1

BACKGROUNDER: INDICATORS OF FUTURE SUCCESS: GPA AND NONCOGNITIVE SKILLS

U nder No Child Left Behind (NCLB), the federal government, states, and districts have used a

single indicator to hold schools accountable for student success. Some go as far as to generate district and school report cards predominantly based on standardized test scores and shutting down those schools that fail to make satisfactory grades. NCLB accountability, however, has not yielded improvements in the number of students succeeding, if success is measured by the number of students graduating from college within six years. Data reveal only a 2.0 percentage point increase from 2002 to 2006 in the total number of full-time students who began seeking a bachelor’s degree at a four-year institution and completed that degree within six years—an insignificant difference when compared to the 2.1 percentage point increase from the pre-NCLB cohorts—1996 to 2000.1 Further, NCLB had no impact on the achievement gap: The Black-White six-year college-completion gap for the pre-NCLB 1996 cohort was 19.20 points compared to the 22.30 point gap for the 2006 cohort. These data also hold true for graduation within four years. Policymakers and stakeholders should move from a single-indicator system and employ multiple measures highly predictive of future student success: grade point average (GPA) and noncognitive skills, for example. True Predictors of Success

GPA Schools must be held accountable for the success of all students. Instead of focusing on standardized test scores, however, students would benefit more if schools focused on reliable predictors of future student success. More accurate predictors of success are GPA and noncogni-tive skills. Researchers argue that GPA may be a strong predictor of success due to its ability to capture content knowledge and skills critical to success, such as persever-ance and self-control. A 2014 study conducted by William C. Hiss and Valerie W. Franks examined the outcomes of optional standardized testing policies at 33 public and private colleges and universities. They concluded that standardized tests are more likely to decrease the number

NEA Education Policy and Practice Department Center for Great Public Schools 1201 16th St., NW, Washington, D.C. 20036

18413.7/15.vn

College Graduation Gap - 4 years

Hispanic-White Gap

Black-White Gap

American Indian/AlaskaNative-White Gap

25

20

15

10

5

0

1996 starti

ng cohort

2000 starti

ng cohort

2002 starti

ng cohort

2003 starti

ng cohort

2004 starti

ng cohort

2005 starti

ng cohort

2006 starti

ng cohort

College Graduation Gap - 6 years

Hispanic-White Gap

Black-White Gap

American Indian/AlaskaNative-White Gap

25

20

15

10

5

0

1996 starti

ng cohort

2000 starti

ng cohort

2002 starti

ng cohort

2003 starti

ng cohort

2004 starti

ng cohort

2005 starti

ng cohort

2006 starti

ng cohort

2

of college applicants than produce valuable predictive results.2 A better predictor of success, according to the study, is GPA, as college and university cumulative GPAs closely tracked high school GPAs (HSGPAs).

“College and university Cumulative GPAs closely track high school GPAs, despite wide variations in testing. Students with strong HSGPAs generally perform well in college, despite modest or low testing. In contrast, stu-dents with weak HSGPAs earn lower college Cum GPAs and graduate at lower rates, even with markedly stronger testing.”3

Further, university students with stronger high school GPAs graduated from college at higher rates than stu-dents with weaker high school GPAs but much stronger testing.

Although the study refers to college-entrance standard-ized tests like the Scholastic Aptitude Test (SAT) and the American College Testing (ACT), the results demonstrate standardized test limitations in predicting success. Ad-ditionally, Geiser and Santelices of Berkeley’s Center for Studies in Higher Education, confirm high-school GPA is consistently the best predictor of four-year college outcomes:

“HSGPA is consistently the best predictor, followed by SAT II Writing scores, for both first- and fourth-year college grades, and this pattern holds for all entering cohorts, UC campuses and academic fields, with only minor excep-tions…. High-school grades provide a fairer, more equi-table and ultimately more meaningful basis for admission decision-making and, despite their reputation for ‘unreli-ability’, remain the best available indicator with which to hazard predictions of student success in college.”4

