Agenda for Alabama’s Third Century: Raising Educational Attainment for All

While Alabama continues to gain on other states in the percentage of adults with a high school diploma, it is falling further behind the national average when it comes to the percentage of the population with a bachelor’s degree or better.

This is important because higher educational attainment is correlated with higher rates of labor force participation, higher personal income, and higher GDP per capita, as well as, better health outcomes.

The visualization below shows the correlation between a state’s per capita income and the percentage of adults (individuals 25 years or older) who hold a bachelor’s degree or higher. The greater the percentage of college-educated adults in the state population, the higher the state’s per capita income. And vice-versa. The correlation holds true for the percentage of the population with an associate degrees.

In response to changes in demographics and the economy, Alabama leaders launched a full-court press to raise levels of educational attainment, recognizing that as a cornerstone of future prosperity. The Success Plus initiative brings government, education, and business together around a coordinated effort to raise educational attainment and post-high school advanced skills training, with the goal of adding 500,000 highly skilled workers to the workforce by 2025. In the face of the Coronavirus pandemic, it’s all the more important to sustain educational momentum and prevent the exacerbation of educational inequities that have held Alabama back.

PARCA opened 2020 with an annual meeting in January that challenged speakers to sketch out a vision for our state’s third century. We’re revisiting the central themes explored at that meeting in the coming months, beginning with this installment on educational attainment.

From PARCA’s work with schools and governments, successful organizations follow a process in order to pursue a vision.

  1. Know where you stand and own your data.
  2. Make a plan for improvement, one that includes goals and evidence-based strategies for how to achieve them.
  3. Measure progress toward your goal. Analyze the results.
  4. Celebrate successes. Shift resources and strategy if tactics aren’t working.

This report, as a first step, describes where Alabama stands in comparison to other states in educational attainment.

Narrowing the Gap in High School Attainment

It’s not a surprise that Alabama trails other states. The state is dragged down by its historic underinvestment in education, by the legacy of racially-segregated, separate and unequal schools, as well as continuing inequities and de-facto segregation in some areas.

Here’s the good news: Alabama has shown it can produce dramatic change. When it comes to raising the percentage of adults with high school education, it has, according to Census Bureau data. 1

  • Between 1940 and 2018, Alabama ranked third highest among states in percentage point gain in high school attainment.
  • By 2018, Alabama had nearly closed the gap with the US average for the percentage of the state’s adult population with a high school degree.
  • In 2018, 87% of Alabama adults had a high school degree compared to 88%.

And while work remains to be done, Alabama has also made significant progress in closing the gap between Blacks and whites in high school attainment. In 1940, after decades of Black flight from the South and with segregation still in full force, only 4% of Black adults in Alabama had a high school diploma, compared to 22% of whites. Both races trailed well behind the national averages. By 2018, 82% of Black adults in Alabama had a high school credential compared to 89% of whites. Across the U.S., 85% of Blacks and 93% of non-Hispanic whites have a high school credential.

In spite of advances, Alabama still ranks No. 44 among the 50 states in the percentage of its population with a high school degree, thought the gap between all states is much narrower than it once was.

Falling Behind in Bachelor’s Degree Attainment

The gap is much wider when it comes to the percentage of the population with a four-year degree. When it comes to bachelor’s degree attainment, Alabama has failed to keep pace with the growth rates in other states.

  • In 1940, the percentage of the population with a college degree was small in Alabama (3%) and the U.S. (5%).
  • Since 1940, the percentage of the population with a college degree has steadily increased but Alabama’s growth rate has been slower than the national rate, resulting in the current gap of 7 percentage points.
  • Since 1940, Alabama had the ninth-lowest percentage point gain among states in bachelor’s degree attainment and ranks No. 44 among the 50 states in the percentage of the adult population with a degree.

Use the menu to select other states for comparison

Alabama’s bachelor’s degree attainment trails the national average for both Blacks and whites. Among Alabama whites, 28% of adults have a bachelor’s degree or higher, compared to 36% of whites nationally. That’s a gap of 8 percentage points.

In Alabama, among Blacks, 17% of adults have a bachelor’s degree or higher, compared to 22% nationally. That’s a gap of five percentage points. Throughout this period, more students began enrolling and graduating from college. However, white students nationally and in Alabama seem to have been better positioned to take advantage of this major shift in higher education.

Nationally, the gap in bachelor’s attainment between whites and Blacks is 14 percentage points, compared to a gap of 11 percentage points in Alabama. For both races, the gap with the US average has widened between 2010 and 2018, according to the Census statistics. The percentage point growth in attainment was particularly slow for Alabama Blacks during the latest period, advancing only two percentage points.

These educational attainment measures are not solely attributable to the preparation levels and graduation rates from Alabama K-12 schools or colleges, though the performance of those schools has an effect. Also influencing the equation, particularly in terms of bachelor’s degree attainment, is where Alabama graduates move after college, which is often a function of where career opportunities are available, and the population and wages are growing. Data suggests that many of the states seeing rapid increases in the percentage of the population with bachelor’s degrees are drawing college-educated migrants from other states. Alabama research from the Alabama Commission on Higher Education also finds that the state exports college graduates.

If you want a closer look at how the state’s compare on educational attainment, the visualizations below present a more detailed view. The charts break down the population into segments based on the percentage of the population in each educational category. An individual’s category is determined by his or her highest level of education. The percentage represents the number of people at that level each attainment divided by the total population 25 years and older. In this chart, the states are sorted by the percentage of the population with an associate degree or higher. Use the tabs to explore the states ranked by the percentage of the population with bachelor’s and graduate degrees or the states’ data displayed in an interactive map.

How Does Educational Attainment Compare within the State?

Overall, Alabama’s educational attainment rates may trail the rates of other states, but measures of attainment also vary widely within the state. Educational attainment, as measured by the percentage of the population with bachelor’s degrees or higher, exceeds the US average in five counties: Madison, Shelby, Lee, Jefferson, and Baldwin. Five additional counties exceed the Alabama average: Montgomery, Tuscaloosa, Autauga, Coffee, and Elmore Counties. Those top counties represent most of the major metro population centers plus the university towns. The exception is Coffee County which is home to Enterprise State Community College and includes the residences of many of the personnel, military and civilian, associated with Fort Rucker.

Within the top 20 are a few counties where educational attainment levels are higher than one might expect based on their rural character: Sumter, Pike, and Macon. On closer examination, all three have a university and a relatively small population. Macon County also has a large Veterans Administration Hospital. Pike County is home to some defense contractors and other manufacturing businesses that likely boost the county’s share of credentialed residents.

As in the national context, there is a relationship between education levels and income. In Alabama, the most predictive relationship can be found by comparing county per capita income and the percentage of the population with an associate degree or higher. In general, counties with higher levels of educational attainment have higher levels of income. Educated workers seek out locations where there are concentrations of higher-paying jobs. Higher-paying jobs generally require higher levels of education or training. Businesses seek out locations where there is an ample supply of educated residents so hiring will be easier. Higher incomes generate more business opportunities.

The quest to close the gap

For its first 150 years of statehood, Alabama intentionally operated separate and unequal schools for blacks and whites. And even for whites, the schools weren’t adequate to keep students on pace with the rest of the United States. Through the 1950s, Alabama’s population was predominately rural with an economy oriented to agriculture or blue-collar industry. Alabama’s primary recruiting tools were low taxes, low wages, and cheap land.

The lingering effects of this historic underinvestment in students become apparent in the period of record-low unemployment prior to the pandemic. Despite a booming economy and low unemployment, Alabama continued to have one of the lowest labor force participation rates in the U.S. Businesses struggled to find employees to meet the demands of more technologically advanced workplaces, while a disproportionate number of citizens, primarily those with lower levels of education, stayed stuck on the sidelines.

If Alabama is to reach its full economic potential, the state needs to continue to improve the equity, effectiveness, and efficiency of its educational delivery system. It needs to reach out to adults who need to advance their education. And it needs to retain the graduates its schools produce.


Alabama First Class Pre-K Research Gains National Exposure

The National Institute of Early Education Research (NIEER) at Rutgers University has recognized research on the persistence of the impact of First Class Pre-K.

Researchers from the University of Alabama at Birmingham, PARCA, ThinkData, and the Alabama Department of Early Childhood Education published a peer-reviewed article in the International Journal of Child Care and Education Policy showing the persistence of gains in math and reading for students who received First Class Pre-K. This article is the culmination of several years of work involving five cohorts of students over time.

First Class Pre-K students in Collier Elementary School in Mobile County public schools. al.com, March 7, 2019

The article, written for an academic audience, offers a detailed statistical analysis of recent findings of the research team already shared with policymakers and pre-K advocates in Alabama and authored by UAB, the Alabama Department of Early Childhood Education, and PARCA.

Results indicate that children who received First Class Pre-K were statistically significantly more likely to be proficient in both math and reading compared to students who did not receive First Class Pre-K. Further, there was no statistical evidence of fadeout of the benefits of First Class Pre-K through the 7th grade, indicating the persistence of the benefits into middle school.

You can read the published article here.

To read previous research about additional academic benefits of Alabama First Class Pre-K, especially as it pertains to educational equity, click here.


COVID-19 and Public Education: Lessons Learned Last Spring

Alabama schools are set to re-open in August, with plans for local systems to offer educational services through traditional on-campus schools, remote on-line education, and a hybrid of traditional and remote learning options.

With the novel Coronavirus still spreading, all plans are subject to change. Already, the state’s largest system, Mobile County, and the Selma City School System have decided not to open school buildings and to proceed with only remote learning for all of its students this fall.

