Research

Owings, W. A., & Kaplan, L. S. (2019). American Public School Finance (3rd ed.). Taylor & Francis. https://bookshelf.vitalsource.com/books/9781351013772

Page 269

Chapter 11: Spending and Student Achievement

FOCUS QUESTIONS

1 Trace the recent history linking school funding to student achievement.

2 Clarify the research findings on the effects of spending on student achievement.

3 Explain why the issue is controversial.

4 Describe the link between teacher quality (effectiveness) and student achievement.

5 Summarize the connection between professional development and student achievement.

6 Describe the findings on the relationships between class size, school size, and student achievement.

7 Explain the relationship between teachers’ salaries and student achievement.

8 Describe the relationship between school facilities and student achievement.

Where we spend our education dollars affects student learning. Substantial research supports the link between

school spending and student achievement through teacher/teaching quality, professional development,

reduced class and school size, teacher salaries, and school facilities. Knowledgeable education leaders can

help their communities wisely invest taxpayer dollars in areas that pay dividends in student achievement.

Likewise, informed school leaders can engage their publics about education dollars’ equity implications: the

lowest-income students generally receive the least amount of funding, the least experienced teachers, and the

poorest school building conditions—all undermining children’s ability to learn apace with more affluent peers.

Education finance professors Kern Alexander and Richard Salmon believe questions about school

funding’s impact on public education lie at the heart of national education reform.1 The Coleman Report (1966)

and A Nation at Risk (1983) prompted some politicians and policy makers to seriously question the public

schools’ value if costly educational interventions could not overcome the destiny of children’s backgrounds.

Many continue this reasoning today. As a result, the issue of public school spending is controversial and highly

politicized.

Does public education make a difference in student achievement apart from family influences? Where

should educators spend money in schools to get the biggest “bang for the instructional buck”? The

Professional Standards for Educational Leaders (PSEL) (2015) states that “Effective educational leaders

manage school operations and resources to promote each student’s academic success and well-being.”2 This

includes seeking, acquiring, and managing the fiscal and other resources needed to support effective teaching

and student learning, building professional capacity and community, and acting as “responsible, ethical, and

acceptable stewards” of the school’s—and taxpayers’—money. Targeting available funding to maximize

student achievement is a key school leadership responsibility. Knowing how the school funding issue became

central to school reform policy will help school leaders and their constituents make sense of the current debate.

RECENT HISTORY LINKING FUNDING AND ACHIEVEMENT

Recent questions about the links between school funding and student achievement have been swirling since

the mid-20th century. As part of the 1964 Civil Rights Act, the U.S. Department of Education invited sociologist

James Coleman to investigate racial segregation and education inequality. Surveying 600,000 students in

3,000 schools across the country, Coleman and his colleagues examined the adequacy of physical facilities,

curriculum, teacher characteristics, and student achievement as measured by standardized test scores. They

factored in student self-attitudes and academic goals, socioeconomic status, and parent education levels. His

1966 Coleman Report, Equality of Educational Opportunity, identified shocking achievement gaps across races

and regions. It also determined that variations in school resources—as measured by per pupil spending and

student–teacher ratios—were unrelated to variations in student achievement on standardized tests. The report

concluded, “Schools bring little influence to bear on a child’s achievement that is independent of his [sic]

background and general social context.”3 In its view, student learning depended mostly on the child’s family

background and to a lesser degree, on peers; schools played a minor role.

Coleman’s study seemed to say that the educational inputs (such as student–teacher ratios, funding

resources, teaching practices, curriculum, quality of school facilities) did not contribute much to student

achievement (outputs). Parental education and affluence had more influence on students’ learning in school

than anything the schools or teachers did in the classroom. In short, when it comes to spending money to

improve public schools, “Money isn’t pixie dust.”4

For the most part, educators reacted to the Coleman Report with cautious silence, but economist Eric

Hanushek and education writer and political activist William Bennett spoke up. Using similar input–output

methodology (production function studies) as Coleman, Hanushek (1981, 1986, 1989, 1996) reached similar

conclusions. His meta-analyses of existing studies found that bigger school budgets were not systematically

related to higher student achievement; their relationship was neither strong nor consistent. Nor did the studies

address the “dramatic differences” in teachers’ performance. With eye-popping titles such as

Page 270

Throwing Money at Schools and The Quest for Equalized Mediocrity: School Finance Reform without

Consideration of School Performance, Hanushek incited public and policy opinion against increased education

funding.5

Likewise, William Bennett, U.S. Secretary of Education (1985–88), argued that public schools did not

need more taxpayer dollars to increase student achievement because the states with the highest SAT

scores—Iowa, North and South Dakota, Utah, and Minnesota—had low per pupil spending.6 Bennett

neglected to mention that only the few top academic students in these states took the SAT as part of their

application to highly competitive elite East Coast universities; most college-going peers in these states took the

ACT. In other words, these SAT students did not accurately represent the normal population distribution, and

drawing such conclusions was invalid.