Noncognitive SkillsGPA has proven successful at predicting future success because of its capacity to capture both cognitive skills and those critical to success (i.e. noncognitive ability).5 GPA not only captures mastery of content knowledge, but also skills not usually captured in tests—self-control, for example.6 Self-control is an example of a noncognitive skill. Noncognitive skills—sometimes referred to as “social and emotional learning,” “soft skills,” and/or “meta-cog-nitive learning skills”—consist of the skills not captured in cognitive tests such as aptitude tests, standardized tests, or course exams. Nonetheless, these skills are critical to academic success: earning course credits, for example, requires a set of behavioral skills, including self-regulation. Mastery of such skills has proven predictive of future ac-ademic success, without such skills, changes in cognitive ability may be unlikely.7 Growth in noncognitive skills has been tied to increases in course grades and future educa-tional attainment. Noncognitive skills—depending on the study one employs—can include self-control, persistence, grit, optimism, curiosity, conscientiousness, social fluidity, and self-confidence—a range of personality and motiva-tional habits.8

Variables Impacting Noncognitive Skills

Noncognitive skills begin to develop in infancy, continue to develop through adulthood, and are malleable. This means that students can improve their cognitive skills and improve their chances at academic success through vari-ous programs (e.g. early childhood education and physical education) and strategies. Many noncognitive skills such as self-control, conscientiousness, self-confidence, and social fluidity are related to executive function, which is primarily located in the prefrontal cortex of the brain. The prefrontal cortex is critical in emotional and cognitive self-regulation, also known as executive function. The prefrontal cortex is most affected by childhood trauma and stress, and chil-dren raised in poverty are more likely to experience early trauma and or stress.9 In fact, gaps in noncognitive and cognitive skills between disadvantaged students and their peers can be traced back to adverse early environments.10 Trauma and stress negatively impact the development of noncognitive skills, but such impacts can be mitigated and skills can be improved through strategies and programs, including early childhood education.

Research conducted by James Heckman formalizes the notion that noncognitive skills foster the acquisition of

3

cognitive skills, and parental inputs affect the formation of both.11 Previous studies on the Abecedarian Project, Perry Preschool, Nurse Family Partnership program, and the Chicago Parent-Child program demonstrate the capability of early childhood investments to increase parental inputs and positively impact noncognitive skills. Treated students were more likely to graduate from high school, obtain more years of education, and attend a four-year college. Although early investments are expected to produce pos-itive returns, returns can only be maximized if early invest-ments are followed by later investments.

“But it is also essential to invest late to harvest the fruits of the early investment. Such dynamic complementari-ty helps to explain the evidence by Currie and Thomas (1995) that for disadvantaged minority students, early investments through Head Start have weak effects in later years if not followed up by later investments.”12

Aside from investments in early childhood education, which has been proven to increase a child’s chances at

future success, noncognitive skills can be improved further through a range of later stage interventions. Past successful interventions include implementing physical education programs, which stimulate the growth of con-nections in the prefrontal cortex and improve executive function.13 The following table is an example of some of the noncognitive skills-based strategies proven to raise student grades and GPAs. These strategies include: 1) dis-cussing how the brain grows new connections and “gets smarter;” 2) praising a student’s efforts, rather than abil-ities; 3) emphasizing communal and cooperative goals over individualistic and competitive goals; 4) discussing long-term goals such as completing college; and 5) stressing contributions to society over self-oriented goals.

The below chart only captures a portion of the studies and strategies conducted in the field, but nonetheless, prove noncognitive skills are indeed malleable and impact cog-nitive ability, as measured by GPA and individual course grades. In addition to the strategies employed in the below chart, noncognitive skills are impacted by additional

Selected Interventions with Academic Outcomes

AUTHORS RESEARCH PARTICIPANTS EFFECT (RELATIVE TO CONTROL GROUPS)

Teaching students that intelligence can be developed (a growth mindset)

Blackwell, Trzesniewski, & Dweck (2007)

Urban, low-income, African American and Latino 7th grade students

High math grades

Good, Aronson, & Inzlicht (2003)

