2008 October 10, 2008 High Schools A & C December 6, 2008 SAT I & II November 4, 2008 November 18, 2008 High Schools B & C January 24, 2009 SAT I & II December 26, 2008 January 6, 2009 High Schools A & B March 14, 2009 SAT I February 10, 2009 February 24, 2009 High Schools A & C May 2, 2009 SAT I & II March 31, 2009 April 9, 2009 High Schools B & C June 6, 2009 SAT I & II May 5, 2009 May 15, 2009 High Schools A & B

ACT Test Dates for 2008/09 Test Date Main Deadline Late Fee Deadline Nearby Test Locations September 13, 2008 August 12, 2008 August 22, 2008 High Schools A & B October 25, 3008 September 19, 2008 October 3, 2008 High Schools A & C December 13, 2008 November 7, 2008 November 20, 2008 High Schools B & C February 7, 2008 January 6, 2008 January 16, 2008 High Schools A & B April 4, 2008 February 27, 2008 March 13, 2009 High Schools A & C June 13, 2008 May 8, 2009 May 22, 2009 High Schools B & C Source: Adapted from materials created by a National College Advising Corps program site.

test preparation.111 If the school partners reer interests and future plans, encourag- with another organization to provide exam ing students to consider factors such as: prep classes, an academic coordinator can arrange the promotion, enrollment, and • Geography/location content of the class.112 • Tuition cost • Financial aid 2. Assist students in their college search. • School size • Admission requirements Students should receive assistance in find- • Retention rates ing a postsecondary program that matches • Demographics their qualifications, interests, and goals.113 • Available majors Schools should set up one-on-one meetings with students to discuss the types of Schools should help students understand schools that are a good fit for them to con- how to gather information on colleges sider and submit applications. School staff that will help in their search. For exam- should help students coordinate their ca- ple, students can be given an assignment that requires them to fill in information about a college that can be obtained from 111. Maxfield et al. (2003). the college’s website. The assignment can 112. Johnson (1998). be used as a starting point for a discus- 113. Constantine et al. (2006); Johnson (1998); sion about the type of information stu- Mehan (1996); St. John et al. (2002). dents should be seeking, where they can

( 33 ) Recommendation 4. Engage and assist students in completing critical steps

Exhibit 4. Example of a college visit schedule

Time Activity Description

9:00am Arrive at college campus

9:30am–10:30am Group information session College admissions staff provide information about the college and a summary of the college search and application process and financial aid.

10:45am–11:45am Small group tours Small group tours led by current college students (each group includes a high school chaperone).

12:00pm–1:00pm Lunch in college dining hall High school students have lunch with student tour guides and college admissions staff and have an opportunity to ask additional questions.

1:00pm–2:00pm Observe college class Small groups of students visit college classrooms.

2:30pm Leave college campus

Source: Adapted from materials created by a National College Advising Corps program site.

find that information on college websites, students to college and the college environ- and how to compare different colleges. ment, inform students about the college Students should also be guided to books application and selection process, and help or websites, such as the U.S. Department them consider different college options. of Education’s College Navigator website These trips should be more than a campus (http://nces.ed.gov/collegenavigator/) or tour—students should have a chance to database of accredited institutions (http:// explore campus resources, observe cam- ope.ed.gov/accreditation/), that can assist pus life, and interact with college students them in searching for colleges.114 Students (see Exhibit 4 for an example of a college should be encouraged to apply to mul- visit schedule). For example, students can tiple colleges to help them find a match shadow college students, possibly alumni school,115 or to consider applying to a from their high school, throughout their backup college to which they are more day, attending classes, eating lunch, and confident about being admitted. walking around campus together.116

3. Coordinate college visits. High schools should contact the admissions office of a college to set up meet- The panel recommends that high schools ings that allow students to hear from an organize trips for students to visit col- admissions officer, a college professor, lege campuses. These visits can introduce and a panel of students. Student groups on campus that represent the background or culture of high school students, such 114. Other useful resources include the Col- as a Latino student group, can provide a lege Board Matchmaker (http://collegesearch. collegeboard.com/search/adv_typeofschool. useful perspective on college and campus jsp) or the Sallie Mae College Answer (http:// life.117 A high school can plan activities www.collegeanswer.com/selecting/schoolcost/ that allow students to interact with cam- isc_index.jsp). These and other college plan- pus resources, such as visiting a science ning sites are accessible through the Pathways to College Network College Planning Resource laboratory or using the college library. Directory (http://www.pathwaystocollege.net/ collegeplanningresources/). 116. Calahan et al. (2004). 115. Roderick et al. (2008). 117. Gandara (2004).

( 34 ) Recommendation 4. Engage and assist students in completing critical steps

Students also should be encouraged to occur over the course of the year (see Ex- participate in college access programs or hibit 5 for an example of a timeline).120 summer activities that facilitate overnight Schools can provide a handout that lists stays on a campus.118 the key dates that students need to consider for the application process in their 4. Assist students in completing college junior and senior years.121 The compo- applications. nents of a timeline could include college entrance exams, college applications, the Without assistance, students may not in- Free Application for Federal Student Aid vest sufficient time in completing college (FAFSA) and state financial aid forms, ad- applications or may not devote attention mission acceptances, and financial aid and to key details. Schools should provide housing acceptances. The timelines can students who plan to attend a four-year be formatted so that students can enter college with hands-on assistance in com- in dates specific to the colleges they are pleting their college applications. High considering. Schools can use the timelines schools should work with students to en- in one-on-one meetings with students to sure that their applications are complete, track student progress in meeting key submitted by deadlines, and (if applicable) deadlines. Schools should make sure that of sufficient quality for acceptance. Be- all informational materials about college cause each student’s needs and interests applications are available in multiple lan- are unique, the panel recommends that, guages and formats. to the extent possible, school staff provide assistance to students one-on-one In providing one-on-one assistance, high or during small workshops or classes de- schools can help students fill out an appli- signed to assist students with completing cation, coach students on how to write a college applications, writing application college application essay, and discuss col- essays, or reminding them about applica- lege application fee waivers. Schools should tion deadlines.119 encourage students who need letters of recommendation to request these letters in The panel suggests that schools develop advance of the deadline to ensure that they mechanisms for clearly communicating can be completed on time. Since much of timelines for application milestones that the communication with college admissions offices is electronic, schools should ensure 118. Bergin, Cooks, and Bergin (2007); Cave and that students create or designate an email Quint (1990); Myers et al. (2004). account early in the application process to 119. Bergin, Cooks, and Bergin (2007); Calahan use in communicating with colleges. et al. (2004); Cave and Quint (1990); Maxfield et al. (2003); Mehan (1996); Kahne and Bailey (1999); Kuboyama (2000); Seftor, Mamun, and 120. Mehan (1996). Schirm (2009). 121. Ibid.

( 35 ) Recommendation 4. Engage and assist students in completing critical steps

Exhibit 5. Example of a college admissions timeline

April 2008  Visit a college during spring vacation.

June 2008  Ask teachers for letters of recommendation before summer vacation.  Visit two colleges by the end of the month.

July 2008  Brainstorm college essay topics.  Visit two more colleges by the end of the month.

August 2008  Obtain admission applications for colleges being considered.  Write a rough draft of the college application essay.  Search for college scholarships.

September 2008  Complete a final draft of the college essay.  Check in with the high school’s College and Career Center on a regular basis.  Request that high school transcripts be sent.

October 2008  Complete college applications (or the Common Application, a general application form used by more than 150 independent colleges) by the end of the month.

November–December 2008  Early action or early decision deadline for some colleges.  Continue to search and apply for scholarships.

January 2009  Application deadline for most colleges and universities (January 1 or 15).  Contact colleges to make sure your application materials were received.  Fill out the FAFSA (released January 1).

February 2009  Complete the FAFSA prior to the deadline for most schools (February 1 or 15).  Search for scholarships at the colleges you are considering.

March 2009  Update FAFSA application, if needed.  Receive college acceptance letters.

April 2009  Attend open houses for colleges that offered admittance.

May 2009  Select a college and send a deposit to the school.  Request final high school transcripts be sent.

Source: Timeline adapted from materials created by a National College Advising Corps program site and an application timeline created by Sallie Mae at www.salliemae.com/before_college/students_ plan/select_school/getting_in/understanding/application_timeline.htm.

( 36 ) R meCOm eNDATION 4. Engage and assist students in completing critical steps

Potential roadblocks and solutions to college at the beginning of the year and provide follow-up information at critical Roadblock 4.1. Our counselors have large points throughout the year, such as when caseloads, making it difficult for them to college applications are typically due. help each student fill out applications or Schools may want to consider providing register for tests. guidance counselors with extended professional development opportunities to Suggested Approach. Guidance counsel- update their knowledge about the college ors do not need to be the only resource application process. Schools also should for helping students with critical steps to provide a workshop or other training for college entry. Schools can consider part- teachers on how to write effective letters nering with college access programs that of recommendation. provide adults to assist students. Schools also can invite volunteers from the com- Roadblock 4.4. Parents have limited munity, including college students, gradu- time to participate in college visits orga- ate students, or other college graduates to nized by the school because of their work assist students with the process. Another schedules. potential source of volunteers is retired professionals who previously worked in Suggested Approach. Although involving college admissions or who have knowledge parents in college visits can be difficult be- of the college entry process. Schools also cause of conflicts with work schedules, stu- can ensure that teachers have adequate dents and parents should be encouraged to knowledge about the college application visit a local college campus at their conve- process to assist students who approach nience. These visits can be less formal than them with questions. the school-organized trips, but they should provide an opportunity for parents to ex- Roadblock 4.2. The time required to travel plore college life on campus with their child. to SAT/ACT test prep sites is a barrier for High schools can provide parents with a col- students because the sites are located so lege trip itinerary that includes key places far away. to visit on campus and contact information for the college admissions office. Suggested Approach. Consider making the school a test prep site. This might Roadblock 4.5. College visits require staff make it more convenient for the students time and funding that are not available at to enroll in test prep classes and might our school. give students more incentive to register for and take the entrance exam. Suggested Approach. High schools should contact the admission offices of local col- Roadblock 4.3. Our staff do not have leges to determine whether there are avail- current information about college able resources to support college visits. requirements. High school alumni who attended a local college may be willing to sponsor a trip for Suggested Approach. Teachers and coun- students or may be able to link the school selors may be the first people whom stu- to other alumni who can assist with the dents turn to with questions about the college visit. High schools also can collabo- college application process. Schools may rate with college access programs that have want to provide teachers with a short over- funding to support college visits and can view of the critical milestones for applying coordinate trips.

( 37 ) Recommendation 5. Level of evidence: Moderate

Increase families’ The panel judged the level of evidence financial awareness, for this recommendation as moderate be- and help students cause there is evidence that the recommended practices increased the financial apply for financial aid assistance application rate and the college enrollment rate. Two programs with studies meeting standards with reserva- Financial aid plays an important role tions informed students about financial in making college affordable and aid opportunities and provided hands-on improving access to college, especially assistance in completing financial aid ap- for first-generation students and plications.124 Both of these programs had students from low-income families. a positive impact on financial aid applica- However, these students and their tions and college enrollment. Although families often have limited knowledge five other programs with studies meeting of financial aid opportunities and may standards with and without reservations overestimate the cost of college.122 provided financial aid services, the level This can create the impression that a and intensity of these services were often college education is out of reach and unclear.125 Since the two programs that might discourage students from taking closely correspond with the panel’s rec- the necessary steps in high school to ommendation had a positive impact on prepare for college. Even when parents financial aid application and college en- and students are aware of financial rollment, the panel assigned a moderate aid opportunities, the financial aid level of evidence. application process can be a barrier, with complex rules and complicated Brief summary of evidence to forms that can create additional support the recommendation roadblocks for students.123 The types of practices recommended by the panel formed a major part of Talent High schools can ensure that students Search and the FAFSA Experiment, and take the necessary steps to obtain both had a statistically significant im- financial aid by educating students and pact on application for financial aid and their parents early in high school about postsecondary enrollment.126 Almost all college affordability and the availability Talent Search projects provide individ- of financial aid and by helping them ual financial aid counseling, financial aid identify potential sources of aid. Students workshops for students and/or parents, benefit from hands-on assistance in assistance with financial aid applications, meeting financial aid deadlines and and scholarship searches.127 The FAFSA completing application forms. 124. FAFSA Experiment—Bettinger et al. (2009); Talent Search—Constantine et al. (2006). 125. EXCEL—Bergin, Cooks, and Bergin (2007); Career Beginnings—Cave and Quint (1990); Spon- sor-a-Scholar—Johnson (1998); QOP—Schirm, 122. Grodsky and Jones (2007); Horn, Chen, and Stuart, and McKie (2006); Upward Bound—Seftor, Chapman (2003); King (2006); MacAllum et al. Mamun, and Schirm (2009). (2007); Tomas Rivera Policy Institute (2004). 126. FAFSA Experiment—Bettinger et al. (2009); 123. Dynarski and Scott-Clayton (2007); Roder- Talent Search—Constantine et al. (2006). ick et al. (2008). 127. Calahan et al. (2004).

( 38 ) Recommendation 5. Increase families’ financial awareness

Experiment compared two financial aid How to carry out this interventions. The first had a tax profes- recommendation sional provide one-on-one assistance in filling out the FAFSA based on a family’s 1. Organize workshops for parents and stu- income tax return, provide information on dents to inform them prior to 12th grade the family’s financial aid eligibility for local about college affordability, scholarship and colleges, and offer to submit the FAFSA at aid sources, and financial aid processes. no cost. A second intervention provided families only information on their finan- High schools should inform students and cial aid eligibility. The first intervention parents about financial aid and the cost had a positive impact on the proportion of of college early in high school. The panel individuals who submitted the FAFSA and recommends that high schools organize enrolled in college, whereas the second separate workshops to inform parents and had no impact on either outcome. students about financial aid.130 The workshops should address misconceptions Five other programs with studies meet- about college costs and build awareness ing standards provided assistance with of financial aid. The panel recommends the financial aid application process: Up- holding an initial workshop on college ward Bound, Career Beginnings, Sponsor- affordability in 9th or 10th grade, ensur- a-Scholar, EXCEL, and QOP. Upward Bound ing that students and parents understand and Career Beginnings had studies measur- the cost of college and the aid available to ing financial aid outcomes, but neither had make it affordable. A workshop on schol- an impact on these outcomes. Both Sponsor- arship and aid sources should occur in a-Scholar and Career Beginnings had a posi- 10th grade so that students and parents tive impact on college enrollment,128 and can begin to think about the sources of the three remaining programs had no im- different forms of aid. Although students pact on enrollment.129 The level and inten- complete the FAFSA in their senior year, sity of financial aid services varied across information about the financial aid appli- Upward Bound and Career Beginnings sites. cation process should be covered in the Some sites provided financial aid counsel- junior year to prepare students for the ing and hands-on assistance with finan- process. These three financial aid topics cial aid applications, and others offered are discussed next. informational workshops for students and parents. Sponsor-a-Scholar mentors and • College affordability. Students who QOP case managers assisted students with think that college is too expensive or financial aid applications, and both pro- who lack information about the avail- grams held workshops on the financial aid ability of aid may not take the nec- process. The studies of these five programs essary steps early in high school to offered limited information on the financial prepare for college.131 The panel rec- aid services provided for students. ommends that high schools provide information about college affordability— both the cost of college and ways to cover the cost—starting in 9th grade. Schools can create a worksheet that

128. Career Beginnings—Cave and Quint (1990); 130. Advisory Committee on Student Financial Sponsor-a-Scholar—Johnson (1998). Assistance (2008); Bailis et al. (1995); Constantine 129. EXCEL—Bergin, Cooks, and Bergin (2007); et al. (2006); Cunningham, Erisman, and Looney QOP—Schirm, Stuart, and McKie (2006); Upward (2007); Seftor, Mamun, and Schirm (2009). Bound—Seftor, Mamun, and Schirm (2009). 131. Luna De La Rosa (2006).