As policymakers, educators, and parents prepare for what will likely be a most unusual school year—including the possibility of additional shutdowns— PARCA gathered information from local reports and two major national polls that attempt to describe what parents and students experienced during the school closures this spring. As schools plan for the fall, these experiences are important to understand.

According to the polls, parents worried the online school experience was resulting in:

  • learning loss and lack of academic advancement
  • a lack of social interaction for students, negatively impacting student mental health
  • inadequate contact between parents and teachers
  • a mismatch between the resources provided by schools and the aid parents most needed
  • increased inequities in the educational experience

The Shutdown

As COVID-19 spread this spring, schools across the country closed. By March 20, 45 states had closed all schools. By early May, the number climbed to 48 states and the District of Columbia—affecting more than 55 million students. Only Montana and Wyoming allowed schools to remain open, although some systems in those states did close. 1

Almost overnight, schools entered uncharted waters. States, systems, and local schools mobilized resources for parents and students and reimagined teacher-student interaction. For most schools, this entailed some version of virtual education.

According to a Gallop Survey conducted in March 2020, 70% of parents of K-12 students not in school at that time reported their child was participating in an online education program run by his or her school. The survey found that among parents whose children were not enrolled in a formal online education program, 52% were homeschooling with their own materials, 25% were using a free online learning program not associated with their child’s school, and 35% were not engaged in any formal education. 2

Some schools had the capacity to respond to COVID-19 closures comparatively easily. That includes schools in Alabama and elsewhere that were already designed as virtual schools. Other systems in other parts of the country are more experienced in online education because of long winter breaks with harsh weather. Conversely, most schools, educators, parents, and students were thrust into a new learning environment for which they were little prepared.

Parents and students around Alabama reported a wide variance in student experiences, varying according to system, school, grade, and teacher. Some reported students having more work than before the shutdown and spending hours each day with regular virtual check-ins. Others reported that work was considered optional or that students finished nine-weeks of work in just a few days. The long-term effects of the academic transition and the inconsistency of students’ experience will take time to assess.

Parent Reactions

The national nonprofit educational organization Learning Heroes conducted a survey of parents in April and March 2020. The survey, which reached 3,645 parents from across the nation, was conducted in conjunction with the Council of Chief State School Officers (CCSSO), National Board for Professional Teaching Standards (NBPTS), and the National Association of State Boards of Education (NASBE).

Results show that parents, now in the role of educators or critical partners in their child’s learning, have gained a new appreciation for what teachers and schools do. 3

Figure 1

Source: 2020 learning Heroes Survey

Some parents reported being overwhelmed, while others reported more involvement in their child’s learning has given them a healthy sense of engagement and better insight on how to help their children learn. These parents look forward to being more involved in schools and their child’s education once schools re-open.

Academic ConcernsLoss of Learning Assessed

Seventy percent of parents expressed concern about the loss of learning and how this will be made up.  Fifty-four percent are concerned their children will not be ready for the upcoming school year. These issues raise more fundamental questions about the nature of teaching and learning. High-quality teaching and learning can presumably occur in different forms. With state testing postponed, measuring the impact on student learning gain or loss will be complicated but is an important objective.

Figure 2

Source: 2020 Learning Heroes Survey

States such as California and South Carolina are planning to implement new tools for assessing learning loss. Quick, real-time assessments conducted by teachers in the classroom, or virtually, will likely be most effective. Assessments that take time to report results will have limited utility for teachers but may be instructive for administrators and researchers.

Researchers have tried to predict the magnitude of pandemic-related learning loss by analyzing normal summer learning loss — the degree of academic regression between the end of one school year and the beginning of the next – and treating the COVID shutdown as an extended summer. Some researchers estimate that students likely ended the school year with only 40% to 60% of learning gains achieved during a typical school year. [efn-note] M., Soland, J., Tarasawa, B., Johnson, A., Ruzek, E., and Liu, J. (2020). Projecting the potential impacts of COVID-19 school closures on academic achievement. Edworking Papers, May 2020. Annenberg Institute for School Reform at Brown University [/efn_note] Other studies estimated much lower losses. 4 5 6

Some experts believe the projected learning loss is over estimated.

They note that estimates using summer loss as a baseline are not taking into account the learning that occurred through virtual forums and support provided by schools this past spring. Likewise, most of the content students were expected to learn was already introduced to students by March, although students did not have an opportunity in class to practice skills, and develop mastery. Furthermore, teachers are prepared to work with students coming back at different levels of preparation after the summer break, so they will not be caught off-guard. At the same time, this will likely be much more challenging and will be taxing for teachers who are less prepared and motivated to work with diverse learners.

These same experts, however, are alarmed about the challenges facing beginning readers, who usually need continued re-enforcement throughout the year. This could affect future literacy rates and have implications for implementing Alabama’s Literacy Act in the lower grades. 7

Social and Mental Health Concerns

Parents expressed fear about the impact of COVID-19 on their children’s social-emotional well-being, and 59% worry about the impact of reduced social interactions. For young children in unsettled or abusive home environments, the school can be a safe place. Long-term absence from this safe place can become a source of heightened trauma with long term consequences.

Many children may be coping well, but medical experts are concerned about the stress and trauma children (and adults) are experiencing during the pandemic, especially those with underlying mental health conditions. School-aged children experienced sudden changes in their educational setting and routines. Many experienced shock. Some families have had the stress of sickness and death in their families as a result of the virus – though overall a relatively small percentage. Many more families are under financial strain. Concerns have been raised about abuse, neglect, loneliness, and isolation. The virus has affected every facet of the life of children and adults. 8

Symptoms of trauma in school-aged children can include:

Physical SymptomsOver-or under-reacting to stimuli (physical contact, doors slamming, sirens)
Increased activity level (fidgeting)
Withdrawal from other people and activities
CognitiveRecreating the traumatic event (e.g., repeatedly talking about or “playing out” the event) or avoiding topics that serve as reminders
Difficulties with attention
Worry and fear about safety of self and others
Disconnected from surroundings, “spacing out”
Social-EmotionalRapid changes in heightened emotions (e.g., extremely sad to angry)
Difficulties with controlling emotions angry outbursts, aggression, increased distress)
Emotional numbness, isolation, and detachment
Language and CommunicationLanguage development delays and challenges
Difficulties with expressive (e.g., expressing thoughts and feelings) and receptive language (e.g., understanding nonverbal cues)
Difficulties with nonverbal communication (e.g., eye contact)
Use of hurtful language (e.g., to keep others at a distance)
LearningAbsenteeism and changes in academic performance/engagement
Difficulties listening and concentrating during instruction
Difficulties with memory 9

Parents can reduce the risk of stress by creating a calm, safe, and predictable environment, communicating and building a positive-supportive relationship with their children, and encouraging their children to develop self-regulation skills.

Teacher Interaction

Parents expressed concern about the lack of regular ongoing contact they and their children have with their children’s teacher(s). This is perhaps less critical for self-motivated students with highly resourceful parents or guardians with time devoted to learning at home. But many students depend on regular high-level teacher interaction. Parents indeed may have the will and skill to perform in this role but are working in fulltime jobs. Others express concern about not having the background to adequately support their children. Still, others may be in stressful life situations that rob them of the motivation and energy to serve in this role. In each of these situations more ongoing contact with teachers and community support specialists would likely make a significant difference.

Figure 3

Source: 2020 Learning Heroes Survey

Though parents find communication with teachers extremely helpful, the majority did not receive this support on an ongoing basis.  Teachers have found themselves in uncharted territory for which they were not prepared. They too may not have the skills and background needed for online teaching and tutoring.  The awkwardness of online communication and technical hiccups can generate additional frustration. Everyone is learning and adapting.

Resources Provided by the School

An especially important issue for schools this past spring was providing guidance and resources to parents to assist them in working with their children. The figure below shows the percent of parents indicating they received key resources from their child’s school during the pandemic this past Spring.

Figure 4 

Source: 2020 Learning Heroes Survey

But sometimes what parents received was not what they needed or found most useful. In Figure 5 below, resources are ordered by the percent of parents who found the assistance useful (red bar), from highest to lowest, and the percent receiving the guidance or resource.  

Figure 5

Source: 2020 Learning Heroes Survey

The most useful assistance included:

  • school provided personal technology
  • online guidance
  • one-to-one tutoring with teachers
  • ongoing regular contact with teachers
  • printed versions of class materials
  • remote classes delivered online

Parents found printed materials more helpful than digital materials.

The gap between what was offered and what was found most useful, when offered, was largest for the following:

  • personal guidance in supporting your child’s learning at home
  • remote one-to-one tutoring by teachers
  • school provided technology
  • access to mental health services

COVID-19 and Equity

A number of observers have focused attention on the profound inequities in education magnified by COVID-19.  Systems vary in funding, resources, curriculum, extracurricular offering, teacher experience and in many other ways. These disparities are likely exacerbated when the home becomes, not by choice, the primary learning environment for all students.  

Virtual education has the potential for system-by-system and house-by-house differences in capacity to compound each other.

Differences in capacity across households include the following:

  • Income and educational attainment of parents.
  • The knowledge and experience of parents, guardians, or other adults.
  • The time and availability of parents, guardians, or other adults to actively facilitate or assist in their children’s learning.
  • Family structure: One and two-parent families where responsibilities are shared.
  • Relationships between parent-child, parent-teacher, and student-teacher.
  • Access to communication, guidance, and support from teachers and schools. 
  • Access to community supports and enrichment.
  • Access to computer technology and high-speed internet. Capacity and motivation to make use of these resources.
  • Access to nutritional food daily.