African American and Latino middle school students at a rural school

Higher state test scores for all in reading and for girls in math

Aronson, Fried, & Good (2002) African American and white college students

Higher GPA; greater vauing and enjoyment of academics

Helping students to feel that they belong or are valued in school

Walton & Cohen (2007) African American college students Higher GPA

Cohen et al. (2006; 2009) African American and white middle school students

Higher grades among African American students in the targeted class

Helping students to see how the curriculum is relevant to their own lives

Hulleman & Harackiewcz (2009)

White, African American, Latino, and Asian high school students

Higher grades in the targeted class among students with low initial expectations of success

Helping students to set goals, identify obstacles, and learn self-control strategies

Oyserman, Bybee, & Terry (2006)

African American and Latino middles school students

Higher grades; fewer absences; fewer disciplinary referrals

Brigman & Webb (2007) Students in grades 5-9 who scored below 50th percentile on state math test

Higher state test scores in reading and math

Source: Dweck, C.S., Walton, G.M., & Cohen, G.L. (2011). Academic Tenacity: Mindsets and Skills that Promote Long-Term Learning. White Paper prepared for the Gates Foundation.

NOTE: All interventions were controlled randomized trials

4

variables such as those listed below. The chart below lists some of the factors impacting noncognitive skills. Both charts demonstrate that a purposeful focus on improving noncognitive skills through school resources and practic-es impact the growth of noncognitive skills.

Noncognitive Skills in the Field

The focus on noncognitive skills, particularly through early education programs, has shown promising results. Early childhood education has proven successful at addressing deficiencies in noncognitive skills and raising cognitive test scores. Heckman’s study of Perry Preschool uncovered early childhood education’s ability to increase motivation and reduce negative externalizing behaviors, positively impacting outcomes in education, employ-ment, and health.21 It is no surprise that strategies and programs to develop noncognitive skills have grown in popularity. A Mathematica study found that over 195 in-struments, including surveys, are used to measure growth in noncognitive skills.22 Current examples of schools and organizations examining and measuring noncognitive skills include the California Office to Reform Education (CORE) and KIPP charter schools. These groups vary in

the data they collect and in the methods used to track and improve growth.

CORECORE*, a nonprofit organization composed of ten Califor-nia school districts, believes noncognitive skills are integral to helping students succeed in college, career, and life.23

CORE focuses on four noncognitive skills: growth mind-set (a student’s belief about their ability to achieve and the connection between effort and success), self-efficacy, self-management, and social awareness. CORE collects data on noncognitive skills through the use of self-report student surveys. CORE piloted student self-report surveys in the spring of 2014 and plans to begin reporting on the four areas in the spring of 2016. School effectiveness will be determined by the following:

Academics (60 percent) 1. Standardized test scores 2. Graduation rates 3. Middle school persistence (rate of 8th graders who go on to 10th grade)

Social and Emotional Factors (20 percent) 1. Chronic absenteeism rates 2. Suspension/Expulsion rates 3. Noncognitive factors

Indicator Rationale

Early Childhood Education Impacts the development of noncognitive skills.14 Noncogni-tive skills impact cognitive ability.

Parent Engagement Associated with lower rates of high school dropout, increased on-time high school completion, highest grade completed, and has been found to have a strong association with non-cognitive skills such as self-esteem and susceptibility to peer pressure.15 Heckman also finds that parental inputs affect the formation of noncognitive skills.16

School Climate There is a significant relationship between school climate and student engagement. Engagement is a noncognitive skill and a predictor for academic success.17

Class Size Class size impacts noncognitive skills such as student en-gagement.18

Specialized Instructional Support Personnel (SISP) SISP, specifically mental health services, have the ability to mitigate childhood trauma and, thus, improve noncognitive skills.19

Physical Education Physical education along with moderate to vigorous physical activity has been proven to strengthen executive function.20

5

School Culture and Climate Factors (20 percent) 1. Student, staff, and parent surveys 2. English language redesignation rates 3. Rates of identifying special education students

KIPPKIPP* charter schools only recently began to report on seven noncognitive skills: zest, grit, optimism, self-control, gratitude, social intelligence, and curiosity.24 One method that KIPP uses to promote growth in the seven character strengths is to require educators to model the seven skills in the classroom. Growth is then measured on a Character Growth Card.