By contrast, other studies indicated positive connections between school spending and student

outcomes. Examining individuals’ later earnings in the labor force, economists found a significant association

between adult earnings and school spending.7 In the 1990s, Larry V. Hedges, a national leader in using

educational statistics for evaluation, and colleagues Richard Laine and Rob Greenwald (1994) reanalyzed

Hanushek’s data using more sophisticated methods and found overwhelming evidence that school funding had

a “positive” impact on student achievement.8 Their later meta-analysis (1996) of 60 studies showed a broad

range of resources were positively related to student outcomes with effect sizes large enough to suggest

moderate increases in spending may be associated with significant gains in student achievement.9 Likewise,

education finance professors Deborah Verstegen and Richard King (1998) affirmed that these positive findings

have been strong and consistent over time.10 Until the 21st century, however, Coleman’s conclusions greatly

overshadowed studies linking school spending and student achievement, allowing them little political air.

Newer studies also affirm the link between carefully targeted school funding and student achievements.

A 2003 four-state research report by the Southwest Educational Development Laboratory found “a strong

relationship between resources and student success.”11 Several investigations on spending and early

childhood programs for economically disadvantaged children—including the Perry Preschool Project and the

Carolina Abecedarian Project—have found positive effects on students’ IQ scores at an early age12 and

increased high school graduation rates among females, higher adult income, increased employment among

males, and reduced number of arrests.13 Head Start has been shown to increase students’ reading skills.14

Looking to older disadvantaged students, various rigorous studies have found that money spent on dual

enrollment programs in high school have many significant positive effects on educational outcomes, including

high school attendance and academic achievement, college readiness, access, and enrollment; and

educational attainment.15

In a notable 2016 study, Northwestern University professors C. Kirbo Jackson, Rucker Johnson, and

Claudia Persico studied the effects of court-mandated education finance reform on school spending, linking

court-ordered school finance reform (1972–2010) to a nationally representative longitudinal data set that tracks

individuals from childhood into adulthood. They found that increased spending on educational quality and

quantity yields large improvements in educational attainment and improved labor market outcomes (i.e.,

increased wages and family income, reduced adult poverty), especially for low-income children.16 Although

critics question some of the study’s conclusions as unrealistic, they agree that how schools spend money

matters.17

A 2016 study by economics professors Julien Lafortune, Jesse Rothstein, and Diane Whitmore

Schanzenbach found that investing “sharp, immediate, and

Page 271

sustained” funding into low-income school districts can have large effects on students’ educational

achievement (measured in National Assessment of Educational Progress (NAEP) scores) as compared with

states that did not increase funding to lowest-income districts.18 Given this mounting evidence, Hanushek

modified his outlook, writing in 2016 that “a general consensus” appears to agree that “how money is spent is

much more important than how much money is spent.”19 Investigators conclude that increased funding by

itself may not ensure improved outcomes, but providing adequate funding may be a necessary condition. But

simply giving schools more money—without directing and administering them differently—is not likely to

generate systematic improvements in student outcomes.

Most recently, a 2018 study focused on the relationship between school budgets and student success

in California. The state’s Local Control Funding Formula (LCFF), established in 2013 to address resource

inequities, sent $18 billion in increased state support over 8 years to districts based on the number of enrolled

“high-need” students.20 Districts largely targeted their new dollars to reduced student–teacher ratios,

increased average teacher salaries, instructional expenditures, enlarged public PreK (4-year-old), and special

education. The study found that increased spending directed to student needs led to significant increases in

high school graduation rates and academic achievement, especially among children from low-income

families.21

Yet despite the data consistently showing the relationship between increased targeted spending and

student achievement, many states are reducing their school funding. In A Punishing Decade for School

Funding (2017),22 the Center on Budget and Policy Priorities, a nonpartisan research and policy organization,

reported that as of 2015 (the latest education spending data from the Census Bureau), 29 states were

providing lower per pupil funds than in 2008, and 12 states had cut their general funding formula for

education.23 Instead of restoring education funding, seven of the 12—Arizona, Idaho, Kansas, Michigan,

Mississippi, North Carolina, and Oklahoma—had enacted income tax cuts costing millions from state

treasuries.24

Of course money, by itself, cannot improve student outcomes. But an adequate amount is necessary;

and when correctly targeted, money has been shown to do just that.

WHERE SCHOOL FUNDING IMPROVES STUDENT ACHIEVEMENT

The data show that increased spending directed to delivery of quality instruction directly to students produces

the greatest achievement return for the dollars spent. In the following sections, we provide evidence that

increased spending on teacher quality (effectiveness), professional development, reduced class size and

school size, increased teacher salaries, and improved school facilities can produce improved student

outcomes. We also provide data to challenge when and how these strategies might—or might not—be

cost-effective.