These obstacles may be greater, but in no way limited, to lower-income areas.

The general public and government leaders most frequently cite a child’s school and teachers as the primary difference in their education. But research has long noted that children do not enter school as a blank slate, and that inequalities begin at birth as a result of different prenatal conditions, and too often are made worse during those early years before school.  Children enter school with vastly different levels of preparation. The achievement gap, from this point of view, is a symptom of broader inequality, past and present. Improving education on campus and on-line and building a solid workforce calls for addressing these inequalities in the home and school. 10 Strauss, V. (2020). How COVID-19 has laid bare the vast inequities in U.S. public education. The Washington Post, April 14, 2020. /efn_note]


School Discipline and Race in Alabama

Black students face harsher disciplinary measures than white students for similar offenses, a PARCA review of public school disciplinary records has found. Meanwhile, Alabama is one of few states nationally, and the only state in the Southeast, that has no uniform statewide policies requiring due process when students are suspended or expelled.

Suspensions have been shown to increase rates of school failure.   

Students with a record of numerous disciplinary infractions are at a higher risk for trouble in school and life. Research has shown that students experiencing early problems with attendance, academics, and discipline more frequently experience negative outcomes, such as leaving high school without a diploma or graduating, but unprepared for college or work. Such problems in school have also been linked to a greater likelihood of poor health and criminal activity.[1]

There is very little evidence showing positive impact of out-of-school suspensions on student behavior, and in many cases, they are more likely to cause more harm than good. Out-of-school suspensions are disproportionately applied to Black students, and potentially contributing to a pipeline that leads from school to prison.

Background

In 2018, PARCA Research Coordinator Joe Adams collected data from the Alabama State Department of Education (ALSDE) that included extensive disciplinary records for the school years, 2014-15, 2015-16, and 2016-17. Using these data, PARCA has found that students who are Black are more likely to receive out-of-school suspensions than white students for the same offense, and students who are white are more likely to receive the less restrictive in-school suspensions than Black students. These findings provoke concern considering the broader context of racial inequities in education, healthcare, criminal justice, corrections, employment, and housing conditions.

In recent years research has found significant differences in the use of suspension and expulsion based on race. These data, along with research showing the long-term negative impact of suspensions and expulsion, have led states to re-examine their disciplinary policies, though Alabama is behind this curve.

A report issued by the Education Commission of the States in 2018 found that most states are limiting the use of suspension or expulsion. [2]

  • Sixteen states, plus the District of Columbia, limit the use of suspension or expulsion by grade level, usually by disallowing suspension or expulsion in the early grades. Alabama’s state regulatory statutes do not currently have such limitations. Though suspensions are more common in middle and high school, they also frequently occur in the lower grades in Alabama.

  • Several states limit the use of exclusionary discipline for certain violations. Of those, about 17 states, plus the District of Columbia, prohibit suspension or expulsion solely for a student’s attendance or truancy issues. This limitation does not exist in Alabama. Suspension is a common disciplinary action for truancy and tardiness in the state.

  • About 30 states, plus the District of Columbia, encourage districts and schools to utilize non-punitive, or more supportive, school discipline strategies. Of those, 22 states, plus the District of Columbia, mention the use of specific, evidence-based interventions — such as schoolwide Positive Behavior Interventions and Supports (PBIS), restorative practices, Response to Intervention (RTI), trauma-informed practices and social-emotional learning. Alabama statutes do not address the school-wide programs cited above, though individual schools are choosing to implement PBIS, restorative justice, and other such programs, and the Alabama State Department of Education developed a guidebook to PBIS.

A student’s right to a fair hearing in the state is another important issue. Based on a 1975 U.S. Supreme Court ruling (Godd v. Lopez, 419 U.S. 565 [1975]), students who are suspended or expelled have a right of due process to defend themselves in a fair hearing. Since this ruling, states have enacted policies to ensure local communities follow through and protect a student’s constitutional right to a hearing. Some states outline very detailed procedures for local boards to use. Most directly state that no student will be suspended or expelled for more than 10 days without an objective hearing. Alabama is one of the few states nationally and the only state in the Southeast that does not have uniform statewide procedures that must be followed before removing a student from school. See https://safesupportivelearning.ed.gov/school-discipline-laws-regulations-state.

Some local boards in the state have developed due process procedures, with varying levels of clarity, and some have not. If a system fails to protect due process, the only recourse is a legal proceeding. Under these circumstances, it is conceivable that a student could be suspended, or even expelled, because of an infraction in which they are innocent, without a hearing and investigation into what happened.

Alabama lawmakers recognize these issues are a problem.

In the 2020 session of the Alabama Legislature Senate Bill 189 proposed limits on suspensions and expulsions. The bill proposed:

  • Requiring local boards of education to hold a hearing when a student is expelled or suspended for more than 10 days.
  • Prohibiting suspension of students enrolled in Pre-K through the fifth grade unless the safety of other students was endangered.
  • Prohibiting suspension for truancy or tardiness.

At the close of the shortened 2020 legislative session, SB 189 had passed the Senate and was referred to the Education Policy Committee in the House of Representatives.

Analysis of Disciplinary Infractions

The dashboard charts below show data on reported infractions in Alabama schools, beginning with the number of reported incidences for all specific infractions collected in the state Student Incident Report (SIR) system. This includes charts showing the infractions with the highest number of incidents, as shown below.

Disturbances or disruptions generated the most reported infractions all three years, followed in 2017 by assault, disobedience, and defiance. These categories give some idea of the offense committed, but with limitations. Local schools and systems may vary in how they interpret these categories. Mystery offenses are unspecified offenses.

The total number of reported infractions by race are shown below. Students who are Black comprise 33% of students in Alabama’s public schools; however, they account for 60% of all reported disciplinary incidents.

Extensive research has demonstrated that students in low-income or poverty households are more likely to experience family stress, verbal and physical abuse, neglect, drug and alcohol use in the home, and other sources of emotional pain and trauma—and may develop different responses or reactions to stressful situations.[3] Cultural conflicts between students and teachers and administrators may also result in a disproportionate number of Black students cited for infractions, as well as teacher-student exchanges that escalate and fail to properly address situations that arise.  [4] [5]

Responses to Disciplinary Incidents in Alabama Public Schools

The following dashboard charts show the punitive actions (dispositions) for all offenses in Alabama, as found in the data provided to PARCA by ALSDE.

Among the incidents that are reported to the state in the SIR data, the preferred disposition for responding to the vast majority of incidents was suspension. Out-of-school suspensions are more common than in-school suspensions, though the 48-40% split is not an alarmingly large gap. Both are used frequently.

More problematic are dispositions by race. The following dashboard chart displays state-level disposition data for Black and white students, divided between schools that utilize corporal punishment and those that do not. In Alabama, systems that use corporal punishment tend to be majority white systems. Majority Black systems are more likely to prohibit corporal punishment. Systems with greater racial diversity are mixed in their use of corporal punishment. Consequently, by default, Black students are less likely to receive corporal punishment.

The proportion of Black students receiving out-of-school suspensions is markedly higher than for white students, who are more likely to receive the less severe in-school suspensions.  

These patterns hold when looking at dispositions for specific infractions committed by both Black and white students. Looking at dispositions recorded for nearly 60 infractions, in 90% of the infraction types, Black students were more likely to receive an out-of-school suspension than white students for the same infraction.

The dashboard charts show the percentage of students receiving out of school suspension by race for the most common infractions.

Infractions are ordered from the highest gap to the lowest between the percentage of offenses cited for Black and white students punished with out-of-school suspensions for the same offense. This order varies somewhat each year. Black students are more likely to receive an out-of-school suspension than white students for all of the 10 most common infractions.

Among the most frequently occurring infractions, the gap between cases in which Black and white students received out-of-school suspension was largest in 2017 for tobacco-related offenses and smallest for drug-related offenses. In 2016, the gap was largest for disobedience. Infractions for disobedience resulted in out-of-school suspension in 19% of the cases involving white students, but when Black students committed the offense, out-of-school suspension was used in 47% of the cases.

This is a gap of 28 percentage points – showing that Black students were twice as likely to be removed from school than white students for this same offense.

The charts below provide more detail for different infractions. Use the filters to explore results for all infractions. Examples are posted below.

Suspensions – Background and Research

In the 1990s and early 2000s, schools across the United States employed exclusionary discipline, namely in-school and out-of-school suspensions, at increasing rates. Suspensions are a punishment for “bad behavior” and may have also been seen as a “cooling off period” for students. In other situations, suspensions may be used to bring more control to a hostile situation and create easier conditions for maintaining a safe learning environment. In any case, suspensions are a signal to parents/guardians that their children are demonstrating behavioral problems.

With in-school suspensions, students are removed from their regular class but still attend school in a designated classroom or another school in the system. Thus, the students are still receiving some form of instruction and supervision. Out-of-school suspension prohibits the student from attending school for a temporary period, usually five or fewer days. Very serious infractions may result in expulsion—removing the student from the school system.

Clearly, there are situations where the safety of students and school personnel must take priority and call for removing a student from school grounds. However, there is very little evidence supporting the positive impact of suspensions on student behavior.[6] Researchers documenting the increased use of suspensions over the last three decades found they were not effective in changing student behavior and were associated with other negative outcomes, including lower academic achievement, grade retention, increased drop-out rates, and involvement with the juvenile justice system.