Schools or organizations seeking to implement programs and/or strategies that impact noncognitive skills and em-ploy noncognitive skills as a tool to inform education prac-tice and policy should be aware, however, that the use of self-report surveys to measure such skills may produce paradoxical results due to reference bias, according to a study conducted by West et al.25 Another way to capture data on noncognitive skills might be to combine surveys or indexes with data on impactful variables such as parent engagement and early childhood education. Further, there is no consensus on which noncognitive skills are the most salient, which strategies are the most effective, nor is there consensus on a preferred measure of noncognitive skills.

Moving Forward

NCLB accountability proved unable to predict and grow student success. GPAs’ capacity to capture salient skills and predict future academic success, combined with noncognitive skills’ predictive power, makes GPA and measures of noncognitive skills more robust accountabil-ity tools than standardized test scores and instruments

worthy of focus. Previous research demonstrates that a focus on noncognitive skills increases cognitive skills and, ultimately, narrows the achievement gap.26 Policy makers and practitioners would be able to narrow achievement gaps effectively and better predict future success by work-ing on noncognitive skills and giving greater weight to GPA. Schools and districts must be held accountable for student success—success beyond increasing standardized test scores.

Families and communities would be better served if state, district, and school report cards moved from using a single indicator of success to multiple indicators. These report cards should be expanded to include data on stu-dent GPA, as well as noncognitive skills scores, disaggre-gated by race, income, gender, ability, and migratory sta-tus. Further, stakeholders would be better able to pinpoint the factors influencing student GPA and noncognitive skills if report cards were expanded to include data on the variables known to impact noncognitive skills. Salient indicators include: 1) percentage of students enrolled in early childhood education; 2) parent engagement; 3) school climate; 4) access to SISP; 5) class size; and 6) physical education. Additional indicators can be found at http://www.nea.org/gpsindicators. Finally, accountability report cards must also include data on students’ access to college gateway courses like algebra I, algebra II, geome-try, trigonometry, calculus, biology, chemistry, and physics. College gateway courses provide the foundation needed to achieve college success, noncognitive skills improve a student’s chances at success, and GPA better predicts success.

The lessons from NCLB are clear: standardized tests have limitations, more robust predictors of academic success exist, and multiple measures of success can provide a clearer picture of what is happening in a school, in com-parison to test scores alone.

*The examples from CORE and KIPP were chosen based on their use of noncognitive factors and should not be construed to represent NEA policy or position.

