Esquire Theme by Matthew Buchanan
Social icons by Tim van Damme

17

Aug

Is Success Academy the Climate Change of K-12 Education? (Part 3 of 3)

The first two parts of our analysis show that Success Academy students have very high achievement, and that most of this feat is not caused by the selection of their students, but by the effectiveness of the network’s schools.  Therefore, identifying unique elements of Success Academy’s approach is worthwhile to facilitate the spread of best practices to other charter networks as well as traditional schools.  In this final part of our analysis, we analyze elements of the Success Academy model and attempt to point out components that seem unique or particularly well executed.

As a starting point, consider the work of Harvard economist Roland Fryer who identified practices common among New York City’s most successful charters and later implemented those practices in Houston and Denver schools.  According to his paper “Getting Beneath the Veil of Effective Schools: Evidence from New York City,” the following five practices are commonly employed at successful NYC charters.

  • Teacher Feedback
  • Data Driven Instruction
  • High Quality Tutoring
  • Instructional Time
  • High Expectations

From reading Mission Possible, a book about educational practices by Success Academy founder Eva Moskowiz, it is clear that each of the five are implemented at the Success Academy schools. These practices likely account for a portion of the overachievement at the network’s schools.  However, they don’t explain why Success Academy schools outperform other charter networks (Uncommon Schools and KIPP are good examples) that also implement these practices.

One unique area of the Success Academy approach that has received significant attention is its literacy program.  In June, Chalkbeat’s Patrick Wall published an article discussing how other charter networks have been visiting Success Academy to observe literacy lessons. 

Two areas that stand out are the early-grade book discussions and significant amounts of time dedicated to independent writing.  Consider the video below that captures a group for Success Academy first graders discussing The Araboolies of Liberty Street.

[Link to First Grade Book Discussion]

A number of things stand out from this discussion.  First, the framework of the conversation is highly structured.  Students sit with hands folded in their laps and put their thumbs on their nose when they would like to share an idea.  It is clear that students have been coached in how to introduce their thoughts to the audience using complete sentences.

Second, there is a focus on contextualizing individual thoughts within the broader conversation.  Phrases such as “I agree with Alan because…,” “The connection I am trying to make with him is…,”  “What I hear you saying is that…,” and “I disagree with Kaiser because…” all show that students are coached to present their ideas as responses to others in the group.  This consideration of the audience is an important component of communication, and it appears the Success Academy schools are teaching it explicitly.

Third, it is clear that the school’s culture is permeated with the common core standards and terminology.  Multiple students reference elements of the book to support their thoughts and one even says explicitly “I’m going to use evidence from the book.”  This frequent exposure likely makes students more comfortable with the terminology of the common core they encounter on state exams.

With new common core aligned tests, writing is a larger component of ELA scores. Therefore, another area that deserves attention is the Success Academy approach to independent writing.  Consider the video below where a first grade teacher introduces students to what the network calls writing workshop.

[Link to Writing Workshop Video]

Again the structure is evident.  Another thing that stands out from the clip is the sense that elements of the writing workshop are being presented to kids as games.  They time themselves to see how quickly they can set up their stations and will attempt to beat their record the next time.  They also build up the amount of independent writing incrementally, each time writing longer, until they reach 40 minutes at the end of the year. 

This approach seems to fit well with a recent New York Times opinion piece where UW-Madison math professor Jordan Ellenberg argued that teaching is most effective when kids experience it as a game.

Finally, the network makes efforts to standardize instruction across its schools.  At most traditional schools, teachers across a district are expected to accomplish the same general goals each year, but much of the specifics are undefined, resulting in drastically different learning experiences and outcomes for students.

At Success Academy schools, “the same lessons are taught on the same day across the same grades, not only in that one school but across the network.”  A few days before a lesson is taught across the network, a teacher is identified to deliver the lesson to his/her students, recording the delivery, and sharing it for other teachers to watch in preparation sessions led by assistant principals and leadership residents.[1]

In sum, practices identified by Roland Fryer in combination with these other elements that seem to make Success Academy’s unique are a starting point for traditional school principals and charter leaders looking to experiment with practices which may lead to better student outcomes in their schools.