Critics argue that suspensions are adversely related to student learning. Suspensions remove students from the classroom or an optimal learning environment. Achievement gaps are widened as out-of-school suspensions are disproportionately given to students who are male, Black, economically disadvantaged, from single-parent families, or who have disabilities. Suspension can create more sense of separation between the student and school, increase feelings of not belonging, and negative feelings about school. Unsupervised students are vulnerable to getting in more trouble and consuming alcohol and drugs. Suspensions can cause further stress for children and a sense of isolation.

Suspension can be a way to remove the “problem” without addressing the underlying causes.

References 

[1] Balfanz, R. (2009). Putting Middle Grades Students on the Graduation Path: A Policy and Practice Brief. Everyone Graduates Center and Talent Development, Middle Grades Program. Middle School Association.

[2] Rafa, A. (2018). 50-State Comparison: State Policies on School Discipline. Education Commission of the States. August 28, 2018

[3] Nixon, B. (2012). Stress Has Lasting Effect on Child Development. The Urban Child Development Institute.

[4] Investigating the Association between Home-School Dissonance and Disruptive Classroom Behaviors for Urban Middle School Students. The Journal of Early Adolescence, Vol. 38, pp. 530-553. April 2018. By K. Tyler, J. Burris, and S. Coleman. 

[5] Classroom Disruptions, the Teacher-Student Relationship and Classroom Management from the Perspective of Teachers, Students, and external Observers: A Multi-Method Approach. In Learning Environments Research, Vol. 22, Issue 1, pp. 101-116, April 2019.  By M. Scherzinger and A. Wettstein.

[6] Ritter, G. (2018). Reviewing the progress of school discipline reform. In Peabody Journal of Education, Vol. 93, No. 2, p. 133-138, 2018.


2019 ACT WorkKeys: Preparing Students for Work in a Time of Uncertainty

High school seniors and recent graduates are preparing to enter a workforce that has been disrupted, hopefully only temporarily. In a matter of weeks, the nation has gone from record low unemployment to levels not seen since the Depression. Employers are adapting to a new set of work conditions and the demand for new skills appropriate for remote work is rapidly rising.

The WorkKeys Assessment may now be more valuable than ever in helping employers assess the practical skills and adaptability of potential hires.

WorkKeys Assessments in Alabama

WorkKeys is a standardized test given to 12th graders in Alabama public schools. The assessment is meant to measure skills relevant to many of today’s work environments. This is a test that generates positive results in Alabama with steady progress.

WorkKeys is one of the measures through which a student can be designated as College and Career Ready in the state. They may not have the specialized training needed for particular occupations, but if a student earns a certification at the higher levels, the measure should increase the confidence of employers in hiring staff who are adept at applying useful cognitive skills and knowledge to applied work tasks in contemporary work settings.

The Assessments. The assessments consist of three tests of applied cognitive skills which are relevant, according to ACT’s research, to over 20,000 occupations:

  • The Applied Math test
  • The Graphic Literacy test
  • The Workplace Documents test (applied reading for understanding)

Students are awarded a National Career Readiness Certification if they score a Platinum, Gold, Silver, or Bronze score on the WorkKeys. Platinum is the highest level, followed by Gold, Silver and Bronze.

After modifications to the 2018 assessment created significant fluctuation among the different certificate levels, the results in 2019 showed steady gains in the Platinum and Silver levels, and slight decreases in the percent of students earning Gold and Bronze certificates, as well as the percent of students not receiving any certificate.

In Alabama students earning a Silver certificate or above are considered career ready.

Highlights of the WorkKeys test results for the Class of 2019 include:

  • 66% of Alabama high school graduates were deemed workforce ready as measured by the ACT WorkKeys assessment, an improvement over 64 % in 2018.
  • At 94%, Hartselle had the highest percentage of workforce-ready graduates, as measured by WorkKeys.
  • The Silver Certificate continues to be the level with the highest percentage of students.

WorkKeys is emerging as another source of information for making employment decisions and helping prospective workers learn about strengths and weaknesses and opportunities for growth. Using job profiling data provided by ACT, the Alabama Department of Labor maintains data listing the median WorkKeys scores for high demand occupations requiring an associate’s degree or less.


2019 Assessment Results

The following charts show the percentage of graduates in Alabama who demonstrated workforce readiness on WorkKeys assessments.

Percent Workforce Ready Increased in 2019.

In 2019, 66% of high school graduates taking the assessment were deemed workforce ready. This percent has steadily increased since 2015.

Workforce Ready at the Local Level  

The following charts focus on workforce readiness among graduates at the local level. A county view is appropriate when thinking about the quality of a local workforce.

Of course, local school systems and schools are the places where students develop knowledge and skills related to work readiness.

Listed below are the top ranked systems based on workforce readiness assessed through WorkKeys:

  • Mountain Brook City – 95%
  • Arab City – 94%
  • Satsuma City – 94%
  • Cullman City – 89%
  • Hartselle City – 87%
  • Trussville City – 87%
  • Homewood City – 86%
  • Madison City – 86%
  • Brewton City – 85%
  • Guntersville City – 85%

Mountain Brook has made positive gains over the past two years and appears to remain committed to WorkKeys as a valuable assessment.

With committed effort, WorkKeys is an assessment where systems can show growth, as demonstrated by a variety of systems and schools in Alabama.

A number of large and medium-sized systems made noteworthy gains over 2018. Auburn City topped all, showing remarkable gain (up by 24 percentage points).

Among the schools, Loveless Academy continued its top ranking with 100% of students deemed workforce ready. Top schools include:

  • Loveless Academic Magnet High School, Montgomery County – 100%
  • New Century Technology Magnet High School, Huntsville City – 95%
  • Mountain Brook High school – 95%
  • Brewbaker Technology Magnet High School, Montgomery County – 94%
  • Arab City High School – 94%
  • Ramsay High School, Birmingham City Schools – 94%
  • Satsuma High School, Satsuma City Schools – 94%

A number of large and moderate sized high schools experienced positive gain over 2018. This is an area of growth for a number of schools failing to show gains in more purely academic assessments.

Change in Certificate Levels

Students are deemed workforce ready if they achieve certification at the Platinum, Gold, or Silver levels. The charts show that the percentage of students at the Gold and Silver levels increased moderately from 2015 through 2017, with Platinum barely making a dent. After modifications to one of the tests in 2018, Gold and Platinum level certificates grew substantially, while Silver certificates dropped significantly. The percent at each level in 2019 is comparable to 2018, but with positive gains in Platinum and Silver. The percent of students at the lowest level, with no certificate, continues to drop.

Overall, this resulted in a higher percent workforce ready for the state, with a positive trend toward higher certification levels. 

Change in Certificate Levels at the System Level

Platinum: Mountain Brook and Homewood City finished in a tie with the highest percent of Platinums, followed by Madison City, Hoover, Hartselle and Arab City. Aburn City climbed up in 2019 to join these state leaders.

Gold: Oneonta, Arab, and Mountain Brook are the systems with the highest percentage Gold. Though the majority of systems decreased at the Gold level, a number of systems increased their percentage. Leading the way in percentage point gain was Chickasaw City and Jacksonville City, followed by Piedmont City and Talladega County. Arab City continues to show nice gains as well.

Silver: This is the minimum level to be counted as workready. A number of less affluent systems are among the leading systems for Silver certificates. Systems with the highest percentage at this level include Opp City, Satsuma City, Haleyville City, and Lamar County. The vast majority of systems increased in the percentage of students earning Silver certificates in 2019, led by Opp City.

Subgroup Analysis

Analysis of subgroup results shows a continuing disparity between subgroups. Use the filters to see how systems differ in subgroup performance. Some schools may be better at assisting struggling groups than others.

The highest performing groups include students who are Asian, white, non-poverty, female, and military-affiliated. The lowest-performing groups are black, migrant, and special education. While all of the racial groups are increasing their percentage of workforce ready, black students have made the most gain since 2015.

Females continue to outperform male students, though both have shown comparable positive growth since 2015.

In looking at trends, all racial groups are showing progress from year to year, especially Asian and black students.

Finally, special populations are also showing positive growth in workforce readiness, especially students qualifying for free lunch (poverty) and students living in foster homes.

Discussion

The 2020 report, Education Matters, written by PARCA on behalf of the Business Education Alliance, recommends increasing the awareness of WorkKeys as a valuable tool for employers, communities, and schools.

In addition to the assessments, WorkKeys provides a full suite of resources that can help provide training for teachers, test preparation for students, and design career-related curricula to help students improve their “hard” and “soft” skill levels. ACT Career Ready 101 is designed to help teachers bring work readiness skills into the classroom. This raises questions about how students become workforce ready in school. Beyond special resources for preparation, is the curriculum and content delivered in schools changing to strengthen career readiness, even in traditional academic subjects?

The long-term future of WorkKeys will be determined by the continuing value educators give to these assessments and suite of resources, and the degree to which employers value WorkKeys. A growing number of communities and employers are aware of the utility of WorkKeys, but this still varies across the state. Currently, schools also vary in the degree to which they emphasize WorkKeys.


College Freshmen in Need of Remedial Education

The percentage of first-time college students from Alabama public high schools assigned to remedial courses in math and English has continued to drop overall and, in the most recent data available, dropped dramatically for those attending two-year colleges, according to data from the Alabama Commission on Higher Education (ACHE). Graduates studying in four-year colleges needing remediation increased for the first time in several years. 

The data follows Spring 2018 graduates of Alabama high schools who enrolled in Alabama public colleges in the fall after graduation. The data shows that 26% of those who enrolled in higher education were required to take a course in either remedial math or remedial English or both. That’s down from 32% five years ago.