Page 272

Teacher Quality (Teacher Effectiveness)

Effective teachers are widely recognized as the school’s most important contribution to student learning. The

Coleman Report suggested that teacher effectiveness was a critical school factor in student achievement.

Later research indicates that qualitative variations among teachers can make large positive differences in their

capacity to generate student learning, year after year.25 Since studies confirm that consistently working with

highly effective teachers can overcome the academic limitations placed on students by their family

backgrounds,26 when teacher quality varies across schools and districts in ways that systematically

disadvantage poor, low-achieving, and racially isolated schools, it becomes an equity issue.27

Every school wants effective teachers in every classroom but are not sure how to identify them. Many

researchers have identified state certification (or license) in the subject they are teaching as a positive

predictor of higher student achievement.28 Nonetheless, investigators also find that traditional methods of

identifying teacher quality—a bachelor’s degree, a state license, and a relevant college major—although

important, are not enough to differentiate teacher effectiveness in generating student learning. Fortunately,

improvements in teacher observations and evaluation of performance,29 advances in assessment strategies

and data processing capacity, the availability of statistical modeling, and access to large longitudinal databases

of student achievement information—plus the ability to manipulate these data—now make it possible to identify

teachers’ classroom behaviors that help them generate student learning.

Recognizing the complexity of teaching, educators and researchers agree that the following teacher

effectiveness factors relate to increased student academic growth.30

• Verbal and cognitive ability

• Subject matter knowledge (tied to state standards)

• Skillful teaching behaviors

• Ability to create a classroom climate for learning

• Ability to adapt instruction to differing student learning needs

• Participation in intensive, content-focused professional development with classroom practice and feedback

• Teaching experience (those with fewer than 3 years’ experience tend to be less effective, but there is little

evidence that, in general, more than 3 years produces greater student achievement)

• Professional behaviors (including reflection and maintaining expertise in content and pedagogy).

These teacher effectiveness characteristics make sense. Effective teachers need to be able to speak

and think fluently if they are to clearly communicate lessons to students and make necessary instructional

adjustments. They also need the content knowledge—aligned with state standards—if students are to learn the

material

Page 273

the community expects them to know and be able to do (and on which the state will assess them). Classroom

experience linked to student performance and professional reflection helps hone their expertise. In addition,

research in the past decade has confirmed that certain instructional practices (i.e., planning, instructional

delivery, student assessment, learning environment, professional behaviors, and personal qualities) can

differentiate more effective from less effective teachers in generating higher student academic growth.31

Several studies show effective teaching behaviors linked with increased student achievement. In a

2012 meta-analysis of independent studies, education researcher Robert Marzano and colleagues found a

16-percentile point gain in student achievement in reading and mathematics on state achievement tests for

students whose teachers used specific instructional strategies (empirically shown to increase student learning)

over students whose teachers did not. And the more effective strategies teachers use and the more proficiently

executed, the higher their students’ achievement.32 Likewise, a 2011 Chicago Public School study found that

their new teacher evaluation program (including classroom observations, trained observers, and performance

feedback around a rubric describing varying degrees of teaching effectiveness) led to students of teachers with

the highest ratings on the performance rubric showing the greatest achievement growth (whereas students in

classrooms with the lowest rated teachers had the lowest academic growth).33 Similarly, the 2013 Bill and

Melinda Gates Foundation’s Measurement of Effective Teaching (MET) study found that randomly assigning

students to teachers previously identified as effective on a performance rubric showed greater achievement

growth than students with other teachers in the same school, grade, and subject rated less highly.34 Lastly, a

2017 study found that the effects of individual teachers on student achievement are large, especially among

the most and least effective teachers.35

Likewise, value-added measures (VAM) studies—numerical analyses using sophisticated statistical

algorithms to estimate or quantify how much of a positive (or negative) effect individual teachers have on

student learning during a given school year—are finding teacher effects on student learning can be cumulative

and long-lasting.36 Students in classrooms with very high value-added teachers can achieve a whole year’s

difference in learning gains as compared with a student working with a low value-added teacher.37 Similarly,

students who work with an above-average teacher for five years in a row are predicted to overcome the

achievement gap typically found between students qualifying for free or reduced-price lunches and those from

higher income backgrounds.38 Students assigned to high value-added teachers are more likely to attend

college, earn higher salaries, and are less likely to become teenage parents (than peers with low value-added

teachers); and replacing a very ineffective teacher (bottom 5%) with an average teacher would increase the

value of students’ lifetime income (in today’s dollars) by about $250,000 per classroom.39 One VAM study

calculated that a teacher one standard deviation above the mean on teacher effectiveness generates more

than $400,000 in future student earnings in today’s dollars with a class size of 20; and replacing the bottom

5–8% of teachers (highly ineffective) with average teachers (moderately effective) would advance the United

States near the top of international math and science rankings with a present value of $100 trillion.40 These

breath-catching VAM analyses reinforce the idea of teachers’ importance to student learning gains that carry

over into adult earnings potential.