( 39 ) Recommendation 5. Increase families’ financial awareness

displays potential costs for college they can search more broadly for next to potential sources of financial scholarships (e.g., www.fastweb.com, aid to demonstrate the realistic cost www.latinocollegedollars.org). Al- to families. Students should receive though high school advisors often information on the typical tuition cost maintain information on scholarship for two- and four-year colleges, differ- opportunities, students may not access ences between public and private insti- this information unless they regularly tutions, and tuition estimates for local visit a school’s advising office.133 High and regional colleges. High schools schools can disseminate scholarship can provide a worksheet that has a information during the workshop and side-by-side comparison of the cost follow up with updated or additional of these schools and should help stu- information on the school’s website or dents and parents distinguish the dif- in its monthly newsletter. Schools can ferent types of college costs, includ- designate a staff member to collect and ing tuition, fees, room and board, and update financial aid, scholarship, and books and supplies. grant opportunities for students.

Students need to understand the types • Financial aid application process. of financial aid available to cover these High schools should hold workshops costs, including grants, loans, scholar- to inform students and parents about ships, tax credits, and work-study pro- the financial aid application process, grams. Descriptions of financial aid, including details about the process loan obligations, and grants can be for submitting the FAFSA. Students confusing for individuals who may have should understand the information that limited interactions with banks and is needed to complete the FAFSA and lending agencies; accordingly, conversa- should know about the online and hard- tions should be developed in a manner copy versions of the application. High that is understandable to the student schools should explain that the FAFSA and his or her family. The workshops plays a role in determining eligibility should encourage students and parents for federal loans and grants as well as to estimate their financial aid eligibility state grants, scholarships, and other using a tool to forecast eligibility based forms of aid. Informing students about on FAFSA (e.g., FAFSA4caster, http:// key concepts, such as the estimated www.fafsa4caster.ed.gov).132 family contribution (EFC), can help students understand the meaning of their • Scholarship and aid sources. One financial aid materials. Students should workshop should assist students in understand the steps in the process navigating the vast array of financial that occur after submitting the FAFSA, aid sources to identify relevant op- including receipt of the student aid re- portunities. A list of available federal port and a financial aid package. and state grants and their eligibility requirements can help students de- Workshops on financial aid should be termine likely sources of aid. During held for parents as well as for students.134 the workshop, high schools also can High schools should develop a plan for provide a list of local and regional engaging parents and encouraging them sources of scholarships available for to become invested in the financial aid students, as well as websites on which 133. Luna De La Rosa (2006). 132. Advisory Committee on Student Financial 134. Calahan et al. (2004); Gandara (2004); Assistance (2008). Schirm, Stuart, and McKie (2006).

( 40 ) Recommendation 5. Increase families’ financial awareness

and college application processes. For library or computer lab so that students can example, a parent institute that includes complete the FAFSA on the Internet. sessions on financial aid and other aspects of the college entry process could be held Even though high schools can reach a throughout the school year.135 Inviting broad group of students through line- parents to informal social gatherings at by-line assistance at a workshop, stu- the school, such as picnics or family din- dents may have complex questions spe- ners, can encourage parent involvement as cific to their financial situation or may well.136 Offering child care at these events be uncomfortable raising questions at a can make it easier for parents to attend group meeting.139 Therefore, high schools and participate. The workshops for par- should provide individual assistance or ents should discuss how they can help stu- counseling following a workshop to fur- dents complete the financial aid process ther assist students in completing the and encourage them to assist students in FAFSA or other aid applications.140 For meeting key deadlines. high schools that provide mentoring services (see recommendation 3), mentors 2. Help students and parents complete finan- can provide one-on-one assistance if they cial aid forms prior to eligibility deadlines. are knowledgeable about financial aid or if they receive training.141 Individual fi- The panel recommends that high schools nancial aid counseling also can be helpful provide hands-on assistance to students for answering questions about the Stu- and parents in completing financial aid dent Aid Profile, award letter, or financial forms prior to critical deadlines. In addi- aid decisions that are made after a stu- tion to workshops providing information dent submits the FAFSA. about financial aid, high schools should hold workshops to assist high school se- Potential roadblocks and solutions niors and their parents in completing the FAFSA form, to answer student questions, Roadblock 5.1. Our school does not and to explain the information requested have staff who are trained on financial on the form.137 The workshops should in- aid policy. clude volunteers who are knowledgeable on the FAFSA and can provide one-on-one help Suggested Approach. Financial aid of- in completing the application form.138 High ficers in local colleges will be knowledge- schools should reach out to financial aid able about financial aid and can be in- officers from local colleges who can train vited to assist students during a workshop teachers or volunteers on the FAFSA and or through one-on-one sessions.142 High who can assist individual students during schools also could invite the financial aid the workshop. Students should be notified officer to train teachers on financial aid of the information needed to fill out the and the application process so that they FAFSA, such as income information from can assist students. parents’ tax forms, before the session. High schools can coordinate with the school

135. Tierney and Jun (2001); Standing et al. 139. Luna De La Rosa (2006). (2008). 140. Advisory Committee on Student Financial 136. Gandara et al. (1998). Assistance (2008); Bailis et al. (1995); Calahan et 137. Advisory Committee on Student Financial al. (2004); Maxfield et al. (2003). Assistance (2008); Cave and Quint (1990). 141. Johnson (1998); Pell Institute for the Study 138. Bailis et al. (1995); Calahan et al. (2004); of Opportunity in Higher Education (2006). Seftor, Mamun, and Schirm (2009). 142. Cave and Quint (1990).

( 41 ) R eCOmmendation 5. Increase families’ financial awareness

Roadblock 5.2. Guidance counselors may consumer science teachers may have back- not have information about college costs or grounds that are useful for understanding information about the changing nature of the financial aid process. Establishing con- college costs. tacts with financial aid staff at local colleges can make it easier for teachers to stay Suggested Approach. The panel suggests current with information on college costs. that high schools identify and train staff at The financial aid staff from local colleges the school who are willing to learn about could be useful for training teachers and financial aid and to serve as a resource other staff at the high school on financial for students. Math teachers or family aid topics.

( 42 ) Appendix A. particular types of studies for drawing causal conclusions about what works. Postscript from Thus, one typically finds that a strong the Institute of level of evidence is drawn from a body of Education Sciences randomized controlled trials, the moderate level from well-designed studies that do not involve randomization, and the low What is a practice guide? level from the opinions of respected authorities (see Table 1). Levels of evidence The health care professions have em- also can be constructed around the value braced a mechanism for assembling and of particular types of studies for other communicating evidence-based advice to goals, such as the reliability and validity practitioners about care for specific clini- of assessments. cal conditions. Variously called practice guidelines, treatment protocols, critical Practice guides also can be distinguished pathways, best practice guides, or simply from systematic reviews or meta-analy- practice guides, these documents are sys- ses such as What Works Clearinghouse tematically developed recommendations (WWC) intervention reviews or statistical about the course of care for frequently en- meta-analyses, which employ statistical countered problems, ranging from physi- methods to summarize the results of stud- cal conditions, such as foot ulcers, to psy- ies obtained from a rule-based search of chosocial conditions, such as adolescent the literature. Authors of practice guides development.143 seldom conduct the types of systematic literature searches that are the backbone Practice guides are similar to the prod- of a meta-analysis, although they take ad- ucts of typical expert consensus panels vantage of such work when it is already in reflecting the views of those serving published. Instead, authors use their ex- on the panel and the social decisions that pertise to identify the most important come into play as the positions of individ- research with respect to their recommen- ual panel members are forged into state- dations, augmented by a search of recent ments that all panel members are willing publications to ensure that the research to endorse. Practice guides, however, are citations are up-to-date. Furthermore, the generated under three constraints that do characterization of the quality and direc- not typically apply to consensus panels. tion of the evidence underlying a recom- The first is that a practice guide consists mendation in a practice guide relies less of a list of discrete recommendations that on a tight set of rules and statistical algo- are actionable. The second is that those rithms and more on the judgment of the recommendations taken together are in- authors than would be the case in a high- tended to be a coherent approach to a quality meta-analysis. Another distinction multifaceted problem. The third, which is is that a practice guide, because it aims for most important, is that each recommen- a comprehensive and coherent approach, dation is explicitly connected to the level operates with more numerous and more of evidence supporting it, with the level contextualized statements of what works represented by a grade (strong, moder- than does a typical meta-analysis. ate, or low). Thus, practice guides sit somewhere be- The levels of evidence, or grades, are tween consensus reports and meta-anal- usually constructed around the value of yses in the degree to which systematic processes are used for locating relevant 143. Field and Lohr (1990). research and characterizing its meaning.

( 43 ) Appendix A. Postscript from the Institute of Education Sciences

Practice guides are more like consensus These are people the chair believes can panel reports than meta-analyses in the work well together and have the requisite breadth and complexity of the topic that expertise to be a convincing source of rec- is addressed. Practice guides are different ommendations. IES recommends that at from both consensus reports and meta- least one of the panelists be a practitioner analyses in providing advice at the level with experience relevant to the topic being of specific action steps along a pathway addressed. The chair and the panelists are that represents a more-or-less coherent provided a general template for a practice and comprehensive approach to a multi- guide along the lines of the information faceted problem. provided in this appendix. They also are provided with examples of practice guides. Practice guides in education at the The practice guide panel works under a Institute of Education Sciences short deadline of six to nine months to produce a draft document. The expert panel The Institute of Education Sciences (IES) members interact with and receive feed- publishes practice guides in education to back from staff at IES during the develop- bring the best available evidence and ex- ment of the practice guide, but they under- pertise to bear on the types of systemic stand that they are the authors and, thus, challenges that cannot currently be ad- responsible for the final product. dressed by single interventions or programs. Although IES has taken advantage One unique feature of IES-sponsored prac- of the history of practice guides in health tice guides is that they are subjected to care to provide models of how to proceed rigorous external peer review through the in education, education is different from same office that is responsible for inde- health care in ways that may require that pendent review of other IES publications. practice guides in education have some- A critical task of the peer reviewers of a what different designs. Even within health practice guide is to determine whether care, where practice guides now number the evidence cited in support of particular in the thousands, there is no single tem- recommendations is up-to-date and that plate in use. Rather, one finds descriptions studies of similar or better quality that of general design features that permit point in a different direction have not been substantial variation in the realization ignored. Peer reviewers also are asked to of practice guides across subspecialties evaluate whether the evidence grade as- and panels of experts.144 Accordingly, the signed to particular recommendations by templates for IES practice guides may vary the practice guide authors is appropriate. across practice guides and change over A practice guide is revised as necessary to time and with experience. meet the concerns of external peer reviews and to gain the approval of the standards The steps involved in producing an IES- and review staff at IES. The process of ex- sponsored practice guide are first to select ternal peer review is carried out indepen- a topic, which is informed by formal sur- dent of the office and staff within IES that veys of practitioners and requests. Next, a instigated the practice guide. panel chair is recruited who has a national reputation and up-to-date expertise in the Because practice guides depend on the topic. Third, the chair, working in collabo- expertise of their authors and their group ration with IES, selects a small number of decisionmaking, the content of a practice panelists to co-author the practice guide. guide is not and should not be viewed as a set of recommendations that in every case 144. American Psychological Association depends on and flows inevitably from sci- (2002). entific research. It is not only possible but

( 44 ) Apendip x A. Postscript from the Institute of Education Sciences

also likely that two teams of recognized individual school district might obtain on experts working independently to produce its own because the authors are national a practice guide on the same topic would authorities who have to reach agreement generate products that differ in important among themselves, justify their recom- respects. Thus, consumers of practice mendations in terms of supporting evi- guides need to understand that they are, dence, and undergo rigorous independent in effect, getting the advice of consultants. peer review of their product. These consultants should, on average, provide substantially better advice than an Institute of Education Sciences

( 45 ) Appendix B. specialties in education, labor economics, About the authors and econometrics. Jill Constantine, Ph.D., is a senior econ- Panel omist and associate director of research at Mathematica Policy Research. Dr. Con- William G. Tierney, Ph.D., (Chair) is cur- stantine has led a number of large-scale rently University Professor and Wilbur- research and evaluation projects related Kieffer Professor of Higher Education at the to education programs, including several University of Southern California’s Rossier in the field of higher education. She was School of Education, as well as director the project director for the Evaluation of of the Center for Higher Education Policy College Corps. She also served as the dep- Analysis. His research interests pertain uty project director and, later, as project to access, issues of equity, organizational director for the U.S. Department of Educa- effectiveness, and the changing nature of tion’s National Evaluation of Talent Search. academic work. He recently concluded a Dr. Constantine also has led a number of three-year project that examined how to im- other education projects, including those prove financial aid strategies for low-income in the areas of Beginning Reading, teacher youth and their families. He also recently preparation, and Early Head Start. Dr. Con- concluded a project on how to improve gov- stantine has published extensively in the ernance in four-year colleges and universi- education field. ties. His recent books include Urban High School Students and the Challenge of Access Neal Finkelstein, Ph.D., is a senior re- (Peter Lang), Preparing for College: Nine Ele- search scientist at WestEd, developing re- ments of Effective Outreach (SUNY), and In- search and evaluation designs that study creasing Access to College (SUNY). program implementation in K–12 public schools, and overseeing randomized field Thomas Bailey, Ph.D., is the George and trials implementation in education settings. Abby O’Neill Professor of Economics and His areas of expertise include K–12 school Education in the Department of Interna- finance, academic preparation programs tional and Transcultural Studies at Teach- for high school youth, school-to-work, and ers College, Columbia University. He also early childhood education. Dr. Finkelstein is the director of the Institute on Educa- served as director of Educational Outreach tion and the Economy at Teachers Col- Research and Evaluation for the University lege. In 1996, Dr. Bailey established the of California, where he implemented re- Community College Research Center at search and evaluation designs on 10 cam- the Institute, which pursues a wide variety puses to study the effectiveness of K–12 of quantitative and qualitative research student and school academic programs, focused primarily on improving educa- particularly on postsecondary education tional outcomes for students at commu- opportunities. He also served as senior nity colleges, especially low-income and program officer for the National Research minority students. Dr. Bailey also directs Council, supporting the investigation of the National Center for Postsecondary Re- equity, adequacy, and productivity in K–12 search, funded by the Institute of Educa- public education financing. tion Sciences, which is conducting rigorous evaluations of widely used strategies Nicole Farmer Hurd, Ph.D., is the execu- for low-skilled students, including inten- tive director of the National College Ad- sive summer bridge programs and learn- vising Corps (NCAC), which is headquar- ing communities. He is an economist, with tered at the University of North Carolina

( 46 ) Appendix B: About the Authors

at Chapel Hill. The corps is a coalition Christina Clark Tuttle, M.P.P., is an ed- of university-based college advising pro- ucation researcher at Mathematica Policy grams serving more than 48,000 students Research. Her work has focused on study- in 12 states. NCAC places recent college ing opportunities for disadvantaged stu- graduates in public high schools to part- dents and school reform, contributing ner with guidance counselors in an effort to Mathematica’s evaluations of Upward to increase their college-going rates. Dr. Bound, QOP, the New York City voucher ex- Hurd, who served as an assistant dean and periment, and charter schools. Ms. Tuttle director of the Center for Undergraduate is the deputy project director of Mathe- Excellence at the University of Virginia, matica’s national evaluation of Knowledge was the founding director of the College Is Power Program. She is a WWC certified Guide Program, which served as the model reviewer and deputy principal investiga- for the Advising Corps. Dr. Hurd received tor for the high school math topic area of the Governor of Virginia’s Award for Vol- the WWC. unteerism and Community Service and was the 2007 faculty recipient of the Uni- versity of Virginia’s Raven Award. Dr. Hurd remains an administrator in higher education, serving in the Office of Undergradu- ate Admissions at the University of North Carolina at Chapel Hill.