Much like there will always be those who dispute pollution’s effect on global warming, most Success Academy detractors aren’t likely to be persuaded anytime soon.  Some of their arguments are legitimate and explain a small portion of the schools’ achievement, but ultimately they miss the larger picture.  Something is working at the Success Academy schools, and educators who refuse to pay attention are missing out on an opportunity to maximize the potential of their students.

by John Keltz and Jarod Apperson

John Keltz is a researcher for the Atlanta Public Schools.  He graduated from Case Western Reserve with a bachelors degree in Economics and Math.  He earned a Masters of Economics from The University of Wisconsin - Madison.

Jarod Apperson is a Graduate Research Assistant at Georgia State University.  He graduated from New York University with a degree in Finance and Accounting.



Is Success Academy the Climate Change of K-12 Education? (Part 1 of 3)

Former Vice President Al Gore released An Inconvenient Truth in 2006, making the case that the Earth had experienced climate change resulting from rising carbon dioxide emissions.  Eight years later, a Gallup poll showed that forty-three percent of Americans still do not believe effects of pollution have caused a rise in the Earth’s temperatures.

In the face of climate models which predict that adding carbon dioxide to the atmosphere would lead to higher temperatures, and strong evidence that both temperatures and the levels of carbon dioxide in the atmosphere have actually risen, a surprising number remain convinced that climate change is a farce, a hoax manufactured by leftist scientists.

The same year that Al Gore released his documentary, former New York City Councilwoman Eva Moskowitz opened the doors of her first charter school, Success Academy - Harlem 1.  In the eight years that followed, her schools have also faced a large amount of skepticism.

Despite mounting evidence to the contrary, a large and vocal group of observers dismiss the performance of Success Academy as unremarkable, believing it to flow from bad actors and incessant test-prep, rather than legitimate strategies which might be implemented elsewhere.

In this three-part series, we analyze Success Academy’s performance on the just-released 2014 New York state assessments, attempt to evaluate claims that Success Academy’s achievement may be the result of biased inputs, and finally look at some aspects of the model which make the school unique.

The crudest (and unfortunately most widely used) way of analyzing school performance is to simply compare schools based on their average scores.  Below are the schools ranking in the top 25 among 1,252 NYC tested schools.  Five of the 25 are Success Academy schools.

As most readers will be quick to observe, this sort of approach leaves much to be desired.  With no consideration of the population served by each school, comparisons are often unfair.  Instead, schools are best evaluated when they can be benchmarked against others who serve similar students.  Consider the graph below which plots the scores of all NYC schools against the Economic Needs Index developed by the Department of Education.

Remarkably, 62 percent of the variation in school scores is explained by the economic needs of the population served.  By considering the population served, we are able to reach a much more reasonable expectation for each school’s achievement and can identify peer schools to serve as benchmarks.

The graph below highlights Success Academy schools in this context, and using the selection tools to the right, readers can also highlight performance of other groups including the citywide gifted program and other charter networks. 

At every single Success Academy school, from the relatively low needs Success Academy - Upper West to the relatively high needs Success Academy - Bronx 1, performance is remarkable.  On average, Success Academy schools are scoring 39 points above what would be expected based on the economic needs of the students served.

To put these 39 points into context, education researchers often convert points to the corresponding number of years of education they represent.  In this context, we see that Success Academy students are on average 3.0 years ahead of the students who attend schools with similar economic needs.[1] 

This achievement is not only remarkable; it is also unique.  No other charter network overachieves at a similar rate.  Even the citywide gifted and talented programs (which typically require students to score in the 99th percentile for admission) do not show similar overachievement.  

After Success Academy, the two groups which show the greatest level of overachievement are the citywide gifted schools and the Ichan Charter Schools.  While the achievement at both of these is also impressive, the overachievement at Success Academy is double these groups. 

It is clear from the above analysis that Success Academy students significantly outperform their peers on state exams; yet, a number of important questions remain.  For example, can comparisons really be made between traditional schools and charters where parents must opt-in?  What about attrition?  What about English Language Learners and Students with Disabilities?  These questions and more are the subject of part two of our analysis. 

Continue Reading Part 2.

by John Keltz and Jarod Apperson.