VAM studies come with cautions. Although useful for diagnostic purposes as a measure of student

progress or as a formative guide for instruction, some policy makers, economists, and educators express

reservations about VAM’s technical and implementation issues and question their usefulness for evaluating

individual teachers’ performance for compensation, promotion, assignment, employment, or making policy

decisions.41 First, problems with missing data, the lack of

Page 274

random assignment of students to teachers or families to schools (selection bias), and absence of information

about available school resources may distort findings. These factors make teacher effectiveness appear to

vary over time and be unevenly distributed across districts.42 Second, different VAM models produce different

results, and results for individual teachers can be notoriously volatile and inaccurate. Given VAM’s sampling

error, some successful teachers might receive low ratings and some unsuccessful teachers could receive high

ratings. Third, VAM can be especially unfair to teachers in more challenging classrooms—students who

achieve above or below grade level, English language learners, students with special needs, or in high-needs

schools—while unduly rewarding teachers in less challenging classrooms. This could discourage effective

teachers from working with more difficult students. Fourth, multiple data sources beyond standardized test

scores—including skilled observations and evaluations by trained supervisors, mentor teachers, and peer

evaluations, student growth scores, and other evidence of student learning—are also needed to fairly and

accurately assess an individual teacher’s effectiveness or make human capital decisions. In short, VAM scores

are not valid or reliable proxies for an individual’s teacher quality and must be used carefully and appropriately

in high-stakes contexts.43

Knowing how much high-quality teachers and teaching contribute to student (and to later adult)

achievement, education dollars appear well spent in hiring, developing, and retaining the most effective

teachers possible. Effective teachers leading every classroom is a school leader’s priority.

Professional Development

A growing body of research shows that improving teacher knowledge and teaching skills are essential to

raising student performance.44 Once in the classroom, teachers continue to master new subject content, adopt

more effective instructional methods, develop techniques to address more students’ learning needs, infuse new

technology into lessons, and address changing laws and procedures. Since students spend most of their

school hours either interacting with teachers or working under their direction, what teachers know and can do

directly affects the quality of student learning.

A meta-analysis of more than 1,000 studies found that students scored 21 percentage points higher

than average on standardized tests when their teachers had access to ongoing professional support and

training.45

Although we do not yet have evidence to indicate which professional development (PD) features can

generate improvements in student learning (although the findings suggest effects),46 research consensus

suggests that the following six core features of effective PD increase teacher knowledge, skills, and practice:47

1. Form: Teachers’ PD is school-based and job-embedded, directly connected to their day-to-day classroom

actions and conducted either alone, with mentors, or in teacher teams.

2. Content focus: Teachers need to know well the content they teach, its alignment with state academic

content standards and assessments, how students learn (including common student misunderstandings or

problems), and successful instructional strategies that connect these.

3. Collaborative participation: Organized around groups of teachers from a school who work together on

meaningful content (such as peer observations of practice, peer coaching analyzing student work and student

data, study groups), and participating in teacher networks beyond the school, PD makes practice public and

open to critique.

4. Active learning: PD is most effective when it includes “sense-making”—opportunities for teachers to

intellectually engage by working directly on incorporating the new techniques into their instructional practice

that motivate them to expand their beliefs and abilities and apply the PD to classroom teaching.48

5. Coherence: Teacher PD should be a comprehensive, interrelated process aligned with state and district

policy expectations and using relevant data focused on guiding improvement, strengthening student learning,

and consistent with teachers’ knowledge and beliefs.

Page 275

6. Duration: Intensive, ongoing, long-term professional learning activities should total a substantial number of

hours each year (research suggests at least 20 hours or more of contact time).49

Each of these six features has cost implications. PD form, active learning, collective participation, and

duration require teacher and mentor time, likely during and after the regular school day and year, depending on

the strategies chosen. Release time from teaching (with substitute teachers provided), scheduled time in the

contract year, stipends for PD after regular work hours, full or partial reimbursement for college tuition or

conferences, and travel cost money. Expenses also include administrative time, materials, and supplies

required to support PD.