Remedial courses are primarily offered in English and math. They are designed to bring students up to the educational level needed to succeed in a college course. Students assigned to remedial math and English cannot proceed on to college-level courses in those subjects until they have passed the remedial courses. Colleges vary in the degree to which they allow students to take courses in other subjects simultaneously with their remedial courses. Some students enter with a poor background in one of these subjects but not the other. Some may have a specific learning disability in math, for example, or just never gained the foundation needed for higher math, but excel in English and other courses.

Interactive charts are presented below. Each panel presents a different view with the capacity to drill down from the state level to the system and school level, and for schools within a system. Bear in mind that the ACHE report only captures high school graduates who enrolled in the fall after their graduation in Alabama public colleges. The remediation rates for schools that send significant numbers of students to private colleges or to out-of-state colleges will not necessarily reflect the outcomes for the entire graduating class.

A higher percentage of students take remedial math than remedial English. But the progress shown for math over the past year, down from 24% to 21% in remediation is significant, given that in the past English has shown more improvement. The percentage of students taking remedial English remained the same at 12%.

At this point, both subjects have made similar progress in reducing the percentage of students taking remedial courses. Math dropping from 26% in 2013 to 21% in 2018, and English dropping from 17% to 12%. Check it out in the charts. and use the filters to drill-down and make comparisons.

The next set of charts provide remediation rates for all systems and schools in math and English over time.

Finally, take a look at the numbers in two-year and four-year colleges.

Thirty-three percent of Alabama’s high school graduates enrolling in the state’s public two-year colleges were assigned to remediation in 2018, a significant decrease from 41% in 2017. Since 2013, that percentage has dropped from 46% to 33%.

At the same time, remediation among Alabama’s high school graduates attending four-year colleges increased to 18% in 2018, up from 14% in 2017, and 17% in 2017. This reverses a downward trend and becomes the highest percentage assigned to remediation in the state’s four-year colleges over the past six years.

Over time, approximately 75% of Alabama’s high school graduates assigned to remediation attend the state’s public two-year colleges, and 25% attend the state’s public four-year colleges. This ratio has remained steady for years, but these numbers began to shift in 2018. The ACHE data shows that for 2018, two-year colleges account for 65% of graduates taking remedial courses and four-year colleges up to 35%.

This may be an anomaly in which 2018 is a unique year for remediation patterns across these two sectors. In the future we can learn more about the degree to which this pattern holds up.

In the dashboard panels, you can explore these same data for local school systems and schools. For example, the second panel provides a comprehensive view of two and four-year college numbers for all local school systems in 2008.

This is followed by a chart showing results for individual systems over the longer period, 2013-2018. In the case below the filter is selected for Huntsville City Schools.

The fourth panel provides a view of these numbers for all schools in 2018.

The system name filter can be used to see all the schools in a single system. This is shown for the Birmingham City Schools.

Finally, you can drill down to data over time for an individual school. Use the filters to look up systems and schools of interest to you.

Explore on Your Own

The final set of charts shows the remediation rates in math and English for all systems and schools over time.

Turning the Corner: Is Remediation Good or Bad?

A downward trend in the percentage of graduates taking remedial courses is positive. This could mean Alabama is increasing college readiness among students in its public schools, and more of those students attending college are better prepared.

Still, the downward trend in remediation is happening across the nation, and in part, reflects a movement away from using remedial courses in colleges and universities.

College remediation has come under consistent attack from critics over the last couple of decades. Research findings on their effectiveness are mixed with negative and positive results for students.

Some research has found that students taking remedial courses earn more credits and are more likely to graduate than comparable students who did not address fundamental problems through remediation. More commonly, research has found the opposite. Students who could have made progress in their curriculum were held back, lose the will to continue, and their family bears the cost. Financial assistance programs rarely pay for remedial courses. The mixed findings of these studies have fueled ongoing debate, though more rigorous research is now providing a closer look at what might really be happening.

A well designed, highly rigorous study published in 2017 by researchers at Vanderbilt and Harvard found that students who are on the margin would more than likely be better off pursuing regular college courses. Those placed in remedial courses will likely earn fewer credits over time and are less likely to graduate from college. At the same time, students on the lower end of the spectrum who are taking remedial courses will likely fare better in college than comparable students who do not take remedial courses. This was especially true for students taking remedial English.1

These are students who genuinely need a basic foundation in order to make further academic progress. Past research has documented evidence of students enrolling in four-year universities who cannot compose a complete sentence. It does raise questions about the appropriate educational route for students who lack important basic skills, but who would like to attend college. Where would these students best be served?

In response to criticism of remedial education, colleges have moved toward a broader and more complex set of resources for helping students. In this environment, remedial courses are on the lower end of a continuum of resources and strategies designed to support students experiencing challenges. The downward trend in remediation is likely in part due to remediation increasingly not being seen as the preferred alternative for students on the margin of needing assistance. The wider support now found in colleges include:

  • Comprehensive learning centers, writing centers, and math centers with mentors and tutors
  • Student learning communities
  • Peer tutoring and mentoring
  • First-year freshman seminar
  • Summer bridge program before the fall college term
  • Paired courses where students apply basic skills in one course to the content learned in another
  • Support in time management, organizational skills, resilience, and self-regulation

Unfortunately, programs vary in their effectiveness. Fewer students in remedial courses do not necessarily mean colleges are achieving higher graduation rates. But some programs are proving effective. Learning communities where students share and learn together have experienced strong success in many cases.

Still, the best medicine is a strong foundation during the pre-school through high school years, combined with strong family support.


Alabama ACT Scores for the Class of 2019

Alabama public high school students’ performance on the ACT was down slightly in 2019 for the second straight year, after hitting a high point in 2017. Alabama students’ average score was 18.9 on a 36-point scale, compared to 19.1 in 2018 and 19.2 in 2017. Over the past five years, Alabama composite and subject scores have been stable, varying only in a narrow range.

Alabama public schools give the ACT in the junior year of high school. The final reported results here are for the students who graduated in 2019. Those students would have taken the ACT in 2018 at their own high school. If a student took the test subsequent to the administration at their high school, the student’s highest scores in each subject are counted in our data.

Interactive charts in this report allow you to explore the results at the state level and by system and school, subject, and year. 

The results presented here are derived from data requested from the Alabama State Department of Education. It differs from what ACT publically reports for the state. ACT’s reports include private school students. PARCA specifically requested the scores of Class of 2019 graduates. The results show a slight but unexpected variance: a decrease in the statewide average composite score but increases in subject tests that make up that composite score. Though counter-intuitive, the variance is minor and can occur. There are more subject test scores than composite scores, which are only awarded when students complete all subject tests. Rounding can also play a part, and the scores in 2018 and 2019 are very close.

How Does Alabama Compare?

The ACT is a test of college readiness and is used by colleges as a factor in the evaluation of applicants’ qualifications for admission. Alabama is one of 17 states that give the ACT to all public high school students, whether they plan to apply for college or not. Among the states that give the test to all public high school students, Alabama ranks 14th. The average scale score among the 100 percent states is 19.5.  Source: The Condition of College and Career Readiness, 2019. National ACT.

ACT Scores and College Admission

A composite score of 18 is considered a minimum threshold score for college admission at several state colleges.  Others have lower thresholds, including open admissions.  On the other end, a score of 25 or higher is expected at more competitive colleges, while 30 is the minimum threshold at some of the nationally elite schools. The table below lists the median entering ACT composite score of students admitted at Alabama colleges. Historically black colleges consider it part of their mission to admit students who may not have had the academic preparation to perform well on the ACT. Source: College Scorecard, US Department of Education.

There is a difference between making a score on the ACT that will help students gain admission to college versus a score that indicates a student is academically ready to successfully perform in college. institutions with lower admission standards generally have a lower percentage of students successfully performing in college courses and completing their studies with a degree.

ACT has established benchmark scores for the individual subject tests that determine whether, in their analysis, a student is ready for college material. Here, college-ready means the student has a 50% or greater likelihood of earning a B in a first-year college course in that subject. The benchmark score varies by subject:

  • English College Ready Benchmark = 18
  • Mathematics College Ready Benchmark = 22
  • Reading College Ready Benchmark = 22
  • Science College Ready Benchmark = 23

Source: What Are the ACT College Readiness Benchmarks?

Results showing the percent college-ready are shown below.

PARCA’s analysis shows that half of Alabama’s graduates in 2019 scored college-ready in English, followed by reading at a much lower percent. The lowest percent of college-ready is perennially in math. This level of college-readiness has been stable in Alabama over time. The ranking of subjects from English as the highest percent to math as the lowest is seen in most other states, though the gap between English and math in Alabama is quite large.


School Systems and Schools

The top ten scoring school systems in 2019 includes:

  • Mountain Brook City: 27.4 average ACT
  • Vestavia Hills City: 25.3
  • Homewood City: 24.2
  • Madison City: 23.6
  • Hoover City: 23.0
  • Trussville City: 22.7
  • Arab City: 22.4
  • Auburn City: 22.4
  • Hartselle City: 22.2
  • Cullman City: 21.9

The same systems made up the top ten in 2018, though Arab City was below Auburn and Hartselle, and Cullman was ranked higher than in 2019. Back in 2017 Arab City was not in the top ten, though close. Homewood and Madison City have been battling it out for the third position over the past five years, though Homewood has firmed up that ranking over the past two years – experiencing significant growth in its average scores.

Systems and Schools Showing Improvement: 2015-2019

In 2019, 56 school systems in Alabama posted higher average composite scores on the ACT than they did in 2015. The ten systems achieving the largest increase in students’ average composite score over that period are listed below. A more complete listing can be found in the interactive charts.