John Keltz is a researcher for the Atlanta Public Schools.  He graduated from Case Western Reserve with a bachelors degree in Economics and Math.  He earned a Masters of Economics from The University of Wisconsin - Madison.

Jarod Apperson is a Graduate Research Assistant at Georgia State University.  He graduated from New York University with a degree in Finance and Accounting.


[1] The standard deviation of New York state exams is 35, so Success Academy has an effect size of 1.11 standard deviations.  The estimated effect size of a year of schooling in grades 3-8 is about 0.37.  Therefore, the overachievement at Success Academy corresponds to approximately 3.0 years of schooling.

Is Success Academy the Climate Change of K-12 Education? (Part 2 of 3)

We demonstrated in part one that Success Academy has very high achievement relative to the economic needs of the network’s students.  However, Success Academy skeptics believe their high scores might be explained by other factors: student opt-in, student attrition, and smaller populations of English learners and students with disabilities.  All these factors involve questions of selected student populations.  We consider each one below, as well as a comparison to schools that clearly have a selected population: citywide gifted programs.

Opt-In Bias

Charter schools are different from zoned schools because parents must choose to send their child and indicate their interest by submitting an application several months before the start of the school year. 

Research suggests students who apply to NYC charter schools, on average, scored higher in the previous year than the students who didn’t submit an application.  Therefore, a portion of the over-achievement that charters experience reflects the fact that applicants are a select group, a motivated group with the awareness and wherewithal to apply in advance.   An estimate of this “opt-in” effect suggests it may be as high as 9 scale score points on the NY exams. (The true opt-in effect is likely lower than 9. Please see the footnote for more details.)[1]

As shown in the graph below, once the opt-in effect is take into account, it is unclear whether some charters which initially appeared to be positively impacting their students are actually getting results beyond the opt-in effect.  Other charters including Success Academy, Ichan, and Uncommon Schools show overachievement significantly above the estimated opt-in effect.

Attrition Bias

A second issue that has been raised as potentially biasing Success Academy’s achievement is the suggestion that the school is “pushing out” students who it suspects will not score well on exams.  This loss of students is referred to as attrition.

In December of 2012, Beth Fertig and her colleagues at WNYC’s Schoolbook published an analysis of attrition across all NYC charters along with comparisons to the five NYC Bureaus as well as the CSD’s with the greatest number of charters.

The analysis revealed a few important things.  First, attrition is significant citywide and particularly high in communities with larger populations of low-income students.  Second, it showed that indeed some charters experience attrition rates above the citywide average and above the CSD’s where they operate. 

As shown in the chart below, Success Academy schools experienced annual attrition of 10 percent, approximately 3 percent less than the citywide average and about 4 percent less than would be expected based on the bureaus where they operate.

Backfill, another component of attrition that is discussed in a thorough piece by Chalkbeat’s Sarah Darville, relates to whether charter schools choose to fill vacated seats or whether they let cohorts shrink as students leave the school.  The Success Academy schools have a policy of backfilling grades K thorough 3 but not the older grades.

If leaving students on average are lower performers, the choice not to backfill would raise the achievement in later grades.  However, limiting analysis to the third grade, where Success Academy does backfill, results in overachievement of 38 points versus 39 for all grades.  This indicates the choice not to backfill does not explain much, if any, of Success Academy’s overachievement.

ELL & Special Needs Bias

A third area of potential bias flows from the percentage of students served who are English Language Learners and qualified for special needs.

Success Academy schools serve fewer ELL students than zoned schools in the districts where they operate.  Reviewing the percentage of ELL students at Success Academy’s tested schools indicates they serve about half as many ELL students (6 percent) as zoned schools nearby (12 percent).