Staff

Jeffrey Max, M.P.A., is a researcher at Mathematica Policy Research, with experience conducting quantitative and qualitative research in the field of education. He has contributed to studies of teacher quality interventions, including evaluations of teacher incentives, teacher preparation, and teacher certification. His past work includes a study in one state of the transfer process from community colleges to teacher preparation programs.

( 47 ) Appendix C. evidence that is documented in the practice guide. In addition, the practice guide Disclosure of potential undergoes independent external peer re- conflicts of interest view prior to publication, with particular focus on whether the evidence related to Practice guide panels are composed of in- the recommendations in the practice guide dividuals who are nationally recognized has been appropriately presented. experts on the topics about which they are rendering recommendations. IES expects The professional engagements reported that such experts will be involved profes- by each panel member that appear most sionally in a variety of matters that relate closely associated with the panel recom- to their work as a panel. Panel members mendations are noted below. are asked to disclose their professional involvements and to institute deliberative Dr. Nicole Farmer Hurd is the executive processes that encourage critical exami- director of the National College Advising nation of the views of panel members as Corps, a national college peer mentoring they relate to the content of the practice program. Although this program is not ref- guide. The potential influence of panel erenced in the text of the practice guide, members’ professional engagements is four of the exhibits are adapted from ma- further muted by the requirement that terials used by National College Advising they ground their recommendations in Corps sites.

( 48 ) Appendix D. determine the extent to which a program Technical information included the recommended practices. on the studies College access programs consist of multiple components that address a variety A search for research on college access of steps students must take to prepare for programs in the United States from 1988 and enter college. The bundling of multiple to 2008 resulted in more than 500 stud- practices within an access program makes ies. Of these, 99 studies had causal de- it difficult to assess the effectiveness of a signs and examined programs that aimed single practice. This presents a challenge to prepare secondary students (grades 6 for the practice guide, which aims to pro- through 12) academically for college, as- vide specific strategies that schools can sist them in completing the steps to col- implement to improve access to college. lege entry, and improve their likelihood of The panel reviewed implementation re- enrolling in college. These were reviewed ports of programs with studies meeting according to What Works Clearinghouse standards to assess the relative impor- (WWC) standards. Sixteen studies of 10 tance of each program component and to programs met WWC evidence standards better understand program implementa- with or without reservations. Correlational tion (see Table D1). The panel assigned a studies that used longitudinal surveys of level of evidence for each recommendation high school students, such as the National after considering the number of programs Educational Longitudinal Study (NELS),145 with studies meeting standards that re- to analyze the relationship between stu- lated to each recommendation; the degree dent characteristics and college outcomes to which programs implemented the rec- were not reviewed against WWC standards ommendation; and the programs’ impacts because they do not provide evidence on on high school academic performance, the effectiveness of a practice or interven- completion of the critical steps for college tion. The panel focused on the programs entry, and college enrollment. with studies meeting standards to define a level of evidence for each recommenda- The panel noted two key issues in its con- tion in the practice guide. sideration of the evidence:

An overview of the programs with studies • Service receipt by the comparison meeting standards is provided in Table D1, group. Several of the studies meeting which also shows the relevance of each standards were effectiveness studies program for the recommendations. In that compared enrollment or partici- many cases, a program implemented only pation in a college access program to one aspect of the panel’s recommendation. the existing college preparation ser- For example, several programs with stud- vices available to students.146 In some ies meeting standards implemented the cases, the comparison group received first step of recommendation 3, mentor- a fair amount of college access ser- ing services, but they did not implement vices. For example, an evaluation of steps two (peer groups) and three (career Upward Bound reported that 54 per- exploration). In addition, many programs cent of comparison group students implemented practices that did not fully align with the panel’s recommendation, or the studies provided insufficient detail to 146. In contrast to efficacy studies that show the effect of a program relative to a control that re- ceives no intervention or services, effectiveness 145. National Center for Education Statistics studies show the effectiveness of a program rela- (2002). tive to other available services or programs.

( 49 ) Appendix D. Technical information on the studies

received college access services, and had to demonstrate baseline equivalence an evaluation of Career Beginnings on at least three measures, including one noted that 35 percent of comparison measure of socioeconomic status (SES). group students received services simi- Establishing baseline equivalence on SES lar to that program’s.147 Since schools is critical because there are differences in often implement more than one college college enrollment rates between high- and access program, students in the com- low-SES students even when controlling for parison group may participate in other race and academic ability.148 Table D2 lists college access programs. Among com- the studies that potentially met standards parison students in the Upward Bound that had insufficient information to estab- study, 14 percent participated in Up- lish baseline equivalence or did not demon- ward Bound Math and Science, and 12 strate equivalence on a measure of SES. percent participated in Talent Search. Similarly, the study of Career Acad- This appendix describes the evidence used emies notes that comparison group to support each of the recommendations students had the option of enrolling made by the panel. After describing the in other similar programs offered by level of evidence for the recommendation, their high school or district. we summarize the evidence used to support each recommendation. Since several • Programs operated by postsecond- of the programs with studies meeting stan- ary institutions and community dards are relevant for multiple recommen- organizations. Although the practice dations, we provide a brief description of guide is designed to help high schools each program and its study below. and school districts take actionable steps to increase college attendance, Talent Search is a federal program that the panel notes that many of the pro- informs students about the high school grams with studies meeting standards courses needed to prepare for college and were implemented by postsecondary the availability of financial aid to cover institutions or community organiza- the cost of college. The program provides tions. Four programs were primar- hands-on assistance with financial aid ap- ily implemented by a postsecondary plications and helps students complete institution and two by nonprofit or the college application process. Talent community-based organizations. The Search projects are operated by two- and panel focused on describing the prac- four-year colleges and also provide col- tices it deemed were effective for high lege visits, academic support, and coun- school students and could feasibly be seling. A quasi-experimental study of the implemented by a high school. program in Indiana, Florida, and Texas met standards with reservations.149 The Studies that potentially study examined the cohort of students in meet standards 9th grade in the 1995/96 school year and used propensity score matching to create The panel used 15 studies that potentially a comparison group of students from the met standards but were not assigned a same schools or districts with the same WWC rating because there was insufficient demographic, socioeconomic, and aca- information to assess baseline equivalence demic characteristics as Talent Search par- of the treatment and control groups. To ticipants. The program had a statistically meet standards with reservations, a study 148. Advisory Committee on Student Financial 147. EXCEL—Bergin, Cooks, and Bergin (2007); Assistance (2006); Black and Sufi (2002). Upward Bound—Myers et al. (2004). 149. Constantine et al. (2006).

( 50 ) Appendix D. Technical information on the studies

significant impact on financial aid appli- aid eligibility and applying for financial cation and college enrollment in all three aid. An ongoing study of the intervention states. The findings for college persistence randomly assigned individuals in Ohio and were smaller than the impact on enroll- North Carolina to (1) the main treatment ment and not consistently positive across group that received assistance in complet- the three states. ing and submitting the FAFSA (an estimate of their eligibility for financial aid) and in- The Sponsor-a-Scholar program selects formation on financial aid opportunities, at-risk students from the Philadelphia (2) an information-only treatment group public school system and offers them an that received a written estimate of each opportunity to participate in a mentoring family’s financial aid eligibility (but no relationship with an adult volunteer for assistance in completing the FAFSA), or five years as well as $6,000 for college- (3) a control group that received an exist- related expenses. Mentors are expected to ing booklet on the importance of college meet with students at least monthly, and and financial aid programs.152 Staff from a program coordinator monitors mentor H&R Block conducted the intervention, relationships. In addition, the program providing services to individuals who re- provides academic support and assistance ceived tax preparation assistance from the with the college application and financial company. The most recent study reported aid processes, including SAT prep classes, initial findings within a year of the inter- college visits, and workshops on financial vention. For dependent participants, who aid; selecting a college; and preparing were primarily high school seniors, the for the challenges of college. The study main treatment had a significant positive of Sponsor-a-Scholar matched a total of impact on submission of the FAFSA and 180 participants across four cohorts to a college enrollment, whereas the informa- comparison group of Philadelphia public tion-only treatment did not significantly school students based on race, gender, impact either outcome. school, and grade point average (GPA).150 The author found that the program in- Career Beginnings is a collaboration creased college attendance in the first two among local colleges, public schools, and years after high school but had no impact the business community that offers sum- on college retention.151 mer jobs, education services (e.g., tutoring), assistance in preparing for college, The FAFSA (Free Application for Federal and adult mentors. The program enrolls Student Aid) Experiment offered an in- students in their junior year of high school tervention to assist low-income families and holds workshops on preparing for col- with obtaining information on financial lege, taking college entrance exams, and completing college admissions forms. A 150. Johnson (1998). study of seven Career Beginnings sites con- 151. College enrollment in the second year after ducted in the second year of the program high school was significant at the 0.10 level. The randomly assigned 1,233 eligible students What Works Clearinghouse (WWC) adjusts for to Career Beginnings or a control group.153 multiple comparisons when a study examines The study did not find a statistically sig- many outcomes simultaneously since the statistical significance of findings may be overstated. nificant impact on enrollment in two-year However, an adjustment for multiple compari- colleges or enrollment in four-year colleges, sons could not be conducted for this study be- but it reported a significant impact when cause there was insufficient information to calcu- these two outcomes were combined (i.e., late an effect size and measure significance based on WWC standards (the study did not include the mean outcomes and standard deviations for the 152. FAFSA Experiment—Bettinger et al. (2009). treatment and control groups). 153. Cave and Quint (1990).

( 51 ) Apendip x D. Technical information on the studies

enrollment in two- and four-year colleges). tative sample of projects.154 At each site, The program did not have a significant ef- applicants were randomly assigned to the fect on scholarship receipt, taking out a program or to a control group, and the student loan, or college persistence. most recent follow-up study measured impacts seven to nine years after the ex- Talent Development High School is a pected date of high school graduation. high school reform model for high schools This study reports no statistically sig- struggling with poor student attendance, nificant impacts on high school outcomes discipline problems, low student achieve- (i.e., completion of course credits, GPA, ment, and high dropout rates. The model or diploma), financial aid receipt, college consists of Ninth Grade Success Academies enrollment, college selectivity, or number that organize first-year high school stu- of college credits earned. The program in- dents into self-contained learning commu- creased the rate of postsecondary enroll- nities, and Career Academies for students ment and the likelihood of receiving a de- in grades 10 through 12 that form learning gree, license, or certificate among students communities around a career theme. The with lower educational expectations. main components of the model include a college preparatory curriculum; extended Career Academies are a high school class periods; an after-school program for reform initiative in which students are students with severe attendance and disci- organized into small learning communi- pline problems; and partnerships between ties around a particular career theme. schools, families, and communities. Kem- The academies combine academic and ple et al. (2005) conducted an evaluation technical curricula and partner with local of the model using a quasi-experimental employers to provide career exploration design that matched five Talent Develop- and work-based learning opportunities ment High Schools in the Philadelphia pub- for students. Kemple (2000, 2001, 2004, lic school system to similar high schools 2008) conducted a random assignment based on race/ethnic composition and evaluation of about 1,400 students in nine promotion rates. The program had sig- schools within large, urban districts serv- nificant positive impacts on dropping out ing a disadvantaged student population. of school and the number of high school The author reports that the program did course credits earned. not have an impact on high school graduation or college enrollment, persistence, Upward Bound is a long-standing federal or degree completion eight years after program designed to assist low-income students’ expected graduation. students in preparing for, enrolling in, and succeeding in college. Projects are hosted The Quantum Opportunities Program primarily by four-year colleges and offer (QOP) is an intensive program that offers academic courses throughout the school case management, mentoring, tutoring, and year and during an intensive six-week other education and support services for summer session often held on a college high school students. Students are matched campus. Other program services include to a case manager in 9th grade and can re- tutoring, preparation for college entrance ceive services for up to five years, even if exams, extracurricular activities, college they drop out of school or move to another tours, and financial aid workshops. Three district. Students receive financial incen- reports cover the evaluation of Upward tives for participating in program activi- Bound and include a nationally represen- ties. The random assignment study of QOP

154. Myers and Schirm (1999); Myers et al. (2004); Seftor, Mamun, and Schirm (2009).

( 52 ) Apendip x D. Technical information on the studies

by Schirm et al. (2006) included 1,069 stu- 70 students by Bergin et al. (2004) found dents in seven sites and found no impact that the program did not have an impact on attainment of a high school diploma or on college enrollment. General Education Development (GED) test, enrollment in college or vocational training, Middle College High Schools are alter- persistence in college, or earning a bach- native high schools designed to help stu- elor’s or associate’s degree. dents who have dropped out or are close to dropping out of high school remain in EXCEL is a scholarship incentive and school and earn high school diplomas. support program sponsored by a public The schools are located on a college cam- university in the Midwest that provides pus and emphasize experiential learning, academic enrichment activities through employ a thematic curriculum, offer team summer institutes and weekend semi- teaching, and provide additional support nars on the university’s campus. Students services. Dynarksi et al. (1998) evaluated receive a scholarship to the sponsoring a Middle College High School program in institution that covers tuition, fees, and Seattle, Washington, using a randomized books if they complete a college prepara- controlled trial. Students were randomly tory curriculum, maintain a B average in assigned to Middle College High School high school, participate in program activi- or to a control group condition in which ties, and earn a score of 18 on the ACT. they attended their traditional local high The program assists students through- school. The study found no impacts on out high school and enlists parental par- dropping out of school or the attainment ticipation and commitment. A relatively of a high school diploma or GED. small random assignment study of about

( 53 ) Appendix D. Technical information on the studies X X X X X 5. Aid Financial Financial (continued) X X X X 4. Entry College X X 3. Networks Recommendations X X X 2. Assess X X X X 1. Courses -

(1995) (2004) Kemple Calahan Calahan Relevant et al.et (2004) and Herlihy tion Studies Implementa Hershey al. et

Target rice lunch students Students dropouts academic discipline, discipline, students in p Students in Population urban areas urban Low-income Low-income schools that that schools achievement achievement achievement with average with average achievement, achievement, Students who and dropouts have difficulty difficulty have families with a with families and eligible for are high school or a recent high free or reduced- or free potentially first- Low-income andLow-income with attendance, school graduate Minority students dropouts or likely generation college high school senior f d 8–12 9–12 9–12 8–12 9–12 11–12 Grade Grades Grades Grades Grades Grades Grades Levels Grade 12 Program

Study Florida, Florida, Indiana, Midwest and Ohio Ohio and Location and Texas city in the California, California, Seattle, WA Seattle, states: New Seven of 24 Philadelphia Philadelphia Charlotte, NC program sites York, Indiana, York, Medium-sized Medium-sized in the following following the in g e

c a 73 394 Size 866 1,233 43,414 Sample 10,297 to 10,297 Analysis (students) 139 to 401 139 11 schools Program and Study Details and Study Program