6

Notes

1 Graduation rates of first-time, full-time bachelor’s degree-seeking students at four-year postsecondary institutions, by race/ethnicity, time to completion, sex, and control of institution: Selected cohort entry years, 1996 through 2006. (2014, January 1). Retrieved March 18, 2015, from http://nces.ed.gov/programs/digest/d13/tables/dt13_326.10.asp2 Hiss, W., & Franks, V. (2014, February 5). Defining Promise: Optional Standardized Testing Policies in American College and University Admissions. Retrieved March 18, 2015, from http://www.nacacnet.org/research/research-data/nacac-research/Documents/DefiningPromise.pdf3 Ibid4 Geiser, S., & Veronica Santelices, M. (2007, June 1). Validity of High-School Grades in Predicting Student Success Beyond the Freshman Year: High-School Record vs. Standardized Tests as Indicators of Four-Year College Outcomes. Retrieved April 6, 2015, from http://www.cshe.berkeley.edu/sites/default/files/shared/publications/docs/ROPS.GEISER._SAT_6.13.07.pdf5 Heckman, J. (2008). Schools, Skills, and Synapses. Economic Inquiry, 289-324.6 Duckworth, A., Quinn, P., & Tsukayama, E. (2012). What No Child Left Behind leaves behind: The roles of IQ and self-control in predicting standardized achievement test scores and report card grades. Journal of Educational Psychology, 104(2), 439-451.7 Bransford, J.D., Brown, A.L., and Cocking, R.R. (2000) How people learn: Brain, mind, experience, and school. Washington, DC: National Academy Press8 Rosen, J., Glennie, E., Dalton, B., Lennon, J., & Bozick, R. (2010, September 1). Non-cognitive Skills in the Classroom: New Perspectives on Educational Research. Retrieved April 28, 2015, from https://www.rti.org/pubs/bk-0004-1009-rosen.pdf9 Felitti, V. J. and R. F. Anda. The Adverse Childhood Experiences (ACE) Study. Centers for Disease Control and Kaiser Permanente, 2005. Cunha, F., J. J. Heckman, L. J. Lochner, and D. V. Masterov. Burke, N., Hellman, J., Scott, B., Weems, C., & Carrion, V. (n.d.). The Impact Of Adverse Childhood Experiences On An Urban Pediatric Population. Child Abuse & Neglect, 408-41310 Heckman, J. J. (2012). The Case for Investing in Disadvantaged Young Children. European Expert Network on Economics of Education.11 Cunha, F., & Heckman, J. (2006). Formulating, Identifying and Estimating the Technology of Cognitive and Non-cognitive Skill Formation. Journal of Human Resources, 738-782.12 Ibid13 Colcombe, S.J. 2004. “Cardiovascular fitness, cortical plasticity, and aging.” Proceedings of the National Academy of Science, 101(9): 3316.14 Heckman, J., Stixrud, J., & Urzua, S. (2006). The Effects of Cognitive and Non-cognitive Abilities on Labor Market Outcomes and Social Behavior. Journal of Labor Economics, 24(3).15 Wendy Miedel Barnard, “Parent Involvement in Elementary School and Educational Attainment,” Children and Youth Services Review 26 (2004): 39–62. Marjory R. Gray and Laurence Steinberg, “UnpackingAuthoritative Parenting: Reassessing a Multidimensional Construct,” Journal of Marriage and Family 61, No. 3 (August 1999): 574–587.16 Cunha, F., & Heckman, J. (2006). Formulating, Identifying and Estimating the Technology of Cognitive and Non-cognitive Skill Formation. Journal of Human Resources, 738-782.17 Eccles, J.S., Wigfield, A., Midgley, C., Reuman, D., MacIver, D., & Feldlaufer, H. (1993). Negative Effects of Traditional Middle Schools on Students’ Motivation. Elementary School Journal, 93, 553–574.18 Dee, T. and West, M. (2008). The Non-Cognitive Returns to Class Size. NBER Working Paper 13994.19 Dorsey, S., Briggs, E., & Woods, B. (2011). Cognitive-Behavioral Treatment for Posttraumatic Stress Disorder in Children and Adolescents. Child and Adolescent Psychiatric Clinics of North America, 255-269. doi:10.1016/j.chc.2011.01.00620 Colcombe, S.J. 2004. “Cardiovascular fitness, cortical plasticity, and aging.” Proceedings of the National Academy of Science, 101(9): 3316.

7

Notes (continued)

21 Heckman, J., Pinto, R., & Savelyev, P. (2013). Understanding the Mechanisms through Which an Influential Early Childhood Program Boosted Adult Outcomes. American Economic Review.22 Atkins-Burnett, S., Fernández, C., Akers, L., Jacobson, J., & Smither-Wulsin, C. (2012, July 30). Landscape Analysis of Non-Cognitive Measures.23 Social-Emotional Skills. (2014, September 24). Retrieved March 11, 2015, from http://coredistricts.org/wp-content/uploads/2014/10/SE-CC-Domain-Social-Emotional-Skills-updated-10.3.14-1.pdf24 Character & Academics - KIPP Public Charter Schools | Knowledge Is Power Program. (n.d.). Retrieved March 11, 2015, from http://www.kipp.org/our-approach/character25 Martin R. West, et al., “Promise and Paradox: Measuring Students’ Non-cognitive Skills and the Impact of Schooling” (Cambridge, MA: Harvard Graduate School of Education, February 2014).26 Dweck, C., Walton, G., & Cohen, G. (n.d.). Academic Tenacity: Mindsets and Skills that Promote Long-Term Learning. Retrieved June 1, 2015, from https://web.stanford.edu/~gwalton/home/Welcome_files/DweckWaltonCohen_2014.pdf

8