While this is an important starting point, there is a confounding factor at play that must be considered.  The basic idea of ELL interventions is that the students eventually learn English and then move out of the program.  At Success Academy, 27 percent of students passed the reading & writing ELL exam compared to 19 percent at the district schools.  Similarly, 65 percent of Success Academy students passed the spoken ELL exam compared to 50 percent at their zoned counterparts.  This suggests that part of the reason participation rates are lower at Success Academy schools is because students attending the schools are more likely to master English and move out of the program.[2]

It is also worth noting that the performance gap between Success Academy ELL students and non-ELL students is relatively small, suggesting that if the schools were to serve a larger number of ELL students, scores would likely only fall marginally.  Ninety-one percent of ELL students passed the math exam while 41 percent passed the English exam.  Both rates are substantially higher than the city.[3]

Special needs is another area which often receives attention in the media; however, administrative data indicate that Success Academy schools serve a similar share of special needs students relative to their zoned counterparts.  Both serve approximately 12%.  Further, the data show the Success Academy schools are more effective at mainstreaming the students, with twice as many students moving to less restrictive settings.[4]

Comparison to Citywide Gifted and Talented

A final context within which to consider bias and the Success Academy achievement is in comparison to other schools with clear selection bias.  Recall in part one of this analysis, we highlighted the performance of citywide gifted programs.

In order gain admission to a citywide gifted program, students must take a standardized test.  InsideSchools reported that for the 2014 year, three times as many students scored in the 99th percentile as there were seats available at the five most selective citywide programs.

As a result of the limited number of seats and the large number of test-takers, students admitted to a citywide Gifted & Talented program are a select bunch.  In fact, they start out wholly in the top 1%.

For the issues of opt-in and ELL status discussed above, the Success Academy students may to some extent also be a select bunch.  However, it is quite challenging to believe they could ever be a more curated group than students required to score in the 99th percentile on an admissions exam.  Yet, surprisingly, the overachievement seen at Success Academy is double the overachievement seen at Gifted and Talented schools.

Consideration of potential sources of bias indicates that a portion of Success Academy’s overachievement may be explained by factors related to the student population served.  However, a majority of Success Academy’s overachievement remains unexplained by these factors; therefore, it seems the schools’ curriculum and operations are responsible for much of their success.  In part three of this analysis, we explore the Success Academy approach and look for elements that may contribute to the network’s performance.

Continue Reading Part 3.

by John Keltz and Jarod Apperson

John Keltz is a researcher for the Atlanta Public Schools.  He graduated from Case Western Reserve with a bachelors degree in Economics and Math.  He earned a Masters of Economics from The University of Wisconsin - Madison.

Jarod Apperson is a Graduate Research Assistant at Georgia State University.  He graduated from New York University with a degree in Finance and Accounting.


[1] This is likely an upper bound. Stanford economist Caroline Hoxby’s 2009 NBER working paper “Charter Schools in New York City: Who Enrolls and How They Affect Their Students’ Achievement” showed that students who applied to charter schools from tested grades scored 0.22 standard deviations (median) in math and 0.23 standard deviations (median) in reading above their peers who did not apply.  The standard deviation of scores on New York exams is approximately 35 points.  This suggests an opt-in effect may be somewhere around 7 to 9 scale score points. Since the time that Hoxby’s data was collected, the rates of application have risen significantly in some of the CSD’s with large numbers of charter schools.  With applicants representing a greater share of total students today than they did in 2006, the effect using 2006 data is likely greater than the effect today.  Recent lottery data would need to be investigated further in order to more accurately estimate the current effect.

[2] Data collected from the New York City Charter School Center.  The composite data reported above was developed weighting by the number of tested students from each school.

[3] Subgroup data was reported in a Success Academy press release.

[4] Data collected from the New York City Charter School Center.  The composite data reported above was developed weighting by the number of tested students from each school.

13

Jul

Can everyone win in the APS/Beltline dispute?

Co-written Op Ed for Creative Loafing.

04

Jun

A New Day In APS: Event Postponed Until Tomorrow

On a fanfare-filled day in late March, the newly elected Atlanta Board of Education announced that Austin Superintendent Meria Carstarphen had been selected to take the helm of the city’s schools. 

Carstarphen promised “a new day for APS” while Board Chair Courtney English said it was time to “believe again.”  The children at Hope Hill Elementary smiled, hugged the new leader, and posed for pictures.

The usual dissidents chirped loudly.  Union representative Verdailia Turner spoke of “red flags” and sent out an incoherent email listing all the conspiracy theories that she believed Carstarphen to be involved with. Ed Johnson complained that Austin might be more “starboard” while Atlanta was more “portside” on the ship of education.