RCT RCT RCT QED QED Study Quasi- Design controlled trial (RCT) Randomized design (QED) experimental

- Free Free High High Talent EXCEL Talent Career School Search Middle Middle College Scholar Student Schools Develop Program ment Highment Sponsor-a- for Federal Beginnings Aid (FAFSA) Application Experiment

emple Constantine et al. (2006) Cave al. et (1990) Citation Brief Johnson (1998) et al. (2005) K Positive effects on one or more relevant outcomes Bettinger et al. (2009) Dynarksi Dynarksi et al.et (1998) No detectable effects on relevant outcomes Bergin al. et (2007) Table D1. Studies of college access programs that met WWC standards with or without reservations without or with WWC standards met that programs access college of Studies D1. Table

( 54 ) Appendix D. Technical information on the studies - - X X 5. Aid Financial Financial X X 4. Entry College X X X 3. Networks Recommendations X 2. Assess X X 1. Courses -

(1999) Moore Maxfield Relevant et al.et (1997) et al.et (2003) Kemple andKemple Rock (1996); (1996); Rock Kemple al. et tion Studies Implementa Target students Population Low-income Low-income dropout rates urban schools and potentially first-generation first-generation students in large, in students schools with high At-risk students in 9–12 9–12 9–12 Grade Grades Grades Grades Levels Program

Study projects Houston, Columbia Location Memphis, sample of Maryland, Maryland, District of Columbia, Nationally District of Cleveland, Cleveland, Fort Worth, Nine Career Philadelphia, Philadelphia, Six QOP sites: Pennsylvania, Florida, Texas, representative California, and California, Upward Bound and Yakima, WA Academiessites: h Size 1,458 1,069 2,292 Sample Analysis (continued) (students) Program and Study Details and Study Program RCT RCT RCT Study Design

(QOP) Bound Career Upward Program Quantum Program Academies Opportunity

emple (2000, Brief Citation 2001, 2004,2001, 2008) No detectable effects on relevant outcomes K Schirm al. et (2006) Seftor al. et Myers (2009); al.et (2004); Myers and Schirm (1999); Myers al. et (1997) . Sample size varies by outcome. lies with a high school senior; a recent high school graduate; or independent an older, adult who was enrolled in college or who wanted enroll to in college. The total analysis individuals. sample included 16,745 Theb. study also targeted low-income families with an independent adult who was enrolled in college or who wanted enroll to in college. the However, review of the study for this prac tice guide focused on the subgroup of dependent participants. c. Sample size varies by state and analysis. had Texas a within-school analysis sample and of 34,869 an across-school analysis sample of 34,346. Indiana had a sample for theFlorida within-school size of 10,927. had a sample size of 43,414 analysis,for the across-school and 14,721 analysis. Searchd. Talent can start as early as 6th grade, but the study focuses on 9th graders. e a. This represents the sample size for the subgroup relevant this to practice guide: dependent participants who were mostly high school seniors. The study targeted low-income fami Studentsf. receive services through the first year after high school graduation. g. The authors used individual student data but did not report the number of students in the sample. All outcome measures had schools, a sample size of 11 but some measures included more cohorts of students than did others. h. Samples size varied for different outcomes. This represents the samplefor size the analysis of college enrollment and financial aid. Table D1. Studies of college access programs that met WWC standards with or without reservations (continued) reservations without or with WWC standards met that programs access college of Studies D1. Table

( 55 ) Appendix D. Technical information on the studies

Table D2. Studies of college access programs that potentially met WWC standards

Program and Study Details Recommendations Brief Program Analysis Study 1. 2. 3. 4. 5. Citation Sample Location Courses Assess Networks College Financial Size Entry Informa- tion Insufficient information to establish baseline equivalence Allensworth et College Between 21,587 Chicago, IL X al. (2008) preparatory and 26,197 coursework students per requirement cohort (10 cohorts) Attewell and Advanced Between 544 and National X Domina (2008) high school 3,279 students curriculum depending on the analysis Crook (1990) College Now 926 studentsa City X University of New York Dougherty et Advanced Between 4,959 and Texas X al. (2006) Placement (AP) 41,458 students courses per subgroupb Gandara (2002, High School 144 students California X X 2004); Gandara Puente high schools et al. (1998) Hargrove et al. AP Courses Between 38,907 Texas public X (2008) and 42,199 high schools students per cohort (4 cohorts) Howell, California Early More than 1 CSU– X Kurlaender, Assessment million students Sacramento and Grodsky Program students (2009) Jeong (2009) AP courses Between 12,130 National X and 12,870 students depending on the analysis Keng and Dodd AP courses/ Varies by analysis University X (2008) exams and cohortc of Texas at Austin Moreno (2002) Puente 62 students California X X (31 matched pairs) high schools Opuni (1999) Project GRAD 555 students for Houston, TX X math analysis; 547 students for reading analysisd Standing et al. GEAR UP 2,687 students National X X (2008) (administrative record analysis) and 2,578 students (survey analysis) Watt et al. AVID 20 high schools Texas X X (2006)

a. Most analyses use a sample of 926 students. One analysis of enrollment has a larger sample of 25,399 students, and another analysis has a sample of 743 students. b. All analysis is by subgroup, resulting in varying sample sizes. c. Sample sizes range across four cohorts and across 10 different AP subjects. d. These sample sizes are for the specific analyses in this study eligible for a WWC review.

( 56 ) Appendix D. Technical information on the studies

Recommendation 1. limited evidence for this recommendation, Offer courses and curricula that the panel believes that offering the courses prepare students for college-level needed to prepare for college and inform- work, and ensure that students ing about those courses are critical steps understand what constitutes a for improving college access. college-ready curriculum by 9th grade Summary of evidence Level of evidence: Low The panel reviewed 25 studies related to Despite the importance of this recom- college readiness and curricula. Of these, mendation, the level of evidence that sup- six were related to curricular offerings, ports it is low by WWC standards. This is none of which provided sufficient evidence primarily a function of the challenge of to assign a WWC rating but could poten- conducting research on course taking in tially meet standards.159 Five of the college any experimental or quasi-experimental readiness studies examined the effects of capacity. More than perhaps any other the Advancement Via Individual Determina- recommendation in this guide, the related tion (AVID) program specifically: of these, evidence is subject to major concerns four did not meet standards,160 and the about selection bias: are the differences in other could potentially meet standards.161 observed outcomes between students who The fourteen remaining studies addressed do and do not enroll in a given course at- elements of academic and curricular advis- tributable to the effects of that course or to ing. Of these, studies of six programs met some unobservable characteristics—moti- WWC standards with or without reserva- vation, for example—that lead students to tions, and two of these had a positive ef- take that course in the first place? fect on college enrollment.162 Three did not meet evidence standards, and the remain- The evidence for taking a college-ready ing five could potentially meet standards. curriculum consists of six studies that potentially met standards.155 Two of the studies provide mixed evidence on the effect Mixed evidence on implementing a of a rigorous high school curriculum,156 college-ready or college prep curriculum and four studies show positive effects of Advanced Placement (AP) courses.157 The The panel drew on an extensive body of evidence for academic advising is stronger, correlational work to reinforce its opinion with six relevant programs that had studies meeting standards, but the impact of aca- Sponsor-a-Scholar—Johnson (1998); Upward demic advising could not be isolated from Bound—Myers et al. (2004); QOP—Schirm, Stuart, other program components.158 Despite the and McKie (2006). 159. Allensworth et al. (2008); Attewell and Domina (2008); Dougherty, Mellor, and Jian 155. Allensworth et al. (2008); Attewell and (2006); Hargrove, Godin, and Dodd (2008); Jeong Domina (2008); Dougherty, Mellor, and Jian (2009); Keng and Dodd (2008). (2006); Hargrove, Godin, and Dodd (2008); Jeong 160. Bailey (2002); Black et al. (2008); Lozano, (2009); Keng and Dodd (2008). Watt, and Huerta (2009); Watt, Huerta, and Lo- 156. Allensworth et al. (2008); Attewell and zano (2007). Domina (2008). 161. Watt et al. (2006). 157. Dougherty, Mellor, and Jian (2006); Har- 162. EXCEL—Bergin, Cooks, and Bergin (2007); grove, Godin, and Dodd (2008); Jeong (2009); Talent Search—Constantine et al. (2006); Mid- Keng and Dodd (2008). dle College High School—Dynarski et al. (1998); 158. EXCEL—Bergin, Cooks, and Bergin (2007); Sponsor-a-Scholar—Johnson (1998); Upward Talent Search—Constantine et al. (2006); Mid- Bound—Myers et al. (2004); QOP—Schirm, Stuart, dle College High School—Dynarski et al. (1998); and McKie (2006).

( 57 ) Apendip x D. Technical information on the studies

that schools must offer courses and cur- equivalence between its treatment and ricula that prepare students for college. comparison groups, given that the sam- The panel identified five studies using data ples are constructed from similar groups from the National Center for Education of students at the same schools at differ- Statistics (NCES)—typically, NELS, the Ed- ent periods of time, it is likely that they ucation Longitudinal Study (ELS), or High would be similar on observable measures. School and Beyond (HSB)163—that showed a The study finds that although more stu- positive correlation between course taking dents completed the 9th grade with cred- and outcomes related to this guide.164 Ad- its in algebra and English, there was no ditionally, six recent quasi-experimental effect on test scores, dropout rates, or the studies were identified and reviewed— likelihood of entering college, and fail- two on rigorous curricula and four on AP ure rates increased. Attewell and Domina specifically—that provide more caution- (2008) use NELS data to demonstrate that ary evidence (and even some unintended taking a more intense curriculum in high negative consequences) and could poten- school has positive effects on 12th-grade tially meet standards.165 Also, five stud- test scores and probabilities of entering ies of the AVID program were reviewed and completing college. Although the au- and demonstrated positive effects of the thors used propensity score techniques program, which mandates enrollment in to match samples of students, evidence of the college prep track for all students, but equivalence between the groups was not most of these did not meet standards and presented for each outcome. one provided insufficient evidence to be assigned a rating (but could potentially Although AP has been extensively studied, meet standards).166 few evaluations use a comparison group design to estimate the effect of AP course Two studies with quasi-experimental de- taking on college outcomes. The panel iden- signs (QEDs) were identified that studied tified four QEDs that could not be assigned the effect of requiring students to en- a rating because they did not provide infor- roll in a college prep curriculum.167 Most mation on the equivalence of the treatment recently and notably, Allensworth et al. and comparison groups. Dougherty’s (2006) (2008) use an interrupted time-series co- student-level analysis indicates that in all hort design to measure the effects of the considered subgroups in Texas (Blacks, policy in Chicago that ended remedial Hispanics, Whites, low-income students, classes and required college prep course- and non-low-income students), the rate of work for all students in the district. Al- bachelor’s degree attainment in each of the though the study does not demonstrate three treatment groups (college-going high school graduates who pass one or more AP exams, those who take but do not pass 163. National Center for Education Statistics (1992, 2002, 2006b). any AP exams, and those who take one or more AP courses but do not take any 164. Adelman (1999, 2006); Gamoran and Hanni- gan (2000); Lee and Ready (2009); Lee, Croninger, AP exams) is significantly higher than the and Smith (1997). rate of bachelor’s degree attainment in the 165. Allensworth et al. (2008); Attewell and control group (college-going high school Domina (2008); Dougherty, Mellor, and Jian (2006); graduates who do not take any AP courses Hargrove, Godin, and Dodd (2008); Jeong (2009); or exams), but no evidence is provided on Keng and Dodd (2008); Spielhagan (2006). the equivalence of those groups although 166. Bailey (2002); Black et al. (2008); Lozano, Watt, and Huerta (2009); Watt, Huerta, and Lo- they are matched on SES and 8th-grade zano (2007); Watt et al. (2006). test scores. Similarly, using ELS data, Jeong 167. Allensworth et al. (2008); Attewell and (2009) finds that enrollment in AP course- Domina (2008). work in 11th or 12th grade is associated

( 58 ) Appendix D. Technical information on the studies

with a significantly higher probability of showed increases in enrollment in courses completing high school within four years of high rigor and in high school graduation after the sophomore year of high school by the authors’ calculations. and a significantly higher probability of entering a four-year college within four years after the sophomore year of high Limited evidence on academic advising school. Hargrove (2008) reports that students who took both the AP course and the The evidence to support the panel’s sug- AP exam had higher college GPAs, earned gestion that schools promote comprehen- more credits, and had higher graduation sion of what constitutes a college-ready rates than did students who took only the curriculum and develop a four-year course AP course or a non-AP course in the same trajectory with each 9th grader that leads subject area. Keng and Dodd (2008) find to a fulfilled college prep track is stronger that students who take AP courses and re- but not sufficient for a moderate rating. ceive college credit for their exam scores The panel reviewed 13 empirical studies have better college outcomes than do stu- with causal designs located in the WWC lit- dents of similar abilities who do not par- erature search, including six that met stan- ticipate in AP, but again, no information is dards with or without reservations,168 four provided to enable the panel to determine that could potentially meet standards,169 the equivalence of those groups. and three that did not meet standards.170 Of the six that met WWC standards,171 the AVID is an academic intervention that tar- evaluations nevertheless could not isolate gets students “in the academic middle”— the effect of the advising practice from those who are potentially first-generation other aspects of the intervention. college students earning average grades— and encourages them to enroll in the col- Talent Search provides its students with lege prep curriculum at their school, in- information about the types of courses cluding honors and AP classes. AVID places they should take to prepare for college. In low-track students in high-track classes and Sponsor-a-Scholar, the mentor and a class provides support for the students to tran- coordinator work with students and school sition to college. AVID works hard to place staff to ensure that students are enrolled in its students in college prep courses. It also a college preparation course of study. Two provides students with exposure to an aca- quasi-experimental studies of the programs demic environment similar to that found in college classrooms. College entry skills and 168. EXCEL—Bergin, Cooks, and Bergin (2007); academic survival skills including study Talent Search—Constantine et al. (2006); Middle habits, organization, management, critical College High Schools—Dynarski et al. (1998); reading skills, and standardized college en- Sponsor-a-Scholar—Johnson (1998); QOP— trance exam preparation are areas targeted Schirm, Stuart, and McKie (2006); Upward Bound—Seftor, Mamun, and Schirm (2009). in the AVID elective class. 169. Puente—Gandara (2002, 2004); Puente— Gandara, Mejorado, Gutierrez, and Molina (1998); Like AP, the intervention has been exten- Puente—Moreno (2002). sively but not rigorously studied. Four 170. Bailis et al. (1995); Math and Science Upward studies of the program with causal de- Bound—Olsen et al. (2007); Early Academic Out- signs did not meet standards. One stud- reach Program—Quigley (2003). ied the effects of AVID on a set of matched 171. EXCEL—Bergin, Cooks, and Bergin (2007); schools in Texas, although not enough in- Talent Search—Constantine et al. (2006); Middle College High Schools—Dynarski et al. (1998); formation was provided to determine the Sponsor-a-Scholar—Johnson (1998); QOP— equivalence of the treatment and compari- Schirm, Stuart, and McKie (2006); Upward son groups, and found that AVID schools Bound—Seftor, Mamun, and Schirm (2009).