Most school districts do not know how much they are spending for educators’ professional

development. Traditionally, states and school districts have had weak financial reporting systems, presenting

data without consistent or uniform definitions, data items, categories, formats, or tracking. Rather, they use

broad accounting codes such as “instructional support” that include more than simply PD spending or ignore

PD in reporting expenditures for special or compensatory education. As a result, isolating professional

development spending is difficult.50 Today, national standards and guidance for school system accounting is

available to remedy this problem.51 Learning Forward, an organization dedicated to educators’ PD,

recommends that school districts dedicate at least 10% of their budgets to staff development, with at least 25%

of an educator’s work time focused on learning and collaborating with colleagues.52

School leadership is essential to create a culture of high expectations, continuous instructional

improvement, and highest-quality professional development tied to identified areas of teacher and student

need. Inviting a motivational speaker to address the faculty at the start of school may be a short-term morale

booster, but such practices should not be confused with a well-designed PD program for lasting impact that

improves teaching and student learning.

Reduced Class and School Size

Teachers and parents have long known that, all else being equal, when it comes to class size, smaller is better.

Over the past 100 years, class sizes and their pupil–teacher ratios have dropped substantially—from 35:1 in

1890, to 28.1 in 1940, to 24.9:1 in 1960, to 15.4:1 in 1990 to 16.2 in 2014.53 Many reasons explain this

Page 276

decrease: hiring more teachers to meet court-mandated desegregation; rising enrollments of special-needs

students who require more intensive and personalized student/teacher interactions; and recognizing that

improved learning occurs in classes with fewer than 30 to 40 students. In addition, the pupil–teacher ratios

dropped as schools attempted to add a modest amount of time for educators to collaborate with colleagues

and to obtain PD during the school day.

The number of students in a class can affect how and how much students learn. It affects how much

time and individual attention teachers can direct toward each pupil and his/her specific needs. Class size can

affect how the teacher organizes and delivers instruction; it is easier to assign more writing assignments and

give more feedback to fewer students. It also influences the extent of active pupil engagement with the

coursework, with the teacher, and with each other while also reducing the number of discipline disruptions.

Many studies of varying quality have addressed the class size–student achievement connection. In

what is arguably the most influential study, Tennessee’s Project STAR (Student/Teacher Achievement Ratio,

1985–89), successfully controlled experiments of class-size reduction in primary grades, showed strong

positive results.54 Findings indicate that attending small classes (average 15 students, as compared with

“regular” classes averaging 22 students) for 3 consecutive years in grade K–3 with an experienced teacher is

associated with sustained, statistically significant achievement differences (equal to about 3 additional months

of schooling). Academic benefits in all school subjects lasted through grade 8. These gains were especially

notable for low-income students.55 Further, an evaluation of Project STAR’s long-term impacts found

kindergarten test scores highly correlated with the individuals’ college attendance, earnings at age 27, home

ownership, and retirement savings (although the effect on achievement test scores faded in later grades),56

and reduced criminal arrests.57 In addition, small kindergarten class size showed benefits in non-cognitive

competencies—such as self-discipline, motivation, persistence, and social skills—that have positive returns in

the labor market.58

Although some have criticized the STAR design, implementation,59 and methodology, other studies

have found similar,60 mixed,61 or no positive62 effects. Notably, a review and synthesis of more than 100

class-size studies suggests that the most positive effect of small classes appears in kindergarten to third grade

for mathematics and reading test scores, with results consistent across schools.63 Findings are mixed about

whether students moving from a small class size to a “normal” sized class can maintain their achievement

gains.64 Research also finds that consecutive years in small classes mostly benefits minority and urban

students.65 Some hypothesize that reducing class size can lead to higher test scores overall and might reduce

the achievement gaps that exist between minority and other students in reading and math.66

Even so, studies conclude that reducing class size by itself does not result in greater academic gains.

Class size effects vary by grade level, pupil characteristics, subject areas, curriculum rigor, teaching quality,

accountability measures, number of years students remain in classrooms with few pupils and effective

teachers, and other variables.

What is more, reducing class size is very expensive. Although a cost–benefit analysis showed STAR’s

return on investment was slightly larger than the costs of implementing it (an estimated $2 return for every $1

spent),67 other studies indicate a small class size has a minor achievement gain as compared to many less

Page 277

costly strategies.68 Also, because class size’s effect on achievement varies with the subject taught and grade

level, these factors as well as student ability also figure into accurately determining cost effectiveness of class

size reductions.

Reduced class sizes and shrinking pupil–teacher ratios have major implications for school finance.

Reducing class size frequently means hiring more teachers (salaries and benefits) and leasing or constructing

additional classrooms with additional utility costs. It will likely require PD (and related costs) to improve

teaching if educators are to gain maximum advantage of the smaller class sizes. Alternate educational

investments such as raising teacher salaries, implementing better curriculum, or investing in high-quality early

childhood education may be more cost-effective. Moreover, if using class size reduction as a school

improvement strategy applied across a state or large local school system means lowering teacher quality

standards in order to fill the additional classrooms, student academic outcomes may be disappointing. In most

cases (except for K–3), the cost of class size reduction tends to exceed its benefits.