RankSystem2015 Average Composite Score2019 Average Composite Score
1Homewood City22.8024.15
2Chickasaw City15.5016.84
3Fayette County17.9819.12
4Thomasville City17.3318.46
5Auburn City21.3222.41
6Macon County15.2516.24
7Trussville City21.7322.70
8Linden City15.4216.37
9Coosa County16.3617.29
10Arab City21.5922.41

Individual schools also showed noteworthy increases. Northridge High School in Tuscaloosa City jumped from an average composite score of 19.9 to 22.2, the highest gain in the state. They were followed by Hubbertville School in Fayette County, increasing from 17.03 to 19.10, and Sardis High School in Etowah County, which increased from 17.67 to 19.53.

Keeping Demographics in Mind

On average, students from economically disadvantaged families tend to score lower on standardized tests than students who are not at an economic disadvantage. Similarly, schools where a higher percentage of the student body is economically disadvantaged, the average test score tends to be lower. Scatterplot charts, like the one featured below, present test score data and student poverty levels at the same time. The vertical position of the school or system is determined by the average test score, while the horizontal location of the school or system is determined by the percentage of the student body directly eligible for a free lunch under the national free lunch program.

Systems and schools with substantial poverty above the line are exceeding expectations given their level of poverty, while those below the line are performing less than expected. Systems and schools close to the line are performing as expected.

When looking across subgroups Asians are the highest performing group. Higher performers are Asian or white, female, and non-poverty.

Among racial groups, Asians have the highest percentage of students scoring college-ready in all subjects, including English, followed by white students – though the gap between Asian and white students in math and science is large. The role of parent’s education level, family income, and coming from a college-going family and school culture plays a role across the results shown below.

The final chart in the demographic story looks at the percent college ready over time for subgroups. This interactive chart can be filtered for different subjects, school systems, and schools.


2018 College-Going Rates for Alabama High Schools

PARCA is sensitive to the challenges of the Coronavirus pandemic. We are responding with reports we hope are helpful. It is also important to continue moving forward. History tells us that disease has always been a part of our lives and that time is on our side. We will eventually win this war.

This brief continues PARCA’s tradition of reporting college-going rates for Alabama and its local systems and schools.

The percentage of high school graduates in Alabama enrolling in college after graduating in 2018 remained the same as the graduating class of 2017, at 62%. The number and percentage attending two-year colleges slightly increased. The number and percentage of recent graduates entering four-year colleges both slightly decreased.

The data, drawn by ACHE from the National Student Clearinghouse, follows Alabama public high school students who graduated in the spring of 2018 and enrolled in higher education in the fall or spring of 2019. The data includes records for in-state and out-of-state institutions, both public and private.

Over the past five years, the college-going rates for Alabama’s high school graduates have declined slightly. In 2014, the first year this set of statistics was produced, 65% of high school graduates enrolled in college the year after their graduation. In both 2017 and 2018, 62% of graduates enrolled.

At the same though, the size of the senior classes has been larger and graduation rates have been higher. That has produced more high school graduates going into college.

While 2018’s 62% college-going rate is tied for the lowest rate over this five year period, the actual number of graduates enrolling in college increased in 2018 compared to 2017. Only in 2016 did more students attend college, 31,414 in 2016, compared to 31, 337 students in 2018.

At the same time though, the larger classes of seniors and higher graduation rates have resulted in greater numbers of students graduating with a high school diploma but not immediately continuing their education. Among graduates of the Class of 2018, 19,191 did not enroll in higher education after graduating high school.

After a dip in 2017, the number of students going to community colleges increased while fewer attended four-year institutions.

Alabama’s public high school graduating class of 2018 totaled 50, 528, which is 764 students more than the Class of 2017. Among those graduates, 16,085 (32%), enrolled in a two-year community college, according to the data. That’s 951 students more than the previous year and the highest number over the past five years.

Enrollment of graduates in four-year colleges decreased from 15,804 in 2017 to 15,252 in 2018, down 552 students (30% of graduates). The total number going to public 4-year colleges in Alabama dipped by 122. The overwhelming majority, 93% of those who enrolled, went to Alabama institutions, and 92% to a public college or university.

The remaining 19,191 high school graduates, 38% of graduates, were not found to have enrolled in higher education. That’s an increase of 365 in the number of students graduating from high school but not enrolling in higher education the following year.

Local Systems and Schools

It is helpful to explore results for individual schools and school systems, which sometimes tells you more about what’s happening underneath the statewide aggregated view.

Additional charts focus on the percentage of students attending four-year colleges by system and school, and the percentage attending two-year colleges at the system and school level, as well as change over time.

The top five systems sending students to four-year colleges includes:

  • Mountain Brook City: 86%
  • Vestavia Hills: 79%
  • Homewood City: 71%
  • Hoover City: 64%
  • Trussville City: 59%

The top five systems sending students to two-year colleges includes:

  • Lamar County: 67%
  • Boaz City: 69%
  • Roanoke City: 60%
  • Marion County: 57%
  • Winfield City and Winston County: 55%

Change in the percentage of students enrolling in higher education from the graduating class of 2014 to the class of 2018 highly varied across systems and schools, ranging from 32 percent growth for Sheffield City to a reduction of 25 percentage points for Tarrant City. The systems with the highest growth include:

  • Sheffield City: 32% growth
  • Anniston City: 16% growth
  • Clay County: 14% growth
  • Conecuh County: 13% growth
  • Roanoke City: 11% growth

Is Your School Adding Value?

Socio-economic conditions are always influential in shaping college-going. A scatterplot is included which presents the college-going rate of Alabama high schools in the context of the poverty percentage at each school. The higher on the chart a school appears, the higher the percentage of its graduates enrolled in college. The farther to the right a school is positioned, the lower the percentage of students in poverty at that school. The poverty measure used is the percentage of students who directly qualified for a free lunch under the National School Lunch Program.

Note that those schools and systems above the line in the scatterplot are exceeding expectations for college-going given their level of poverty, and those below the line are below expectations. Those right on the line or close to it are performing as expected.

Explore the tabs to see a variety of data visualizations, including maps, charts, and tables at the state and local level. Below is an example of a map bringing these issues to light with a statewide geographical view.



Help Wanted: Alabama’s Labor Force Participation Problem

Alabama has record low unemployment, one of the lowest unemployment rates in the country.

And yet, the percentage of Alabamians actually working is lower than almost any other state.

This seeming paradox explains, in part, why Alabama has lower per capita income than most other states and higher rates of poverty, despite those historically low unemployment numbers. To be counted as unemployed, you have to be looking for work, or in other words, participating in the labor force.

In 2018, Alabama’s labor force participation rate was lower than all other states, except West Virginia (we’re tied with Mississippi). As measured by the American Community Survey, 71% of Alabamians, ages 25-64, were working or looking for work. That compares to the U.S. labor force participation rate of 78% for that age group.

If Alabama’s labor force participation rate equaled that national average, Alabama would have almost 160,000 additional workers in the workforce in that age range.

Low labor force participation is a significant problem for Alabama, one that has been recognized by Gov. Kay Ivey’s Administration and the Alabama Workforce Council in its Success Plus Initiative (for more information on this initiative see Education Matters, a report from the Business Education Alliance authored by PARCA. You can also read the Alabama Workforce Council’s latest update on the effort).

Increasing the state’s labor force participation rate is one of the key objectives identified by Success Plus for helping Alabama meet its goal of adding 500,000 highly skilled workers to its workforce by 2025. The state will not reach that goal without bringing significant numbers of those who have dropped out of the workforce back in, which in most cases means equipping them with supplemental education and skills needed to thrive in today’s work environment.

Who is missing?

Before we can bring people back into the workforce, we have to understand who’s missing in the first place, and which groups of Alabamians have dropped out at a higher rate than residents of other states.

Using data from the U.S. Census Bureau’s American Community Survey (ACS) for 2018, PARCA examined the labor force participation in Alabama in comparison to other states and to the nation as a whole. We used five-year estimates because that survey provides estimates for smaller counties not present in the one-year estimates.  

A comparison of one-year estimates yields similar results. However, it should be noted that in the 2018 one-year estimates, labor force participation in Alabama did show an increase over 2017, from 70.5% to 71.4%, after a multi-year downward drift. Whether that is a result or the economy or policy focus, it would be hard to determine, but, hopefully, it is part of a trend. For comparison, before the Great Recession, Alabama’s labor force participation rate was 75% for that age group and the national rate was 80%.

Sharpest differences with the rest of the U.S.: Age, Income, and Disability

Alabama trails the rest of the U.S. labor force participation rate across the board. When you break the population down into subgroups by age, sex, race, economic condition, or educational attainment, in almost every subgroup, Alabama’s rate trails the U.S. rate. However, in some subgroups, the divergence from the national rate is particularly sharp.

In comparison to the same population groups in other states, older Alabamians and Alabamians with lower levels of education are less likely to participate in the labor force, according to Census figures. Alabamians with a disability are also less likely to participate in the workforce than their counterparts in other states. The labor force participation rate for those groups is roughly 10% lower in Alabama than it is for similar individuals across the U.S.

The gap with the U.S is most extreme when it comes to individuals with less than a high school education. In Alabama, only 49% of individuals who lack a high school diploma are working or looking for work. That compares to a labor force participation rate of 61% nationally for those with less than a high school education.