( 59 ) Appendix D. Technical information on the studies

found positive effects of their respective and for African-American students, more programs on college enrollment.172 rigorous courses in general. The Gateway to Higher Education program, designed to Other applicable college outreach programs increase minority opportunities for college involve academic advising in some capac- (particularly in the areas of math and sci- ity. QOP is designed so that case managers ence), requires its participants to enroll in provide advice on selecting college prepara- a curriculum designed to keep students on tory courses in high school. EXCEL provides track to college, including a requirement assistance in signing up for a college prepa- for science courses with lab components. ration curriculum in high school, and it re- One study of Gateway found that it had a quires that students pursue rigorous college positive effect on high school graduation, preparation coursework to receive a schol- the number of college prep courses taken, arship. Middle College High Schools encour- and the number of Regents exams taken, age disadvantaged students to enroll in col- particularly in math and science. Regents lege-level courses through dual enrollment tests are the standardized subject exami- opportunities on a college campus. Upward nations required of New York State high Bound projects offer academic counseling in school students. addition to the host of courses that parallel or anticipate the high school’s college prep Recommendation 2. curriculum.173 However, rigorous studies of Utilize assessment measures these four programs that met WWC stan- throughout high school so that dards did not demonstrate any impacts on students are aware of how outcomes relevant to this guide. prepared they are for college, and assist them in overcoming The panel identified causal studies of three deficiencies as they are identified otherwise relevant programs for this recommendation that could potentially meet Level of evidence: Low standards but did not provide sufficient information to assign a rating. The Puente The panel determined that the level of evi- program ensures that students are placed dence supporting this recommendation is in college prep courses, and a school coun- low. In this case, the rating is not necessar- selor assists students with selecting col- ily a result of limited or poor research: the lege-level classes and may talk about col- ability to implement related data and assess- lege prep requirements during the Puente ment systems is a fairly recent development. class. The reviewed studies of Puente dem- Advances in both capabilities and resources onstrated positive effects on high school devoted to state longitudinal datasets have courses fulfilling California’s A–G require- been promising, but these data do not yet ments and on enrollment in a four-year exist for many jurisdictions. As they become college. GEAR UP provides information more prevalent, the panel expects that re- on the types of courses students should search on their use also will expand. take to prepare for college. Although the national evaluation of the program does Although four programs with studies meet- not yet report on high school outcomes, ing standards included practices related to the authors find that students in GEAR UP data use and additional instruction to as- schools are more likely to take above-grade- sist students, these practices were neither level science courses in middle school, isolated in the evaluation nor necessarily a major component of the program.174 172. Talent Search—Constantine et al. (2006); Sponsor-a-Scholar—Johnson (1998). 174. Talent Search—Constantine et al. (2006); 173. Moore et al. (1997). Sponsor-a-Scholar—Johnson (1998); Talent

( 60 ) Appendix D. Technical information on the studies

Studies of two programs175 potentially More directly relevant to the recommen- meeting standards suggest that the use of dation was a series of eight quasi-exper- data to identify and notify students of their imental studies of four other programs academic progress during high school had more concentrated on using assessment an impact on college outcomes. There also measures to facilitate college readiness: is suggestive evidence that district- or statewide use of assessments associated with California’s Early Assessment Program college readiness (such as PLAN and ACT) (EAP) is a collaboration with the California is associated with improved college out- State University (CSU) system to ensure comes, but this correlation does not mean that college-bound high school graduates that requiring students to take those tests attain the level of English and mathemat- caused improved access to college.176 ics skills articulated by CSU faculty as necessary for college success. CSU modified Summary of evidence the 11th-grade California Standards Tests to assess English and math and provide To support its recommendation, the panel timeline information to juniors about their reviewed 12 studies of eight programs or readiness for college English and math interventions that use assessments and courses. The modified test provides di- data to improve college access for dis- agnostic information at the student level. advantaged students. For those students Students who are not assessed as college identified as not college ready, the panel ready have the opportunity to enroll in recommends that schools and districts an Expository Reading and Writing Course develop individualized plans to support (ERWC) in their senior year to address de- these students. The panel was able to deficiencies. The ERWC materials encourage rive support for this suggestion from the students’ independent thinking and grap- methods college prep and college access pling with text. One quasi-experimental programs use to help students “catch up.” study of EAP179 showed positive effects This includes four programs with stud- on reducing the need for remediation in ies that met WWC standards with or with- college, but it did not provide sufficient out reservations;177 two of which showed information to ensure that the matched positive effects on college enrollment.178 groups of students were equivalent and However, for each of these four programs, had matched on a measure of socioeco- the applicability to this recommendation nomic status.180 Another compared the is most tenuous, since the relevant com- outcomes of students enrolled in a 12th- ponent involves providing tutoring and grade English course piloting the ERWC other academic support to students iden- materials to those in a regular 12th-grade tified as needy. English class.181 The authors found that the treatment group’s performance on a Development High Schools—Kemple, Herlihy, Reading and Composing Skills Test (RCST) and Smith (2005); QOP—Schirm, Stuart, and McKie was higher using a one-tailed significance (2006). test; however, using a two-tailed test and 175. College Now—Crook (1990); California Early adjusting for clustering, the WWC found Assessment Program (EAP)—Howell, Kurlaender, that this difference was not significant. and Grodsky (2009). 176. ACT (2008a, 2008b, 2009a, 2009b). College Now is an early warning and 177. Talent Search—Constantine et al. (2006); support system for students. Students Sponsor-a-Scholar—Johnson (1998); Talent Devel- opment High School—Kemple, Herlihy, and Smith (2005); QOP—Schirm, Stuart, and McKie (2006). 179. Howell, Kurlaender, and Grodsky (2009). 178. Talent Search—Constantine et al. (2006); 180. Ibid. Sponsor-a-Scholar—Johnson (1998). 181. California State University (2005).

( 61 ) Apendip x D. Technical information on the studies

are evaluated in their junior high school ation requirements and to improve GPAs, year based on a combination of their GPA achievement levels, and other student out- and scores on the Regents tests. They are comes. One component of one study of then told if they have the option to earn Project GRAD potentially met standards. college credits free of charge in a dual en- The study did not provide sufficient infor- rollment program, or whether they are in mation to establish equivalence of its treat- need of remediation (in reading, writing, ment and comparison groups (the other or math). One quasi-experimental study approaches do not meet WWC standards). of College Now could not be assigned a The authors report that middle school stu- rating because it did not provide informa- dents in the treatment group outperformed tion on the equivalence of the treatment students in the comparison group on both and comparison groups.182 According to reading and math on the Stanford 9185 reported point estimates, College Now and Texas Assessment of Academic Skills participants are more likely than nonpar- (TAAS)186 tests.187 The other study compo- ticipants to enroll in senior College of New nents examined program impacts on high York (CUNY) colleges, to enroll in bachelor school outcomes but did not use a design of arts programs, and to enroll for sec- that met WWC standards. The panel also ond and third semesters in CUNY; College identified a well-designed, rigorous study of Now participants also take fewer remedial Project GRAD in Houston, Texas, that nev- courses and earn more degree credits in ertheless does not meet standards because the first two semesters in CUNY than do the treatment and comparison schools are nonparticipants. not equivalent in terms of race/ethnicity and the study does not provide evidence of GEAR UP is a schoolwide program de- equivalence on a measure of SES, which is a signed to increase college awareness and necessary condition of meeting standards preparedness for disadvantaged students, for this guide.188 However, the authors re- and it is supposed to provide individual- port that the program had no impact on ized academic support for students. Ac- the completion of algebra in ninth grade, cording to the first report from the na- high school student achievement, four-year tional study of GEAR UP programs (which graduation rates, or completion of a college does not report on any high school or post- prep curriculum on-time. secondary outcomes and does not provide sufficient information to assign a rating), Finally, support comes from correlational the supplemental services provided to studies showing positive effects from im- students, such as tutoring, were targeted plementation of different elements of ACT’s to those students with academic problems College Readiness System:189 EXPLORE and those who were not performing well and PLAN, administered in 8th and 10th on standardized assessments.183 Another grades as precursors to the ACT; the ACT study of GEAR UP in California similarly (along with the SAT, one of two national does not provide information to establish college admissions tests); and COMPASS, group equivalence and shows no signifi- the national college placement test admin- cant effect of GEAR UP by 8th grade.184 istered by ACT. The argument for administering such exams jurisdiction-wide to all Project GRAD staff place a premium on data analysis—to understand and track students’ progress toward meeting gradu- 185. Pearson (1996). 186. Texas Education Agency (2002). 182. Crook (1990). 187. Opuni (1999). 183. Standing et al. (2008). 188. Snipes et al. (2006). 184. Cabrera et al. (2006). 189. ACT (2008a, 2008b, 2009a, 2009b).

( 62 ) Appendix D. Technical information on the studies

students is that it sets signals for academic not provide evidence on the causal effect achievement and provides everyone, even of college-going peers.193 Despite the low students who had not previously consid- evidence rating, the panel believes that ered college, with the chance to identify ac- linking students with college-going adults ademic strengths and weaknesses. The cor- and peers is important for building aspira- relational studies conducted by ACT found tions and supporting college entry. that outcomes improved jurisdiction-wide for the entire population of tested students Summary of evidence at rates similar to those of college-bound students nationally. The panel separately reviewed studies for each of the steps that make up this Recommendation 3. recommendation. Surround students with adults and peers who build and support their Mixed evidence on mentoring college-going aspirations Four programs had studies that met Level of evidence: Low or potentially met standards and provided mentoring services.194 Studies of The panel defined the level of evidence Sponsor-a-Scholar, Career Beginnings, for this recommendation as low because and Puente reported a positive impact of limited evidence that the recommended on college enrollment,195 whereas a practices improve college enrollment rates. study of QOP found no impact on college Although the panel identified evidence enrollment.196 that supports mentoring—the first step of this recommendation—there is limited evi- The Sponsor-a-Scholar program matched dence on the other two steps. Six programs Philadelphia public school students to an with studies meeting or potentially meet- adult mentor for five years starting in 9th ing standards are aligned with one or more grade. Mentors were expected to meet with steps in this recommendation.190 Three of students at least monthly to monitor their these programs had a positive impact on academic performance and assist them in college enrollment,191 and three did not preparing and applying for college. For ex- impact enrollment.192 However, isolating ample, the program asked mentors to re- the impact of the recommended practices view students’ report cards for each grad- is difficult because these programs in- ing period and to help students stay on cluded a variety of other strategies. A few track academically. A program coordina- correlational studies reported a relation- tor recruited mentors, matched them to ship between having friends with college students, and monitored mentoring rela- plans and college enrollment, but they do tionships. The quasi-experimental study of Sponsor-a-Scholar found that the program increased college attendance in the first two 190. Career Beginnings—Cave and Quint (1990); Sponsor-a-Scholar—Johnson (1998); Puente— Gandara (2002); Career Academies—Kemple 193. Horn and Chen (1998); Hossler, Schmit, and (2004); QOP—Schirm, Stuart, and McKie (2006); Vesper (1999); Sokatch (2006). Upward Bound—Seftor, Mamun, and Schirm 194. Career Beginnings—Cave and Quint (1990); (2009). Puente—Gandara (2002, 2004); Sponsor-a- 191. Career Beginnings—Cave and Quint (1990); Scholar—Johnson (1998); QOP—Schirm, Stuart, Puente—Gandara (2002); Sponsor-a-Scholar— and McKie (2006). Johnson (1998). 195. Career Beginnings—Cave and Quint (1990); 192. Career Academies—Kemple (2004); QOP— Sponsor-a-Scholar—Johnson (1998); QOP— Schirm, Stuart, and McKie (2006); Upward Schirm, Stuart, and McKie (2006). Bound—Seftor, Mamun, and Schirm (2009). 196. QOP—Schirm, Stuart, and McKie (2006).

( 63 ) Apendip x D. Technical information on the studies

years after high school but had no impact ing. The study matched 72 Puente students on whether students remained enrolled be- in three California high schools to 72 non- tween the first and second years.197 Puente students based on grades, reading scores, ethnicity, social background, and The Career Beginnings program offered gender.200 However, there was insufficient students a two-year relationship with an information to determine whether these adult mentor in addition to academic tu- two groups were comparable on a mea- toring, summer jobs, and various college sure of socioeconomic status (a require- preparation activities. Mentors were college- ment for quasi-experimental studies in educated professionals who served as role this guide to meet standards with reser- models for students and assisted them in vations). The author reports a significant developing career goals and applying to col- positive effect on four-year college enroll- lege. The random assignment study of Ca- ment but did not provide information on reer Beginnings did not report a significant the statistical significance of other college impact on enrollment in two-year colleges enrollment outcomes. The Puente pro- or enrollment in four-year colleges, but it re- gram recruited mentors from the local ported a significant impact when these two community to serve as role models for outcomes were combined (i.e., enrollment in Latino high school students and to spend two- and four-year colleges). Although men- time with students’ families. As a note tors were an important component of the of caution, the study found that mentors Career Beginnings approach, the program had difficulty establishing a relationship included several other activities, such as with 9th-grade students who were often summer employment, that make it difficult focused on the day-to-day challenges at to assess the role of mentoring in contribut- school rather than on long-term education ing to program impacts. plans. The authors suggest this may have been related to the characteristics of the QOP assigned a case manager to at-risk mentors themselves, and they report that students for five years who was expected group activities for mentors and students to serve in a mentoring role. Case man- and efforts by mentors to reach out to par- agers worked with about 15 to 25 partici- ents were useful strategies. pants and often developed relationships described as similar to “that of a caring The panel relied on three additional pro- aunt or uncle.”198 In addition to mentor- grams for mentoring practices, although ing services, the program offered supple- these programs had studies that did not mental instruction, support services, and meet WWC standards. The College Bound community service activities. The random program recruits Boston College students assignment study of QOP in seven sites to mentor students in local high schools. found no impact on college enrollment, The mentors develop a relationship with retention, or attainment.199 students through monthly meetings and advise them about the college application The High School Puente program had one and selection processes.201 The Washington study that potentially meets standards. State Achievers program, which also in- The quasi-experimental study did not pro- cludes college scholarships, links students vide sufficient information to assign a rat- to “hometown mentors” from the community who assist students with the college 197. The author reports that college enrollment and financial aid application processes be- in the second year after high school was significant at the 0.10 significance level. 198. Maxfield et al. (2003). 200. Gandara (2002). 199. Schirm, Stuart, and McKie (2006). 201. Ladd (1992).