The link between school size and student achievement has also attracted educator and policy interest.

Since 1990, a growing body of evidence questions whether larger schools provide better academic (i.e., more

specialized instruction and varied curriculum) or fiscal results (increased administrative efficiency and reduced

per pupil costs from economics of scale) than smaller schools. The empirical evidence of the relationship

between school size and student achievement is mixed.69 Methodological issues—such as not accounting for

student or community SES—may partly account for these inconsistent findings.70

Studies suggest that students in smaller schools have more qualitative than quantitative benefits.

These include enhanced school climate and student engagement: higher attendance, greater feeling of safety,

fewer behavioral problems, and higher participation in extracurricular activities;71 lower high school dropout

rates;72 and better interpersonal relations.73 Research on small schools also finds that these do not

necessarily result in higher measured academic achievement.74

Importantly, the small school research does point to the connection between student achievement and

their socioeconomic status (SES).75 Studies agree that students’ SES—rather than school size—was the

strongest predictor of student achievement results,76 suggesting that smaller schools may mediate between

student poverty and academic outcomes, somewhat ameliorating poverty’s negative effects.77 This benefit

may result from the individual attention that teachers in smaller schools can provide students in their

classrooms as well as their help outside the classroom navigating the schools’ complex social and

administrative environments. But since studies point to the persistent effect of SES on achievement regardless

of school size, researchers question whether school size has a direct effect on achievement.

By comparison, several studies find that larger schools may benefit high SES students.78 For high SES

students, larger schools may create more opportunities—in availability of courses and activities, smaller class

sizes, and more experienced and effective teachers, administrators, and support personnel—for improved

performance.

Page 278

Of course, large and small are relative terms. Studies suggest that small- and moderate-size high

schools foster more positive social and academic environments than large high schools. Lee and Smith (1997),

using a national longitudinal database to determine achievement gains in mathematics and reading scores of

nearly 10,000 high school students, found a curvilinear relationship between high school size and

achievement. Achievement was highest in “moderate” size high schools (600–900 students) whereas

achievement in very small high schools (enrolling 300 or fewer students) or very large high schools (over 2,100

students) was lowest.79

Cost effectiveness is another consideration. It is true that while smaller schools are more expensive to

operate on a per pupil basis, researchers now assert that small schools can be more efficient when measured

on a cost-per-graduate basis because they graduate more of their students80 (as well as reduce social costs

of lower earnings and higher rates of unemployment, incarceration, and welfare dependence).81 Similarly,

Andrews’ et al. (2002) study determined that moderately sized elementary schools (300–500) and high schools

(600–900) may best balance costs and benefits in student performance and school services (relative to

increased numbers of teachers, administrators, and support staff).82

Although smaller schools may have many climate and academic benefits over large schools, advance

excellence and equity, and exhibit cost effectiveness if measured by cost-per-graduate, the relationship

between school size and student achievement is unclear. Since many of the studies are correlational, it is likely

that school size influences student outcomes indirectly through its academic and social organization—its

curriculum, teaching effectiveness, organizational and interpersonal practices, per pupil expenditures and

community SES. The extent to which school size can better facilitate or hinder these desirable practices is

what impacts student learning and achievement.83

Teacher Salaries

For decades, teachers have been losing ground economically. Since 1994, teachers have seen a

21.4-percentage point drop in their compensation relative to other college graduates and professionals. From

the mid-1990s to 2018, public school teachers’ relative wage gap (regression adjusted for education,

experience, and other factors) went from [minus] –5.3% to a record [minus] –21.4%.84

The Great Recession (2007–09) saw state revenues drop, state budgets cut, and in certain states,

income and business taxes slashed. As of the 2017–18 school

Page 279

year, at least 12 states had cut “general” or “formula” funding, the main vehicle for financing PK–12 schools, by

7% or more per student over the prior decade.85 Local school districts have difficulty making up for this lost

state revenue. For the school year 2016–17, the U.S. average public school teacher earned $59,660, ranging

from a high of $81,902 (New York) to a low of $42,925 (Mississippi).86 Cost of living differences contribute to

these extremes. A 2015–16 national survey found that more than half (55%) of public school teachers say they

are not satisfied with their salaries.87

As Figure 11.1 illustrates, average 2018 teachers’ weekly wages were 32.7% lower than those of other

college graduates ($1,195 vs. $1,777 and 32.7% less in wages than comparable workers.88 Teachers also

receive less premium pay (for overtime), less paid leave, and fewer wage bonuses than do other

professionals.89 As a result, teachers are now five times more likely to have a part-time job than the average

full-time worker.90 Making matters worse, state and local “austerity budgets” after the Great Recession cut

teacher wage growth at the same time as private sector wage growth accelerated.91

Of course, it is a mistake to assume that every college degree-requiring occupation should receive the

same salary, and supply and demand for certain skills—not years of schooling—determine wages. Additionally,

not all agree that teachers are underpaid.92

Some suggest that the trade-off between wages and benefits may account for some of the income disparity

between teachers and other college-educated professionals. For teachers, benefits—including health

insurance and pensions—are valuable parts of their total compensation package (although teachers pay

$460.16 per month, on average in 2016, for family health care coverage).93

In 2018 the Economic Policy Institute (EPI), a non-profit, nonpartisan think tank, found that as a share

of total compensation, teachers’ wages make up 70.9% as compared with 78.5% for other professionals.