Older workers in Alabama are also disproportionately out of the workforce. The labor force participation rate of those 60 to 64 is 10 percentage points lower in Alabama than it is in the rest of the country: 46% of Alabamians ages 60-64 participate in the labor force compared to 56% nationally. For those between 55-59, Alabama’s labor force participation rate is 63% compared to 72% nationally.

Age remains a factor in those between the ages of 45 and 54; Alabama’s rate is almost 7 percentage points lower than the U.S. average. For those 44 and younger, the gap between the Alabama workforce participation rate is narrower, though Alabama’s rate is still 3 to 4 percentage points lower than the U.S. rate.

You find a similar pattern with educational credentials: the more education an Alabamian has, the more likely he or she is to participate in the workforce. As the education level rises, the gap with the U.S. rate closes: 84% of Alabamians with a bachelor’s degree or higher participate in the workforce compared to 86% nationally.

When it comes to individuals who report having a disability, Alabama also has lower rates of labor force participation than other states, with the labor force participation rate among disabled individuals in Alabama trailing the national average by 9%.

You can explore state and national rates of workforce participation in the map below. Use the selector at the top to choose the population group to measure.

The impact of Alabama’s labor force participation deficit in these subgroups is amplified because Alabama’s population is older, less educated, and more disabled than the country as a whole. Alabama’s median age is 38.7 compared to 37.8 in the US; 25% of Alabamians have a 4-year degree compared to 32% across the country; 11.6% of Alabama’s population reported a disability compared to 8.6% nationwide.

What does not impact our labor force participation rate

Alabama’s racial diversity does not affect the state’s labor force participation rate, though if Alabama was more diverse, our labor force participation rate might be higher.

Non-Hispanic whites (ages 16 and up) have the lowest labor force participation rate of the major racial and ethnic groups, just under 57%, lagging non-Hispanic whites nationally by 5.4 percentage points. Blacks in Alabama have a labor force participation rate of just over 57%, lagging the U.S. rate by 5.2 percentage points. Alabama Hispanics have a labor force participation rate of 67%, which is less than a percentage point difference from the U.S. labor force participation rate for Hispanics.

It is sometimes assumed that Alabama’s supposed preference for traditional gender roles might depress the labor force participation rate for females in Alabama compared to the rest of the U.S. However, if those attitudes produce any effect, it is slight. The female labor participation rate in Alabama is 66.4%, trailing the national average by 6.2 percentage points. That’s only slightly higher than the gap for males. The male labor force participation rate in Alabama, 76.6%, trails the U.S. rate by 5.6 percentage points. In fact, one subset of the female population in Alabama participates in the workforce at a higher rate than the U.S. rate. Alabama females with children under 6 are more likely to be in the workforce than similarly situated women nationwide. According to the Census figures, 71.4% of those mothers participate in the labor force.

Geography matters

Alabama’s rural counties have the lowest labor force participation rates. Meanwhile, the most urbanized and affluent of Alabama counties have the highest rates of labor force participation.

One might conclude that it would be best to recruit workers back into the workforce by targeting rural counties where labor force participation is lowest. However, this neglects the fact that the largest population of workers who aren’t participating in the workforce are in metropolitan counties. That’s also where unemployment is lowest and jobs are clustered. Both rural and metropolitan counties need strategies for increasing labor force participation.

The visualization below allows you to calculate how many additional workers could be added to the workforce in the selected geography if that geography matched the U.S. labor force participation rate for the selected subgroup. The two subgroups selectors on the dashboard, one for the subgroup population and one for the subgroup labor force participation rate must be set to the same subgroup to produce accurate calculations.


2019 Test Results for State Schools Show Little Change in Student Proficiency Levels

Alabama is scheduled to implement its new state education tests this spring. In the meantime, statewide results from the Scantron assessment in 2019 have been released by the Alabama State Department of Education. They show little change from 2018. In both years, less than half of Alabama public school students in grades three through eight scored proficient in reading and math.

For the state as a whole, 47% of students were proficient in math in 2019 and 2018, and 46% were proficient both years in reading.

The overall proficiency rate for science in 2019 was 37%, a very slight decrease from 38% in 2018.

This comes at a time when Alabama’s performance on the National Assessment of Educational Progress (NAEP) has dropped.

New State Assessments

2019 is the second year of data from Scantron, which in 2018 replaced ACT Aspire as the primary state assessment used for measuring academic progress in Alabama public schools since 2013-2014. In June 2017, the State Board of Education voted to cancel Alabama’s contract with ACT and move toward developing another set of tests, which are scheduled to be launched in the spring of this school year, 2019-20. In the meantime, the state continued to assess performance in grades three through eight using tests provided by the Scantron testing company.

Scantron Statewide Results for 2019

Comparing Subjects. Selecting the first tab above shows that among the three subjects assessed with Scantron, once again in 2019, math generated the highest percentage of students scoring at or above proficiency (47%) across grades 3-8 combined. Reading was not far behind at 46% proficient. Similarly, from 2014 through 2017 math generated the highest percent proficient among students taking ACT Aspire, though the gap between math and reading was larger.

Under Aspire, math made ongoing progress each year climbing from 40% proficient in 2014 to 48% in 2017. The increase from 40% with Aspire to 47% with Scantron seems about normal.

In contrast, reading proficiency under Aspire showed little change each year rising from 39% in 2014 and to 40% in 2017. The jump for reading from 40% with Aspire to 46% with Scantron was more significant than the change between tests in math.

In 2019, the Scantron science assessment continued to generate the lowest proficiency rate among the three subjects tested, at 37% proficient, as it did in 2018 with 38% proficient.

Comparing Grade Levels. The second and third panels above provide grade level comparisons. Differences in math proficiency by grade for 2019 and 2018 are minimal. Across all grades under Scantron and Aspire, math proficiency levels are at their highest in the third grade, following the trend in which reported proficiency levels drop in the higher grades. This pattern is less substantial when looking at a particular cohort of students moving from grade-to-grade but still applies. With both Scantron and Aspire, math proficiency drops sharply in the seventh grade, though less dramatically under Scantron in both 2019 and 2018.

Third-grade math is the one grade and subject each year in which the majority of students meet or exceed the benchmark for proficiency. In 2019, 58% were proficient in third-grade math, up slightly over 57% in 2018. This level of proficiency is very similar to Aspire’s third-grade math results, which grew from 52% in 2014 to 59% in 2017. After third grade, though, math proficiency drops, winding up at 45 percent of students demonstrating proficiency by 8th grade.

Reading proficiency in 2019, using Scantron, shows a steady drop from 48% proficient in the third grade to 43% in the 8th grade. This reflects very little change from 2018, in which proficiency ranged from 48% in the third grade to 44% in the 8th grade. Though proficiency levels get lower in the higher grades, they are not dramatically lower as found in math. The more dramatic grade-by-grade drop-in math proficiency was evident in Aspire as well, though even more pronounced.

Why are proficiency levels dropping from grade to grade? First, a decrease at the higher grade levels is normal. During the early grades, math and reading are focused on helping young students learn basic skills, but in the higher grades, math becomes more complex and introduces algebra. Reading instruction also becomes more demanding as students move from learning to read to reading for understanding. Attitudes about external state assessments among younger students in the earlier grades could also be different from students in the higher grades.

But why the difference between reading and math? The less dramatic reduction in proficiency levels in reading from grade to grade might suggest that either students are more effective in learning the basics in reading as a foundation for later grades or that the material in math becomes harder to master and teach. Like many places, the demand for math teachers in Alabama far exceeds the supply of available, qualified math teachers. Proficiency scores on the state Aspire math test for tenth graders significantly dropped from earlier grades to as low as 18% proficient in 2016-17, compared to 33% in reading. Tenth-grade proficiency is not assessed under Scantron. Furthermore, high school ACT scores in Alabama for college admission are higher in reading than in math.

Science is a little different. The drop in proficiency among students in the higher grades takes on a different pattern in science. In 2015, Alabama only tested science in the tenth grade, and 21% tested as proficient or above. In 2015 and 2016, grades three through eight and grade ten were tested. Proficiency levels actually increased from grade-to-grade in grades three through six, with the peak level of proficiency in the 6th grade, then began dropping in the 7th grade, reaching the lowest level in the tenth grade. In Aspire’s last year (2016-17), science was tested in grades 5, 7 and 10. Proficiency levels dropped in each higher grade, as would be expected from past performance in these grades. Finally, under Scantron science was tested in grades 5 and 7, dropping from 40% (grade 5) to 35% (grade 7) in 2018, and from 39% to 35% in 2019.

The low scores in science are worthy of concern, for Alabama and the nation. Though reading and math are more fundamentally essential, science relates to problem-solving skills, reasoning, curiosity, critical thinking, good measurement skills, and applied learning. Through science, a student can learn more about the world around them and better prepare for careers in technology, engineering, agriculture, advanced manufacturing, and health services. These are important skills for the workforce of tomorrow in Alabama.

Focus on Third Grade Reading and the Alabama Literacy Act

Early reading is a pivotal predictor of academic success. Children who learn to read in the early years have a foundation that will help them in other subjects. Early brain development and schooling through grade three are extremely important in shaping a person’s education and life chances. The figure above shows the percentage of students in each of the four proficiency categories for third grade reading from 2014 through 2019. Scantron labels used in 2018 and 2019 for the different levels of proficiency are listed below:

Level 1 (Red—Emerging Learner)

Level 2 (Yellow—Developing Learner)

Level 3 (Light Green—Proficient Learner)

Level 4 (Dark Green—Distinguished Learner)

Students who achieve proficiency or above are in Levels 4 and 3 (dark green and light green). Similar to a traffic light, these colors signal “go.” This is followed by yellow (“use caution”) and red (“stop”).