( 64 ) Appendix D. Technical information on the studies

ginning in their junior year.202 The I Have ties cannot be distinguished from other a Dream program does not match students aspects of the program.208 to mentors, but it has a program coordinator who is expected to assist and support The panel used practices from two pro- students throughout middle and high school grams with studies that potentially meet on their path to college.203 standards. The Puente program encouraged the formation of peer groups by mix- Limited evidence on the role of peers ing high- and low-achieving Latino students in an English class for two years.209 Three programs with studies that met In addition, Puente students participated or potentially met standards focused on in a club that supported their college prep- the role of peers, including one that had a aration activities. The study of Puente po- positive impact on college enrollment,204 tentially meets standards because there one that had no impact on enrollment,205 was insufficient information to assess and one with a study that did not measure baseline equivalence of the treatment college enrollment.206 All three programs and control groups. The study reported organized students into groups that facili- a significant positive impact on four-year tated academically oriented friendships. college enrollment, but it did not provide However, these programs do not provide information on statistical significance for direct evidence on the role of peers because other types of college enrollment. they consist of several other components designed to improve college access. The AVID program places promising students on a college preparatory track and Only one program with a study meeting provides multiple activities to prepare and standards relied on an approach relevant support these students academically. The for this recommendation.207 Career Acad- program encourages the formation of a emies group students into small learning peer group by having students take a year- communities of 50 to 75 students who long AVID course that teaches study skills, take academy classes together through- provides tutoring, and assists students in out high school. The program encour- preparing for college. The program pro- aged collaboration among students and motes a group identity by using an AVID expected the academies to serve as “com- logo, setting aside a classroom for AVID munities of support” for students. This students, and providing opportunities for approach promoted the formation of ac- students to collaborate.210 One study of ademically oriented peer groups among AVID potentially meets standards, but it students. Although a large random as- provided insufficient information to assess signment study of Career Academies in equivalence of the matched treatment and nine schools found no impact on high comparison groups. The study matched school graduation, college enrollment, or 10 AVID high schools to 10 non-AVID high postsecondary attainment, the effective- schools based on student race and SES ness of the small learning community as well as school size, type (i.e., urban or approach to career exploration activi- rural), and accountability rating.211 The authors did not report college enrollment 202. Engle, Bermeo, and O’Brien (2006). outcomes but found a positive effect of 203. Kahne and Bailey (1999); Kuboyama (2000); the program on school-level achievement. McGrath and Hayman (1997). 204. Gandara (2004). 208. Ibid. 205. Kemple (2004). 209. Gandara (2004). 206. Watt et al. (2006). 210. Guthrie and Guthrie (2002); Mehan (1996). 207. Kemple (2004). 211. Watt et al. (2006).

( 65 ) Appendix D. Technical information on the studies

However, the study did not report the sta- study of Upward Bound found that a major- tistical significance of this finding. ity of sites provide job-shadowing opportunities. A random assignment study of Up- ward Bound found no significant impact on Lack of evidence on career exploration high school completion, college enrollment, or college credits earned.216 The program The panel identified two programs with did have a positive significant impact on studies meeting standards that offered ca- four-year college attendance for first-gener- reer exploration activities, although neither ation students. However, these impacts can- had an impact on college enrollment.212 not be attributed specifically to career devel- Career exploration activities represent a opment activities because Upward Bound major component of the Career Academies included several other components. approach and are commonly implemented by Upward Bound sites. The panel could The panel relied on practices from four pro- not isolate the effect of career exploration grams that did not have studies meeting activities because both programs consist of standards. The Lansing Area Manufactur- several other components—Career Acad- ing Partnership (LAMP) linked students to emies offer a comprehensive school-to- work-based learning experiences in a local work approach that organizes students into General Motors factory, including several small learning communities and combines hands-on activities. Although a study of the vocational and college preparatory curri- program matched participants to nonpartic- cula, and Upward Bound prepares students ipants based on gender, race, age, GPA, and academically for college through classes of- school, the two groups were not matched fered during the school year and summer. on an indicator of socioeconomic status.217 An implementation study of the School-to- Career Academies implement a variety Work Opportunities Act included surveys of career exploration activities, includ- of school-to-work partnerships, student ing speakers from the business commu- surveys, and case studies of eight states.218 nity, visits to employer sites, career fairs, According to the study, students viewed and job shadowing.213 Career Academies one-on-one interaction with an adult—job often offer a sequence of career activi- shadowing, paid jobs, and unpaid intern- ties, beginning with interest inventories ships—as the most useful type of career and employer site visits, and leading up exploration activity. An implementation of to job shadowing and internships.214 Ca- GEAR UP in Austin, Texas, describes how the reer Academies students take classes in program used online interest inventories to learning communities that are organized identify students’ career interests and help around different career themes. them understand the knowledge and skills needed for their area of interest.219 The Most Upward Bound grantees offer career Roads to Success program, which is being development activities, such as career plan- evaluated in a random assignment study ning assistance, meetings with employers, employer site visits, and job shadowing.215 Although detailed information about these 216. Myers et al. (2004); Seftor, Mamun, and activities is not available, an implementation Schirm (2009). 217. MacAllum et al. (2002). In addition, the study did not provide information to assess the equiv- 212. Kemple (2004); Seftor, Mamun, and Schirm alence of the treatment and comparison groups (2009). used in the analysis. 213. Kemple, Poglinco, and Snipes (1999). 218. Hershey et al. (1999). 214. Ibid. 219. Austin Independent School District, Office 215. Moore et al. (1997). of Program Evaluation (2002).

( 66 ) Appendix D. Technical information on the studies

that has not produced impact findings yet, lead to positive impacts, the panel defined helps students link their career interests and the level of evidence for this recommenda- education plans.220 tion as moderate.

Recommendation 4. Summary of evidence Engage and assist students in completing critical steps Six programs that assisted students with for college entry the college entry process had studies that met standards.225 Three of these programs Level of evidence: Moderate had positive effects on college access or college enrollment outcomes,226 and three The panel judged the level of evidence for had no impact on college enrollment.227 this recommendation to be moderate. Six programs with studies meeting standards A main purpose of the Talent Search pro- with or without reservations provided as- gram is to assist students with the college sistance with the college entry process.221 entry process. The program sites often Although these programs consisted of have staff who work directly with students other strategies to prepare students for to counsel and advise them about the col- college, all of them focused on helping stu- lege admissions process.228 An implemen- dents complete the steps to college entry. tation study found that more than 90 per- Three of the programs had a positive im- cent of projects provide college orientation pact on college enrollment,222 and three activities, college visits, and counseling.229 had no impact on enrollment.223 The panel In addition, project directors described notes that the evidence for this recom- college visits as one of the top two activi- mendation focuses on programs that serve ties that contributed to program objec- low-income and first-generation students tives.230 The large-scale study of Talent with average academic achievement. Only Search participants in three states found one program, which did not find an im- positive effects on the percentage of stu- pact, specifically targeted students at risk dents taking college entrance exams and for dropping out of high school.224 Since college enrollment.231 programs that assisted students with the college entry process did not consistently 225. EXCEL—Bergin, Cooks, and Bergin (2007); Career Beginnings—Cave and Quint (1990); Tal- ent Search—Constantine et al. (2006); Sponsor- 220. Roads to Success (2008a, 2008b). a-Scholar—Johnson (1998); QOP—Schirm, Stu- 221. EXCEL—Bergin, Cooks, and Bergin (2007); art, and McKie (2006); Upward Bound—Seftor, Career Beginnings—Cave and Quint (1990); Tal- Mamun, and Schirm (2009). ent Search—Constantine et al. (2006); Sponsor- 226. Career Beginnings—Cave and Quint (1990); a-Scholar—Johnson (1998); QOP—Schirm, Stu- Talent Search—Constantine et al. (2006); Spon- art, and McKie (2006); Upward Bound—Seftor, sor-a-Scholar—Johnson (1998). Mamun, and Schirm (2009). 227. EXCEL—Bergin, Cooks, and Bergin (2007); 222. Career Beginnings—Cave and Quint (1990); QOP—Schirm, Stuart, and McKie (2006); Upward Talent Search—Constantine et al. (2006); Spon- Bound—Seftor, Mamun, and Schirm (2009). One sor-a-Scholar—Johnson (1998). of the studies finding no detectable effects on 223. EXCEL—Bergin, Cooks, and Bergin (2007); college access or enrollment outcomes found QOP—Schirm, Stuart, and McKie (2006); Upward mixed effects on high school academic outcomes, Bound—Seftor, Mamun, and Schirm (2009). One which the panel deemed not relevant for this of the studies finding no detectable effects on recommendation. college access or enrollment outcomes found 228. Calahan et al. (2004). mixed effects on high school academic outcomes, which the panel deemed not relevant for this 229. Ibid. recommendation. 230. Ibid. 224. QOP—Schirm, Stuart, and McKie (2006). 231. Constantine et al. (2006).

( 67 ) Apendip x D. Technical information on the studies

The Sponsor-a-Scholar program in Phila- An implementation study of the QOP sites delphia had an academic coordinator who suggests several types of activities to as- set up college entrance exam preparation sist students in completing the steps for classes, planned college visits, and held college entry. Case managers encouraged workshops on selecting a college and pre- students to take college entrance exams paring for the challenges of college.232 In and in some sites paid the exam registra- addition, mentors were expected to pro- tion fees.235 Several QOP sites also helped vide one-on-one assistance with the col- participants prepare for entrance exams lege admissions process. The author re- by offering tutoring or purchasing train- ports that the program had a significant ing software. A few sites also purchased positive impact on a broad measure of books and CD-ROMs that provided enroll- college preparation activities that summed ees with test-taking tips and sample ex- up the number of preparation activities in aminations. Case managers provided as- which students participated. sistance in completing college applications and organized college visits for students. There is limited information on the college The large-scale random controlled trial of preparation activities offered by the seven QOP found no significant effects on col- Career Beginnings sites that were studied. lege enrollment. Perhaps most relevant to The random assignment study found that this recommendation, study authors also sites offered workshops or classes on tak- noted that QOP’s policy of having a case ing college entrance exams and classes on manager on duty for a large number of completing college admissions forms.233 hours weekly was somewhat unrealistic Career Beginnings had a positive impact and therefore not always implemented ac- on college enrollment; the direct effect of cording to the QOP model. the college preparation activities cannot be determined, however, since the extent EXCEL staff encouraged and guided stu- of these college preparation activities was dents through the college application pro- not clear. cess. The EXCEL program also held workshops on campus and allowed students Upward Bound projects helped students to live in college dormitories during the prepare for college entrance exams (PSAT summer sessions. The summer institutes and SAT) and provided assistance with provided writing instruction to assist stu- college applications and financial aid dents in writing essays. A small random forms.234 The program also provided op- assignment study of EXCEL in one Mid- portunities for students to visit colleges, western city found no detectable effects and many sites encouraged students to on college enrollment the fall after high live on campus for the summer academic school graduation.236 course sessions. Although there was limited information on the quality or intensity Three additional programs had studies of these services, they are a common part that potentially met standards237 or did of program services. Upward Bound had not meet standards.238 The panel judged no effects on enrollment or persistence in the practices discussed in the studies to two- or four-year colleges for program participants, but it had a significant impact on 235. Maxfield et al. (2003). four-year college attendance for students 236. Bergin, Cooks, and Bergin (2007). with low academic expectations. 237. Puente—Gandara (2002); AVID—Watt et al. (2006). 232. Johnson (1998). 238. I Have a Dream—Kahne and Bailey (1999); 233. Cave and Quint (1990). I Have a Dream—Kuboyama (2000); I Have a 234. Seftor, Mamun, and Schirm (2009). Dream—McGrath and Hayman (1997).

( 68 ) Appendix D. Technical information on the studies

be similar to practices implemented in offered grew over time.241 The study fo- programs with studies that met standards cused on program implementation in the with or without reservations and included 7th and 8th grades, but it did not provide practices described in these studies as sufficient information to assess baseline supplemental evidence of a practice. equivalence. The initial findings described the program’s effect on students’ and par- A study of the AVID program potentially ents’ education expectations and knowl- meets standards but could not be rated be- edge of postsecondary opportunities. cause there was insufficient information to assess baseline equivalence. Students par- In the I Have a Dream program, sites worked ticipating in AVID were given extra coach- individually with and monitored each partic- ing on how to write statements of purpose ipant’s college application process. One site and how to fill out college applications hired a college counselor from a prestigious and financial aid forms, and they were re- private school to meet with each participant, minded about test and application dead- and another site assigned an AmeriCorps lines. At one site, to familiarize students member to focus entirely on this process.242 with college catalogs and to help them There were no studies of the I Have a Dream choose an appropriate college, students program that met standards. received a handout called “Choosing Your College” that contained a checklist of infor- Recommendation 5. mation typically found in college catalogs Increase families’ financial and were instructed to fill in the informa- awareness, and help students tion for a particular college according to apply for financial aid the assigned checklist.239 The study that potentially met standards did not measure Level of evidence: Moderate the program’s effect on college enrollment but reported a positive effect on school- The panel judged the level of evidence for level achievement. However, the study did this recommendation as moderate because not provide information on the statistical there is evidence that the recommended significance of this finding. practices increased the financial assistance application rate and the college enrollment Counselors from the Puente program ar- rate. Two programs with studies meeting range college visits, including overnight standards with reservations informed stu- trips, for students.240 Parents often are in- dents about financial aid opportunities vited on these visits to make them more and provided hands-on assistance in com- comfortable with the college environment. pleting financial aid applications.243 Both The program also introduces students to of these programs had a positive impact representatives from Latino student groups on financial aid application and college on campus. A study of the Puente program enrollment. Although five other programs potentially meets standards and reported with studies meeting standards with and a positive impact on four-year college en- without reservations provided financial rollment. A national study of GEAR UP that aid services, the level and intensity of potentially meets standards found that college visits were common across all of the sites and that the number of college visits 241. Standing et al. (2008). 242. Kahne and Bailey (1999); Kuboyama (2000); 239. Mehan (1996); Watt et al. (2006). McGrath and Hayman (1997). 240. Gandara (2004); Grubb, Lara, and Valdez 243. FAFSA Experiment—Bettinger et al. (2009); (2002). Talent Search—Constantine et al. (2006).

( 69 ) Appendix D. Technical information on the studies

these services were often unclear.244 Two enrollment,249 and three found no impact of these studies measured financial aid on enrollment.250 outcomes and found no impact,245 and the findings for college enrollment were The FAFSA Experiment compared two in- mixed across these five studies. Since the terventions designed to improve the fi- two programs that closely correspond nancial aid and college enrollment rates. with the panel’s recommendation had a The main intervention consisted of an positive impact on financial aid application H&R Block tax professional using a fam- and college enrollment, the panel assigned ily’s income tax return and a structured a moderate level of evidence. protocol of questions to fill out the FAFSA, offering to submit the FAFSA form free Summary of evidence of charge, and providing information on financial aid eligibility for local colleges. Seven programs with studies meeting A second intervention offered families in- standards offered some form of financial formation on their financial aid eligibility aid assistance, often informing students based on their tax return without assist- and parents about financial aid or helping ing them in completing or submitting the students with financial aid applications.246 FAFSA. The main intervention had a sig- The types of practices recommended by nificant positive impact on financial aid the panel formed a major part of Talent application and college enrollment for the Search and the FAFSA Experiment, and dependent participants, who were mostly both had a significant impact on applica- high school seniors. The information-only tion for financial aid and college enroll- intervention did not have an impact on ment.247 Five other programs offered fi- either outcome. nancial aid assistance, but the studies of these programs did not provide sufficient A primary goal of Talent Search is to in- information on the type and frequency form students about the availability of fi- of these services. Among these five pro- nancial aid, provide financial aid counsel- grams, two had studies measuring finan- ing, and assist students in completing the cial aid outcomes and found no impact,248 FAFSA. Almost all Talent Search projects two found a positive impact on college report providing these services, and both students and project staff emphasize the importance of financial aid assistance.251 244. EXCEL—Bergin, Cooks, and Bergin (2007); Programs often provide general informa- Career Beginnings—Cave and Quint (1990); Spon- sor-a-Scholar—Johnson (1998); QOP—Schirm, tion that raises awareness of financial aid Stuart, and McKie (2006); Upward Bound—Seftor, in the middle school and early high school Mamun, and Schirm (2009). grades and then offer counseling and one- 245. Career Beginnings—Cave and Quint (1990); on-one support in filling out the FAFSA in Upward Bound—Seftor, Mamun, and Schirm the junior and senior years. A quasi-exper- (2009). imental study of Talent Search students in 246. EXCEL—Bergin, Cooks, and Bergin (2007); Florida, Indiana, and Texas found that the FAFSA Experiment—Bettinger et al. (2009); Ca- reer Beginnings—Cave and Quint (1990); Talent program had a substantial impact on the Search—Constantine et al. (2006); Sponsor-a- financial aid application rate and postsec- Scholar—Johnson (1998); QOP—Schirm, Stu- art, and McKie (2006); Upward Bound—Seftor, Mamun, and Schirm (2009). 249. Career Beginnings—Cave and Quint (1990); 247. FAFSA Experiment—Bettinger et al. (2009); Sponsor-a-Scholar—Johnson (1998). Talent Search—Constantine et al. (2006). 250. EXCEL—Bergin, Cooks, and Bergin (2007); 248. Career Beginnings—Cave and Quint (1990); QOP—Schirm, Stuart, and McKie (2006); Upward Upward Bound—Seftor, Mamun, and Schirm Bound—Seftor, Mamun, and Schirm (2009). (2009). 251. Calahan et al. (2004).