Teachers’ health insurance and pension benefits comprise 29.1% of their overall compensation versus 21.5%

Page 280

for other professionals. Overall, in 2018, these benefits modestly offset teachers’ compensation penalty (a

21.4% wage penalty minus a 8.4% benefits advantage produces a –13.1% compensational penalty).94

The issues of teacher salaries, teacher quality, and student achievement are connected. The evidence

linking teaching quality and student achievement is overwhelming.95 Plus, investigations find that higher

salaries can attract and keep effective teachers in high-needs schools. Studies in North Carolina96 and

California97 found that targeted financial incentives to attract high-quality teachers to low-performing

schools—annual $1,800 bonuses to certified teachers in high demand subjects in North Carolina and $20,000

fellowships to attract academically talented novice teachers in California—drew high-quality teachers to

challenging schools and reduced teacher attrition by 17% and 75%, respectively. Salary is a proven method of

attracting a larger applicant pool, and selecting the highest quality teachers from that pool is key (although it

may take meaningful adjustments in school culture and climate to keep them).98

Linking empirical evidence of increased teacher quality with higher teacher salaries is difficult because

most school districts have salary schedules based on years of experience and earned academic degrees

rather than on student achievement. Also confounding the link between teacher salaries and student

achievement: more experienced teachers tend to generate more student learning, but beyond the first five

years in the classroom, teaching experience is not significantly related to student achievement.99

Economists Sylvia Allegretto and Lawrence Mishel observe, “The opportunity cost of becoming a teacher and

remaining in the profession becomes more and more important as relative teacher pay falls further behind that

of other professions.”100 States with lower salaries and poorer working conditions have larger teacher

shortages, especially for schools serving children of color, with disabilities, and English learners.101 The

growing wage and compensation differences make it more difficult to recruit and retain talented teachers.

Raising teachers’ salaries is a critical part in any strategy to recruit and retain a better teacher workforce.

School Facilities

School facilities contain the educational process and create the learning conditions. Yet according to the

American Society of Civil Engineers, in 2017, U.S. public schools rated “D+.” Fifty-three percent of public

schools needed to spend money on repairs, renovations, and modernization to put their building in good

condition; 24% of schools were in “fair” or “poor” condition.102 This national underinvestment in public schools’

facilities and their infrastructure—indoor and outdoor lighting, plumbing, heating/air conditioning, ventilation,

electrical, security, fire suppression, telecommunications, furnishings, playgrounds, parking, and

aesthetics—profoundly undermines teaching and learning.

The relationship between unhealthy school facilities and teacher/student productivity has its own name:

“sick building syndrome” (SBS). Its symptoms—the irritated eyes, nose and throat, upper respiratory infections,

nausea, dizziness, difficulty concentrating, headaches, forgetting, irritability, and fatigue—can result from the

schoolhouse’s poor indoor air quality (IAQ).103 Causes of SBS may include poor ventilation, high levels of

dust, rooms with poor lighting, mold or fungus, chemicals in the air from cleaning products, heat or low

humidity, insect or animal droppings, or noisy work environments. Although building

Page 281

components overall may not affect student performance, certain aspects such as noise can make it difficult for

students to think, concentrate, and learn; they hinder students’ cognitive skill development. In addition, poor

IAQ can aggravate teacher and student illnesses, prompting lost attention and absences. Building conditions

are especially influential at the elementary school level.

Strong evidence exists that school buildings’ conditions impact students’ health, student achievement,

and teacher effectiveness. Earthman’s 2002 seminal study found a 5 to 17 percentile point difference in

standardized test scores for students in good facilities (well-maintained buildings with comfortable room and

hall temperatures, satisfactory lighting, appropriate noise levels, good roofs, sufficient space) as compared with

those in inadequate facilities (poorly maintained with too cold or hot rooms, inadequate lighting, high noise

levels, leaky roofs, overcrowding) controlling for the students’ socioeconomic status. Older buildings typically

lack the conditions necessary for proper learning environments, adversely affecting student achievement.104

Many studies confirm,105 although some dispute,106 these findings.