The Figure above shows that 21% of third-graders read at the lowest level in 2019. Under Aspire, in 2017 that percentage was 39%. It will be interesting to see how this comes out in the new assessment.

To address the high percentage of students not reading with proficiency, in 2019 the Alabama Legislature adopted the Alabama Literacy Act.

The Literacy Act refocuses attention on early reading in kindergarten through third grade, with the expectation that all students should be able to read by the end of the third grade. Beginning in the 2021–22 school year, students falling into the lowest group in reading may be at risk of being retained. Though minority students will be disproportionately affected, research shows that students who are held back and learn to read with intensive reading intervention do better in school than comparable students who are not held back.[1] A key issue relates to the capacity of the state and local systems to provide the intensive assistance as students are held back.

The State Superintendent of Education is convening a standing task force to provide recommendations for comprehensive core reading and reading intervention programs, teacher professional development in the “Science of Reading,” and valid and reliable assessments that can be used for screening, diagnostic, and instructional purposes. Research has identified how skilled reading works, and helping teachers learn the science behind reading can make a difference, as demonstrated in Mississippi.


Comparing Systems and Schools – How Did Your School Do?

The tabs above list the percent proficient in each subject for all local systems and schools. Click on any one of those tabs and look up schools or local systems most important to you and see how they compare to other systems.

Across the state, in both math and reading, 42% of school systems met or exceeded the state average for percent proficient in math and in reading. Fifty-two percent achieved this in science. Clustering occurs where systems and schools perform similarly in all three subjects, especially among the wealthiest and poorest systems. Performance among the top ten and lowest ten in all three subjects is consistent. Still, throughout the state, you can also find variation that shows different local strengths. For example, though Haleyville is ranked 109 in Science (28% proficient), and 103 in reading (36%), it is 53 in math (49% proficient).

It would be useful to learn what is causing differences of this nature and why some systems and schools are stronger in particular subjects than others.

Local Change in Proficiency Over Time

In addition to absolute levels of proficiency in a school system or school, the change occurring over time is another important indicator of performance. This can be an indicator of value added by the school. In schools where students come from family backgrounds in which parents have high income and educational attainment, proficiency levels may already be high, with more limited prospects for change in the school’s proficiency level. In schools where a high percentage come from a background of poverty and low parental educational attainment, students are more likely to enter school less prepared and fall short of their actual potential on assessments. Schools and systems making an effective, concerted effort to improve performance can make a difference. Though absolute scores may still be low, positive growth reflects deliberate improvement, sometimes under very challenging conditions.

The dashboards above list and rank the school systems in the state related to percentage point change in the proficiency rate from 2017-18 to 2018-19 in each subject. The bottom section of each dashboard shows the change in individual schools within each system. PARCA’s analysis below highlights the systems with the highest change in each subject over this period.

Table 1: Percentage Point Change in Proficiency, 2018 to 2019

MathReadingScience
Andalusia City: 11%Andalusia City: 9% Piedmont City: 17%
Perry County: 10%Perry County: 8% Coosa County: 11%
Sumter County: 10%Phenix City: 4%Marengo County: 10%
Leeds City: 6%Lanett City: 3%Tallassee City: 9%
Tallapoosa County: 6%Sumter County: 3%Perry County: 8%
Daleville City: 6%Scottsboro City: 3%*Andalusia City: 7%
Elba City: 7%

*Note that the Macon, Piedmont, Dallas, Butler, Walker, Brewton and Elba also improved reading proficiency by 3 percentage points.

Andalusia City and Perry County were among the most improving systems in all three subjects, and Piemont’s 17 percentage point increase in science is outstanding. Across the state, proficiency levels range in math from +11 to -8%, in reading from +10 to -7%, and in science from +17 to -13%.

Table 2 shows the change in proficiency over 2014-2019, which is complicated by comparing Aspire and Scantron results.

Table 2: Percentage Point Change in Proficiency, 2014 to 2019

Change in Math ProficiencyChange in Reading ProficiencyChange in Science Proficiency
Geneva County 26%Jacksonville City 22%Piedmont City 36%
Saraland City 22%Scottsboro City 21%Saraland City 35%
Dale County 22%Trussville City 21%Etowah City 32%
Marengo County 22%Geneva County 18%Satsuma City 30%
Trussville City 21% Saraland City17% Leeds City 29%
Russell County 20% Dale County 15% Geneva County 28%
Lamar County 20% Brewton City 14% Russell County 27%
Houston County 19% Piedmont City 13% Dale County 26%
Troy City 19% Talladega County 13% Opp City 26%
Clarke County: 19% Perry County 13% Geneva City 26%
Haleyville City: 19%Tuscumbia City 13%Cleburne County 26%
Henry County 26%

Over this longer period, a different set of systems are shown. Perry, Leeds, and Marengo systems again are listed, and Piedmont continues to come out on top with improved science proficiency. Systems with significant improvement in two or more of the subjects include Geneva County, Saraland City, Perry County, Dale County, and Russell County. Across the state, change ranged from +26 to -8% in math, +22 to -5% in reading, and +36 to -5% in science.

Performance of Subgroups in Alabama

In Alabama and across the country, differences in proficiency rates among various subgroups of students continue to be a concern, and the gap between students who are white and students of color is increasing. Proficiency rates for all subgroups changed very little from 2018 to 2019. More positive growth occurred between 2014 to 2019, though comparisons are between Scantron and Aspire assessments.

  • Students who are African-American, English Learners, and those in special education all perform at a lower level than do economically disadvantaged students as a group.
  • A higher percentage of Hispanic students, except those who are English Learners, attain proficiency in math and science than do students who are Black. Those two groups have the same percent proficient in reading. Among all the sub-groups, English Learners appear to struggle the most. They are ranked low in all subjects, especially science and reading, where they fall below all racial groups, economically disadvantaged students, and students in special education.
  • Asian students are the highest performing in all three subjects. Wide gaps continue to exist between African-American students and both Asian and White students, and between Hispanic students and both Asian and White students. Comparing proficiency results in 2014 and 2019, the gap between students who are White and those who are Hispanic or Black is increasing. At the same time, Black students have made significant progress. Math has been a stronger subject for Hispanic students than Black students, though in reading the initial gap between these two groups has been closed.
  • Same as last year, female students in the state performed higher than males in math and significantly higher in reading, while male students scored slightly higher in science.

Impact of Poverty

Students growing up economically disadvantaged are less likely to be read to in the early years, are exposed to fewer words, and are more likely to be exposed to health problems that can affect their capacity to learn in school and perform on tests. The education level and income of a student’s parents becomes a significant predictor of performance on standardized tests such as Scantron.

But some schools are better equipped to help all students learn and exceed expectations.

The scatterplot charts found in this section show the general correlation between proficiency levels and poverty levels. A school system’s proficiency rate determines its vertical position on the chart: the higher on the chart, the higher the proficiency rate. A system’s percent of economically disadvantaged students, measured by the percentage of students qualifying for free meals under the National School Lunch Program, determines the system’s position on the horizontal axis. Systems with a higher percentage of economically disadvantaged students will appear to the left of the chart and those with a lower percentage will be to the right of the chart.

Exceeding Expectations or Falling Short. The line displayed in the scatterplot is the average proficiency level for a given level of poverty. Those systems and schools above the line are exceeding expectations given their percent of economically disadvantaged students, and those below the line are falling short of expectations. These charts show that systems with similar poverty levels often show very different proficiency levels. In other words, school systems can and do exceed expectations through effective teaching, student support, and school organization and culture.

As an example, consider the chart on Scantron math performance and percent poverty. Saraland is exceeding expectations with 75% of their students scoring math proficient while 37% are economically disadvantaged. In contrast, in Pike Road only 16% of students are economically disadvantaged but only 47% are considered proficient in math. You can learn about the scores of each of these systems by clicking on the circles in the chart.

Conclusion

The Scantron results from 2019 showed very little change from 2018. The gaps that exist between school systems and among student subgroups continue to be an area where more work is needed, especially if Alabama is to fulfill its vision as a state characterized by a vibrant, innovative and relevant workforce.

This is an exciting year in Alabama as the state rolls out its next suite of assessments at the same time that new math standards have been established. Because of the Literacy Act, the state is refocusing attention on early reading. It will be important to establish a baseline and provide feedback on the validity, reliability, and usefulness of the new assessments. Providing accurate, timely, and accessible data can give the state and its schools clarity on progress being made and a strategic sense of how target resources that can make a difference.

This comes at a critical time when NAEP scores in the state have dropped, when the state’s NAEP proficiency rates continue to be very low, and when the state’s ranking in the nation on math and reading has dropped to the bottom of the barrel.

  • Alabama’s 2019 proficiency level in fourth and eighth-grade math was dead last, 52 out of 52 (the 50 states plus the District of Columbia and the Department of Defense schools). In 2017, the state ranked No. 48 in fourth-grade math and 46 in eighth-grade math.
  • In fourth grade reading the state’s proficiency level was ranked 49 out of 52, dropping from 37 in 2017. In eighth grade reading the state’s ranking dropped from 43 to 49.

These national comparisons provide perspective. Alabama’s schools and teachers in high poverty communities face serious challenges, but hope can be found in the progress made by neighboring states and the steps being taken in Alabama to improve literacy and mathematics instruction.

Reference

[1] West, M (2012). Is Retaining Students in the Early Grades Self-Defeating? See https://www.brookings.edu/research/is-retaining-students-in-the-early-grades-self-defeating/