( 70 ) Apendip x D. Technical information on the studies

ondary enrollment.252 The authors found is not clear from the seven sites included effect sizes between 0.34 and 0.67 for the in the study meeting standards.257 The financial aid application rate, and persis- study reports that one site enlisted the tence of these effects when controlling for help of a financial aid officer and another high school graduation. held classes on completing financial aid forms, although there is limited informa- Five other programs with studies meet- tion to assess the prevalence and quality ing standards provided assistance with of financial aid services. A random assign- the financial aid process, but detailed in- ment study of Career Beginnings found formation on their financial aid practices that the program did not have a significant was lacking.253 impact on scholarship receipt or student loan usage, but it did increase college en- Although Upward Bound focuses on pro- rollment (i.e., enrollment in a two- or four- viding academic coursework during the year college). school year and summer, the program offers several nonacademic services as well. Sponsor-a-Scholar mentors and QOP case An implementation study of Upward Bound managers assisted students with finan- found that most sites provide assistance cial aid applications, and both programs with financial aid, but the intensity of these held workshops on the financial aid pro- services varied widely.254 Some sites offered cess.258 The studies of both programs did a workshop on financial aid, whereas others not provide enough detail to determine met with students and parents on multiple the frequency or quality of these services, occasions to assist with the financial aid and neither study measured financial aid application process.255 The prevalence of outcomes. As described, Sponsor-a-Scholar one-on-one financial aid counseling could had a positive impact on college enroll- not be determined from the available stud- ment, and QOP had no impact on college ies. The random assignment study of the enrollment, retention, or degree attain- program did not find a significant impact ment. Although a study of EXCEL notes on financial aid application or receipt, or on that the program assisted parents with postsecondary enrollment or attainment.256 financial aid forms, there is no additional The study did report a significant impact on information to determine the financial financial aid application for students who aid assistance provided through the pro- participated in the program for a longer gram.259 A relatively small random assign- period of time. ment study of about 70 students found that the program did not have a positive The implementation of financial aid assis- impact on college enrollment.260 tance in the Career Beginnings program The panel examined two programs with 252. Constantine et al. (2006). The study found studies that potentially meet standards differences in the financial aid application rate for practices related to engaging parents. between participants and nonparticipants of 14, Parent involvement formed a major part of 17, and 28 percentage points in Indiana, Florida, the Puente program (described in recom- and Texas, respectively. mendation 3). The program has parents at- 253. EXCEL—Bergin, Cooks, and Bergin (2007); tend interviews along with students, hosts Career Beginnings—Cave and Quint (1990); Spon- sor-a-Scholar—Johnson (1998); QOP—Schirm, Stuart, and McKie (2006); Upward Bound—Seftor, 257. Cave and Quint (1990). Mamun, and Schirm (2009). 258. Johnson (1998); Schirm, Stuart, and McKie 254. Moore et al. (1997). (2006). 255. Ibid. 259. Bergin, Cooks, and Bergin (2007). 256. Seftor, Mamun, and Schirm (2009). 260. Ibid.

( 71 ) Apendip x D. Technical information on the studies

informal parent nights, and holds one-on- in the 9th and 10th grades and provided one parent meetings with a Puente coun- individual assistance with financial aid selor.261 Parents receive financial aid infor- and college applications in the 11th and mation and counseling. A national study of 12th grades. A quasi-experimental study GEAR UP that potentially meets standards of the program compared participants created a matched comparison group of and nonparticipants but did not include GEAR UP and non–GEAR UP schools, but an indicator of socioeconomic status to it did not provide sufficient information to match students. The Kids2College pro- assess equivalence of the analysis sample. gram provided financial aid information As part of the study, the authors found to middle school students as part of its that sites that effectively engaged parents college awareness program that helped used two practices: 9- to 10-week parent prepare students for college. A study of institutes and individual counseling ses- the program examined the pre- and post- sions for parents and children.262 The in- program outcomes of participants, but it stitutes consisted of a series of workshops did not have a comparison group of stu- to inform parents about how to help their dents.263 The Neighborhood Academic Ini- children prepare for college. tiative (NAI) coordinates a Family Develop- ment Institute that includes information The panel considered practices from three on financial aid and works with a financial other programs that provided financial aid aid officer from the University of South- information or assistance, but it did not ern California to assist students in com- have a study that met standards. The Bal- pleting financial aid application forms.264 timore College Bound program conducted The panel did not identify any comparison presentations on financial aid for students group studies of the NAI.

261. Grubb, Lara, and Valdez (2002). 263. Cunningham, Erisman, and Looney (2007). 262. Standing et al. (2008). 264. Tierney and Jun (2001).

( 72 ) References financial aid information. Washington, DC: Author. Achieve, Inc. (2008). American Diploma Alexander, N. A. (2002). Race, poverty, and Project Algebra II end-of-course exam: the student curriculum: Implications 2008 annual report. Washington, DC: for standards policy. American Educa- Author. tional Research Journal, 39, 675–693. Achieve, Inc. (2009). Closing the expecta- Allensworth, E., Nomi, T., Montgomery, N., tions gap: Fourth annual 50-state prog- & Lee, V. E. (2008). College preparatory ress report on the alignment of high curriculum for all: Consequences of ninth- school policies with the demands of grade course taking in algebra and Eng- college and careers. Washington, DC: lish on academic outcomes in Chicago. Author. Chicago, IL: University of Chicago, Con- ACT. (2007). Using EXPLORE and PLAN data sortium on Chicago School Research. to evaluate GEAR UP programs. Iowa American Educational Research Associa- City, IA: ACT, Inc. tion, American Psychological Associa- ACT. (2008a). COMPASS guide to successful tion, and National Council on Measure- high school outreach: Setting the right ment in Education. (1999). Standards for course for college success. Iowa City, IA: educational and psychological testing. ACT, Inc. Washington, DC: AERA Publications. ACT. (2008b). The forgotten middle: Ensur- American Psychological Association. ing that all students are on target for (2002). Criteria for practice guideline college and career readiness before high development and evaluation. American school. Iowa City, IA: ACT, Inc. Psychologist, 57(12), 1048–1051. ACT. (2009a). Readiness and success: State- Attewell, P., & Domina, T. (2008). Raising the wide implementation of EXPLORE and bar: Curricular intensity and academic PLAN. Iowa City, IA: ACT, Inc. performance. Educational Evaluation ACT. (2009b). Statewide administration of and Policy Analysis, 30(1), 51–71. the ACT: A key component in improving Austin Independent School District, Office student access to college and work. Iowa of Program Evaluation. (2002). GEAR UP City, IA: ACT, Inc. Austin: Impacting lives project, 2000– Adelman, C. (1999). Answers in the tool- 2001 evaluation. Austin, TX: Author. box. Academic intensity, attendance Avery, C., & Kane, T. J. (2004). Student per- patterns, and bachelor’s degree attain- ceptions of college opportunities: The ment. Washington, DC: U.S. Department Boston COACH program. In C. M. Hoxby of Education. (Ed.), College choices: The economics of Adelman, C. (2006). The toolbox revisited: where to go, when to go, and how to pay Paths to degree completion from high for it (pp. 355–394). Cambridge, MA: Na- school through college. Washington, DC: tional Bureau of Economic Research. U.S. Department of Education, Office of Bailey, T. (2009). Challenge and oppor- Vocational and Adult Education. tunity: Rethinking the role and func- Advisory Committee on Student Financial tion of developmental education in Assistance. (2006). Mortgaging our fu- community college. In A. C. Bueschel ture: How financial barriers to college & A. Venezia (Eds.), New directions for undercut America’s global competitive- community colleges: Policies and prac- ness. Washington, DC: Author. tices to improve student preparation and Advisory Committee on Student Finan- success (pp. 11–30). San Francisco, CA: cial Assistance. (2008). Early and often: Jossey-Bass. Designing a comprehensive system of Bailey, W. V. (2002). An evaluation of the efficacy of the Advancement Via Indi- vidual Determination (AVID) program

( 73 ) R eferENCES

(Doctoral dissertation, University of Implementation of the Talent Search Virginia, 2002). Dissertation Abstracts program, past and present: Final report International, 63(6A), 2126. from phase I of the national evalua- Bailis, L. N., Hahn, A., Aaron, P., Nahas, J., tion. Princeton, NJ: Mathematica Policy & Leavitt, T. (1995). A new field emerges: Research. College access programs. Waltham, MA: California State University. (2005). Pilot Brandeis University, Center for Human study evaluation of the Early Assessment Resources. Program’s professional development in Baum, S., & Ma, J. (2007). Education pays: English 2004–2005. Long Beach, CA: The benefits of higher education for in- California State University, Office of the dividuals and society. Washington, DC: Chancellor. College Board. Campbell, P. B., Wahl, E., Slater, M., Iler, E., Bergin, D. A., Cooks, H. C., & Bergin, C. C. Moeller, B., Ba, H., et al. (1998). Paths to (2007). Effects of a college access pro- success: An evaluation of the Gateway gram for youth underrepresented in to Higher Education program. Journal higher education: A randomized ex- of Women and Minorities in Science and periment. Research in Higher Education, Engineering, 4(2–3), 297–308. 48(6), 727–750. Cave, G., & Quint, J. (1990). Career Begin- Bettinger, E., Long, B. T., Oreopoulos, P., & nings impact evaluation: Findings from a Sanbonmatsu, L. (2009). The role of in- program for disadvantaged high school formation and simplification in college students. New York: Manpower Demon- decisions: Results from the FAFSA Experi- stration Research Corporation. ment. Unpublished manuscript. Choy, S. P. (2002). Access & persistence: Find- Black, A. C., Little, C. A., McCoach, D. B., ings from 10 years of longitudinal re- Purcell, J. H., & Siegle, D. (2008). Ad- search on students. Washington, DC: vancement Via Individual Determina- American Council on Education. tion: Method selection in conclusions Choy, S. P., Horn, L. J., Nunez, A., & Chen, X. about program effectiveness. Journal of (2000). Transition to college: What helps Educational Research, 102(2), 111–124. at-risk students and students whose par- Black, S., & Sufi, A. (2002). Who goes to col- ents did not attend college. New Directions lege? Differential enrollment rates by for Institutional Research, 107, 45–63. race and family background (Working Christie, K., & Zinth, K. (2008). Ensuring suc- Paper 9310). Cambridge, MA: National cessful student transitions from the middle Bureau of Economic Research. grades to high school. Denver, CO: Educa- Cabrera, A. F., Deil-Amen, R., Prabhu, R., tion Commission of the States. Terenzini, P. T., Lee, C., & Franklin, R. E., Jr. Constantine, J. M., Seftor, N. S., Sama Martin, E., (2006). Increasing the college prepared- Silva, T., & Myers, D. (2006). Study of the ness of at-risk students. Journal of Lati- effect of the Talent Search program on nos and Education, 5(2), 79 – 97. secondary and postsecondary outcomes Cabrera, A. F., & La Nasa, S. M. (2001). in Florida, Indiana, and Texas: Final re- On the path to college: Three critical port from phase II of the national evalu- tasks facing America’s disadvantaged. ation. Princeton, NJ: Mathematica Policy Research in Higher Education, 42(2), Research. 119 –149. Crook, D. (1990). The impact of College Cabrera, A. F., & La Nasa, S. M. (2000). Un- Now: An assessment of the retention derstanding the college choice of disad- and academic progress of program par- vantaged students. San Francisco, CA: ticipants who entered CUNY in fall 1987 Jossey-Bass. [and] College Now alumni survey, fall Calahan, M., Silva, T., Humphrey, J., 1990. New York: Kingsborough Com- Thomas, M., & Cunningham, K. (2004). munity College.

( 74 ) R eferENCES

Cunningham, A. F., Erisman, W., & Looney, S. M. with diverse abilities in early second- (2007). From aspirations to action: The ary school. Educational Evaluation and role of middle school parents in making Policy Analysis, 22(3), 241–254. the dream of college a reality. Wash- Gandara, P. (2002). A study of High School ington, DC: Institute of Higher Educa- Puente: What we have learned about tion Policy. preparing Latino youth for postsec- Dougherty, C., Mellor, L., & Jian, S. (2006). ondary education. Educational Policy, The relationship between Advanced 16(4), 474 – 495. Placement and college graduation. Aus- Gandara, P. (2004). Building bridges to college. tin, TX: National Center for Educational Educational Leadership, 62(3), 56–60. Accountability. Gandara, P., Mejorado, M., Gutierrez, D., & Dounay, J. (2006). Ensuring rigor in the Molina, M. (1998). Final report of the eval- high school curriculum: What states are uation of High School Puente: 1994–1998. doing. Denver, CO: Education Commis- New York: Carnegie Corporation. sion for the States. Gewertz, C. (2009). Ninth grade, by the num- Dounay, J. (2008). Improving outcomes bers. Education Week, 28(24), 26–29. for traditionally underserved students Grodsky, E., & Jones, M. T. (2007). Real through Early College High Schools. and imagined barriers to college entry: Denver, CO: Education Commission for Perceptions of cost. Social Science Re- the States. search, 36(2), 745–766. Dynarski, M., Gleason, P., Rangarajan, A., Grubb, W. N., Lara, C. N., & Valdez, S. (2002). & Wood, R. (1998). Impacts of dropout pre- Counselor, coordinator, monitor, mom: vention programs: Final report. Princeton, The roles of counselors in the Puente NJ: Mathematica Policy Research. program. Educational Policy Journal, Dynarski, S. M., & Scott-Clayton, J. E. 16(4), 547–571. (2007). College grants on a postcard: A Guthrie, L. F., & Guthrie, G. P. (2002). The mag- proposal for simple and predictable fed- nificent eight: AVID best practices study. eral student aid (Hamilton Project dis- Burlingame, CA: Center for Research, cussion papers, 2008-01). Washington, Evaluation and Training in Education. DC: The Brookings Institution. Haimson, J., & Deke, J. (2003). Preparing Ellwood, D. T., & Kane, T. J. (2000). Who for productive careers: Students’ par- is getting a college education? Family ticipation in and use of career-focused background and the growing gaps in learning activities. Princeton, NJ: Math- enrollment. In S. Danziger & J. Waldfo- ematica Policy Research. gel (Eds.), Securing the future: Invest- Hamilton, L., Halverson, R., Jackson, S., ing in children from birth to college Mandinach, E., Supovitz, J., and Way- (pp. 283–324). New York: Russell Sage man, J. (2009). Using student achieve- Foundation. ment data to support instructional Engle, J., Bermeo, A., & O’Brien, C. (2006). decision making (NCEE 2009-4067). Straight from the source: What works for Washington, DC: U.S. Department of first-generation college students. Wash- Education, Institute of Education Sci- ington, DC: Pell Institute for the Study ences, National Center for Education of Opportunity in Higher Education. Evaluation and Regional Assistance. Field, M. J., & Lohr, K. N. (Eds.). (1990). Clini- Hargrove, L., Godin, D., & Dodd, B. (2008). cal practice guidelines: Directions for a College outcomes comparisons by AP new program. Washington, DC: National and non-AP high school experiences Academy Press. (Research Rep. No. 2008-3). New York: Gamoran, A., & Hannigan, E. C. (2000). Al- College Board. gebra for everyone? Benefits of college- Hershey, A. M., Silverberg, M. K., Haimson, J., preparatory mathematics for students Hudis, P., & Jackson, R. (1999). Expanding