Most recently, a 2016 meta-analysis on the relationship between school building condition and student

performance found a significant link between facilities’ conditions and student performance in science,

language arts, mathematics, and composite scores. The size of the correlation varied as a function of the

number of moderator variables (including building condition feature measured, instrument used, subject area,

and school level assessed). These differences may account for the lack of consistent findings in earlier

studies.107

Furthermore, perception often influences (and perhaps creates) a person’s reality. Quality facilities are

significantly related to school climate, and school climate plays a mediating role in the relationship between

facility quality and student achievement.108 Teachers working in buildings in poor condition state that the

facility’s design and appearance has a negative impact on the learning climate whereas teachers in buildings in

good condition report that the building has a positive influence on the learning climate.109 In a national survey,

the U.S. Department of Education reported that 48% of teachers transferred to another school, and 39%

changed careers because their workplaces were in states of disrepair.110 Working in substandard buildings or

in newer buildings that are poorly maintained undercuts teachers’ morale and increases their work frustration;

these attitudes likely transfer negatively into their classroom expectations and practices, reducing student

achievement. Similarly, inadequate school facilities directly impact a student’s motivation, energy, attention

level, listening capacity, and visual retention.111

Fixing school facilities is expensive. School construction projects are costly—and becoming more so. Between

adding space to existing buildings ($3.2 billion), retrofitting and modernizing existing structures ($3.14 billion),

and erecting new facilities ($7.8 billion), U.S. school districts spent more than $14 billion on completing

construction projects during the 2014 calendar year. The median cost for an elementary school was over $16

million, for a middle school was $26.5

Page 282

million, and a high school was $45 million.112 And construction costs keep rising. In March 2018, the producer

price index for goods needed for construction—steel, concrete, lumber and plywood, diesel fuel, and other

building materials—rose 5.8% over the prior 12 months.113 What is more, building schools costs more than

other types of development. Schools must be “high abuse rated” to withstand the wear and tear from

thousands of students who will use the building during its life span. Likewise, schools bring a slower return on

investment (schools are built to last 50 years or more before receiving upgrades as compared with commercial

buildings that renovate after 10–15 years).114

Providing satisfactory (or better) school building and infrastructure is both a learning and an equity

issue. Too often, low-income students learn in the facilities that need the most repair.115 With research clearly

affirming that school funding carefully targeted on enhancing teaching quality and providing safe and

comfortable facilities makes a measurable difference in student achievement, communities and school leaders

benefit when they educate the public about the lost “opportunity costs” of attending school in inadequate

facilities and then acquire the needed resources to improve them.

CONCLUSION

Education dollars appear best spent in hiring and keeping the most effective teachers, providing meaningful

and ongoing professional development, and maintaining school facilities to permit safe, supportive, and

comfortable learning environments. Maintaining low teacher–student ratios in the early grades also has a

positive impact on student achievement but at a rather high cost. As for optimum school size, it depends.

CASE STUDY

You have been hired as a consultant to the Alpha School District, a neighbor of Omega School District. Twenty

years ago, Alpha School District was known as the state’s premiere school system. But for the past 10 years,

Alpha’s test scores have been declining—very little at first but somewhat noticeably in the last 2 or 3 years. By

comparison, Omega School District, a suburban area on the Alpha’s outskirts, now headlines in the papers as

the state’s “Best School District.” Alpha failed to make sufficient academic progress last year.

Page 283

Both Alpha and Omega school districts have adequate fiscal capacity, but effort has been declining in

Alpha while increasing in Omega. Omega’s starting teacher salary now outpaces Alpha’s starting salary. Alpha

School District is willing to spend money on its schools if they have a good, research-based plan. Given the

data in Table 11.1, what suggestions would you offer the new superintendent of Alpha School District to

improve achievement? Justify your answers.

CHAPTER QUESTIONS/ASSIGNMENTS

1. Prepare a public relations brochure for community members in your district about what the research says

about spending and student achievement.

2. Identify positive, proactive steps your school district can take to make certain they hire and retain the

highest quality (most effective) teachers possible.

3. Explain the factors you would consider as you collaborate with others to plan a year’s worth of professional

development for teachers in your school.

4. Describe the research about class size and student achievement. Explain how that relates to what is taking

place in your school district and in comparison to surrounding school districts.

5. Enumerate several specific advantages of small high schools. Explain how to calculate costs so that small

high schools are more cost-effective.

6. Discuss the impact of teachers’ salaries as compared with what other college-educated professionals earn,

on school districts’ ability to attract and retain effective teachers.

7. Discuss how school facilities affect student achievement. Select several criteria and rate the school

facilities in your district as compared to neighboring districts.

8. Examine your school district budget over the last 3 years. Identify the extent to which the budget is putting

money in areas that increase student achievement. How would you change the budget to have a positive

impact on student achievement?

Page 284 - 294