( 75 ) R eferENCES

options for students: Report to Congress on Johnson, A. W. (1998). An evaluation of the national evaluation of school-to-work the long-term impacts of the Sponsor- implementation. Princeton, NJ: Math- a-Scholar program on student achieve- ematica Policy Research. ment. Princeton, NJ: Mathematica Policy Horn, L. J., & Chen, X. (1998). Toward re- Research. siliency: At-risk students who make it to Kahne, J., & Bailey, K. (1999). The role of so- college. Washington, DC: U.S. Depart- cial capital in youth development: The ment of Education, Office of Educational case of “I Have a Dream” programs. Edu- Research and Improvement. cational Evaluation and Policy Analysis, Horn, L. J., Chen, X., & Chapman, C. (2003). 21(3), 321–343. Getting ready to pay for college: What Kallison, J., & Stader, D. (2008). Final evalu- students and their parents know about ation report: 2007 high school summer the cost of college tuition and what they bridge program. Arlington, TX: Texas are doing to find out (Statistical analysis Higher Education Coordinating Board. report, NCES 2003-030). Washington, Kane, T. J., & Rouse, C. E. (1995). Labor- DC: U.S. Department of Education, Na- market returns to two- and four-year tional Center for Education Statistics. college. The American Economic Review, Hossler, D., Schmit, J., & Vesper, N. (1999). 85(3), 600 – 614. Going to college: How social, economic, Kao, G., & Tienda, M. (1998). Educational and educational factors influence the aspirations of minority youth. American decisions students make. Baltimore, MD: Journal of Education, 106(3), 349–384. The Johns Hopkins University Press. Karp, M. M., Calcagno, J. C., Hughes, K. L., Howell, J. S., Kurlaender, M., & Grodsky, E. Jeong, D. W., & Bailey, T. R. (2007). The (2009). Postsecondary preparation and postsecondary achievement of partici- remediation: Examining the effect of the pants in dual enrollment: An analysis of early assessment program at California student outcomes in two states. St. Paul, State University. Davis, CA: University of MN: University of Minnesota, National California–Davis. Research Center for Career and Techni- Ikenberry, S. O., & Hartle, T. W. (1998). Too cal Education. little knowledge is a dangerous thing: Karp, M. M., Calcagno, J. C., Hughes, K. L., What the public thinks about paying Jeong, D. W., & Bailey, T. (2008). Dual for college. Washington, DC: American enrollment students in Florida and New Council on Education. York City: Postsecondary outcomes Ishitani, T. T., & Snider, K. G. (2004, May). (CCRC Brief No. 37). New York: Colum- Longitudinal effects of college prepa- bia University Teachers College. ration programs on college retention. Kemple, J. J. (2001). Career Academies: Im- Paper presented at the 44th Annual pacts on students’ initial transitions to Forum of the Association for Institu- post-secondary education and employ- tional Research, Boston. ment. New York: Manpower Demonstra- Jackson, C. K. (2009). A little now for a lot tion Research Corporation. later: A look at a Texas Advanced Place- Kemple, J. J. (2004). Career Academies: ment incentive program. Ithaca, NY: Impacts on labor market outcomes and Cornell University, School of Industrial educational attainment. New York: and Labor Relations. Manpower Demonstration Research Jeong, D. W. (2009). High school comple- Corporation. tion, college entrance, and Advanced Kemple, J. J. (2008). Career Academies: Long- Placement programs in U.S. secondary term impacts on labor market outcomes, schools. New York: Teachers College at educational attainment, and transitions Columbia University, Community Col- to adulthood. New York: Manpower Dem- lege Research Center. onstration Research Corporation.

( 76 ) R eferENCES

Kemple, J. J., Herlihy, C. M., & Smith, T. J., Lozano, A., Watt, K. M., & Huerta, J. (2009). (2005). Making progress toward gradu- A comparison study of 12th grade stu- ation: Evidence from the Talent Devel- dents’ college anticipations, aspirations, opment High School model. New York: and college preparatory measures. Man- Manpower Demonstration Research uscript submitted for publication. Corporation. Luna De La Rosa, M. (2006). Is opportunity Kemple, J. J., Poglinco, S. M., & Snipes, J. C. knocking?: Low-income students’ per- (1999). Career Academies: Building ca- ceptions of college and financial aid. reer awareness and work-based learning American Behavioral Scientist, 49(12), activities through employer partnerships. 1670 –1686. New York: Manpower Demonstration Re- MacAllum, K., Glover, D. M., Queen, B., & search Corporation. Riggs, A. (2007). Deciding on postsec- Kemple, J. J., & Snipes, J. C. (2000). Career ondary education: Final report (NPEC Academies: Impacts on students’ engage- 2008-850). Washington, DC: National ment and performance in high school. Postsecondary Education Cooperative. New York: Manpower Demonstration MacAllum, K., Yoder, K., Kim, S., & Bozick, R. Research Corporation. (2002). Moving forward: College and Keng, L., & Dodd, B. G. (2008). A compari- career transitions of LAMP graduates. son of college performances of AP and Washington, DC: Academy for Educa- non-AP student groups in 10 subject tional Development. areas (Research Rep. No. 2008-7). New Martinez, M., & Klopott, S. (2003). Improv- York: College Board. ing college access for minority, low- King, J. E. (2006). Missed opportunities revis- income, and first-generation students. ited: New information on students who do Washington, DC: Pathways to College not apply for financial aid. Washington, Network Clearinghouse. DC: American Council on Education. Maxfield, M., Castner, L., Maralani, V., & Klopfenstein, K., & Thomas, M. K. (2009). Vencill, M. (2003). The Quantum Oppor- The link between Advanced Place- tunity Program demonstration: Imple- ment experience and early college suc- mentation findings. Washington, DC: cess. Southern Economic Journal, 75(3), Mathematica Policy Research. 873–891. McCauley, D. (2007). The impact of Ad- Kuboyama, E. (2000). East Palo Alto I Have a vanced Placement and dual enrollment Dream program evaluation report. Stan- programs on college graduation. San ford, CA: Stanford University, School of Marcos, TX: Texas State University. Education. McGrath, R. E., & Hayman, J. (1997). The Ladd, G. T. (1992). College bound: 1989– Paterson, New Jersey, I Have a Dream 1992: The Fund for the Improvement of program: Academic performance and Postsecondary Education, FIPSE final re- outcomes. Teaneck, NJ: Fairleigh Dickin- port. Chestnut Hill, MA: Boston College, son University, School of Psychology. School of Education. Mehan, H. (1996). Constructing school suc- Lee, V. E., & Ready, D. D. (2009). U.S. high cess: The consequences of untracking school curriculum: Three phases of low-achieving students. New York: Cam- contemporary research and reform. The bridge University Press. Future of Children, 19(1), 135–156. Moore, M. T., Fasciano, N. J., Jacobson, J. E., Lee, V. E., Croninger, R. G., & Smith, J. B. Myers, D., & Waldman, Z. (1997). The (1997). Course-taking, equity, and national evaluation of Upward Bound. A mathematics learning: Testing the con- 1990’s view of Upward Bound: Programs strained curriculum hypothesis in U.S. offered, students served, and opera- secondary schools. Educational Evalua- tional issues. Princeton, NJ: Mathematica tion and Policy Analysis, 19(2), 99–121. Policy Research.

( 77 ) R eferENCES

Moreno, J. F. (2002). The long-term out- postsecondary education: Statistical anal- comes of Puente. Educational Policy, ysis report (NCES 98-082). Washington, 16(4), 572–587. DC: U.S. Department of Education, Na- Myers, D. E., & Moore, M. T. (1997). The tional Center for Education Statistics. national evaluation of Upward Bound: Olsen, R., Seftor, N., Silva, T., Myers, D., Des- Summary of first-year impacts and pro- Roches, D., & Young, J. (2007). Upward gram operations. Washington, DC: Math- Bound math-science: Program description ematica Policy Research. and interim impact estimates. Washing- Myers, D., Olsen, R., Seftor, N., Young, J., & ton, DC: U.S. Department of Education. Tuttle, C. (2004). The impacts of regular Opuni, K. A. (1999). Project GRAD: Gradu- Upward Bound: Results from the third ation really achieves dreams. 1998–99 follow-up data collection. Princeton, NJ: program evaluation report. Houston, Mathematica Policy Research. TX: Houston University. Myers, D., & Schirm, A. (1999). The impacts Pathways to College Network. (2004). A of Upward Bound: Final report for phase shared agenda: A leadership challenge I of the national evaluation. Princeton, to improve college access and success. NJ: Mathematica Policy Research. Boston, MA: Author. National Center for Education Statistics. Pearson. (1996). Stanford Achievement Test (1992). High School and Beyond Lon- Series, Ninth Edition—Complete Battery. gitudinal Study of 1980 Sophomores: Pell Institute for the Study of Opportunity Data Files. in Higher Education. (2006). Expanding National Center for Education Statistics. access and opportunity: The Washington (2002). National Education Longitudi- State Achievers scholarship. Washing- nal Study of 1988: Base-Year to Fourth ton, DC: Author. Follow-up Data Files. Perkins, R., Kleiner, B., Roey, S., & Brown, J. National Center for Education Statistics. (2004). The high school transcript study: (2004). Student effort and educational A decade of change in curricula and progress. The Condition of Education, 60. achievement, 1990–2000 (NCES No. 2004- National Center for Education Statistics. 455). Washington, DC: U.S. Department (2005). Student effort and educational of Education, National Center for Educa- progress. The Condition of Education, tional Statistics. 52–63. Perna, L., Rowan-Kenyon, H. T., Thomas, S. L., National Center for Education Statistics. Bell, A., Anderson, R., & Li, C. (2008). The (2006a). Student effort and educational role of college counseling in shaping col- progress. The Condition of Education, lege opportunity: Variations across high 56–71. schools. The Review of Higher Education, National Center for Education Statistics. 31(2), 131–160. (2006b). Education Longitudinal Study Plank, S. B., & Jordan, W. J. (2001). Effects of 2002: Base-Year to Third Follow-up of information, guidance, and actions Data Files. on postsecondary destinations: A study National Center for Education Statistics. of talent loss. American Educational Re- (2008). Student effort and educational search Journal, 38(4), 947–979. progress. The Condition of Education, Quigley, D. D. (2003). The Early Academic 32–43. Outreach Program (EAOP) and its impact National Conference on Citizenship. (2006). on high school students’ completion of America’s civic health index: Broken en- the University of California’s prepara- gagement. Washington, DC: Author. tory coursework (CSE Tech Report No. Nunez, A., & Cuccaro-Alamin, S. (1998). 589). Los Angeles, CA: University of First-generation students: Undergradu- California, CRESST. ates whose parents never enrolled in

( 78 ) R eferENCES

Quint, J., Thompson, S. L., & Bald, M. (2008). socioeconomic status urban youth. Relationships, rigor and readiness: Strat- Education and Urban Society, 39(1), egies for improving high schools. New 128–146. York: Manpower Demonstration Re- Spielhagen, F. R. (2006). Closing the achieve- search Corporation. ment gap in math: The long-term effects Roads to Success: Program evaluation re- of eighth-grade algebra. Journal of Ad- port, fall 2008. (2008a). New York: Roads vanced Academics, 18 (1), 34–59. to Success. Standing, K., Judkins, D., Keller, B., & Roads to Success: Program evaluation re- Shimshak, A. (2008). Early outcomes port #2, December 2008. (2008b). New of the GEAR UP program: Final report. York: Roads to Success. Washington, DC: U.S. Department of Robinson, S., Stempel, A. & McCree, I. Education. (2005). Gaining traction, gaining ground: St. John, E. P., Musoba, G. D., Simmons, A. B., How some high schools accelerate learn- & Chung, C. (2002). Meeting the access ing for struggling students. Washington, challenge: Indiana’s Twenty-First Cen- DC: The Education Trust. tury Scholars program. [New agenda Roderick, M., Nagaoka, J., Coca, V., & series, 4(4)]. Indianapolis, IN: Lumina Moeller, E. (2008). From high school to Foundation for Education. the future: Potholes on the road to col- Texas Education Agency. (2002). Texas As- lege. Chicago, IL: Consortium on Chi- sessment of Academic Skills. cago School Research. Tierney, W. G., & Jun, A. (2001). A univer- Schirm, A., Stuart, E., & McKie, A. (2006). sity helps prepare low-income youths The Quantum Opportunity Program for college: Tracking school success. demonstration: Final impacts. Princeton, Journal of Higher Education, 72(2), NJ: Mathematica Policy Research. 205–225. Seftor, N. S., Mamun, A., & Schirm, A. Tomas Rivera Policy Institute. (2004). (2009). The impacts of regular Upward Caught in the financial aid information Bound on postsecondary outcomes 7–9 divide. Los Angeles, CA: Author. years after scheduled high school gradu- U.S. Census Bureau. (2002). The big payoff: ation. Princeton, NJ: Mathematica Policy Educational attainment and synthetic Research. estimates of work-life earnings. Wash- Siskin, L. S., & Weinstein, M. (2008). Sup- ington, DC: Author. plemental survey to “Creating support U.S. General Accounting Office. (1990). structures and services for Title I high Higher education: Gaps in parents’ and schools implementing the International students’ knowledge of school costs and Baccalaureate programs.” New York: In- federal aid (No. GAO/PEMD-90-20BR). stitute for Education and Social Policy. Washington, DC: Author. Smith, T. M., Young, B. A., Bae, Y., Choy, S. Venezia, A., & Kirst, M. W. (2005). Inequi- P., & Alsalam, N. (1997). The condition table opportunities: How current edu- of education (NCES 97-388). Washington, cation systems and policies undermine DC: U.S. Department of Education. the chances for student persistence and Snipes, J. C., Holton, G. I., Doolittle, F., & Sz- success in college. Educational Policy, tejnberg, L. (2006). Striving for student 19(2), 283 –307. success: The effect of Project GRAD on high Watt, K. M., Huerta, J., & Lozano, A. (2007). school student outcomes in three urban A comparison study of AVID and GEAR school districts. New York: Manpower UP 10th-grade students in two high Demonstration Research Corporation. schools in the Rio Grande Valley of Sokatch, A. (2006). Peer influences on Texas. Journal of Education for Students the college-going decisions of low Placed At Risk, 12(2), 185–212.

( 79 ) R eferENCES

Watt, K. M., Powell, C. A., Mendiola, I. D., Wimberly, G. L., & Noeth, R. J. (2005). College & Cossio, G. (2006). Schoolwide impact readiness begins in middle school: ACT and AVID: How have selected Texas high policy report. Iowa City, IA: ACT, Inc. schools addressed the new accountability measures? Journal of Education for Students Placed At Risk, 11(1), 57–73.

( 80 )