More levelheaded observers promised to wait and see.  Sure, she had raised graduation rates in Austin but no faster than the state of Texas as a whole.  Sure, the district’s TUDA scores were high, but they hadn’t risen much during Carstarphen’s tenure, and most of the performance seemed to reflect Austin’s demographics more than any sort of education miracles. 

If nothing else, Carstarphen seemed energetic, decisive, and unafraid to speak directly, three qualities that themselves would represent steps forward for district leadership.

Nine weeks later, APS is still waiting on Carstarphen to arrive (she is scheduled to start July 7th) and still hoping she will bring with her a new day.  Signs of its future arrival are difficult to spot in the district, which just completed a disappointing hiring cycle of 19 new school principals. 

Impact of Strong Principals

For evidence of the impact an effective principal can have on a school, one needs look no further than Coan Middle School in the Jackson Cluster.  Just two years after taking on the role of school leader, Dr. Betsy Bockman lifted the school from the bottom of the barrel to one of the city’s top performers.  Today, sixth graders at the school outperform eighth graders at King Middle School down the street.  

It is difficult to pinpoint a single strategy that led to this transformation; instead, the results seem to be the culmination of several small things the school got right under Dr. Bockman’s leadership.  She doubled math instruction to help students catch up, she removed ineffective teachers from the classroom, and she called on students to take ownership of their education by setting achievement goals through the year.

After the Davis administration removed several school leaders earlier this spring, there seemed to be a possibility that more high-flying principals like Dr. Bockman would be attracted to lead the districts’ schools.  The need to hire 19 school leaders gave the district a chance to significantly raise the bar for leadership at about 25% of its schools. Weeks later, it appears that opportunity will be largely missed.

The district has now hired all 19 principals.  Ten were hired from outside the district while nine were already APS employees.  Of the ten hired from other districts, six came from Fulton County and four came from Clayton County.  Five had prior experience as a principal while five were assistant principals at their former schools.

Academic Background of New Hires

Before delving into the new hires’ employment history and the academic performance of schools they led, I first looked at the thing they all share—an academic background.  For context, let’s begin by considering the academic background of the district’s high-performing charter school principals.

APS has nine charter schools that outperform what would be expected based on the FRL populations they serve.[i]  Principals who graduated from a top-tier university lead all nine of those schools[ii].

image

This consistency is contrasted by APS where only 5 of the 19 new hires have ever graduated from a Tier 1 institution.  Despite having multiple degrees, seven of the hires have only attended the lowest-quality institutions.

image

Can good leaders come from low-quality academic backgrounds?  Of course.  Can a district find diamonds in the rough? Sure.  But districts seeking transformation may find an easy path forward by moving away from diamonds in the rough and toward pre-cut and polished diamonds. 

Copying best practices from our charter schools is not rocket science.  They aren’t parting the Red Sea.  They are just doing some basic things right, and one of those things is staffing schools with leaders who have themselves been academic success stories.  

Next, I took a look at the academic performance of schools managed by the new hires. 

Experience as Principal in Other Districts

Of the five principals hired who have experience serving as principals of schools in other districts, two have led schools with solid academic track records while three led schools with disconcerting underachievement.

On this front, Buck Greene (Sutton) and Marcus Jackson (Carver – Health) both show promise.  Buck was the principal at Johns Creek High School, one of the highest achieving high schools in Fulton County while Marcus led Kendrick Middle School in Clayton County.  Both schools earned CCRPI scores above what would be expected given the school’s population of FRL students.

On the other hand, Jesse Berger (Gideons), Tim Guiney (Grady), and Richard Fowler (Mays) all led underachieving schools.  In fact, all three led schools that were in the bottom 1/3 of the state relative to what would be expected based on their FRL rates (readers can evaluate CCRPI performance with the graphs posted here).

More troubling, Richard Fowler, whose school ranks in the 16th percentile statewide, has a worrying employment history, having served as principal of the low-scoring Fulton school for less than two years before seeking a job with APS.

image

While past performance is not always indicative of future potential, it is a strong indicator of leader quality, and it should only be ignored after a great deal of thought has been given and reasons for the lack of achievement have been discussed with the candidate.

Did HR consult the Research & Evaluation department to inquire about the historical performance of these principals?  If so, perhaps they determined that other qualities outweighed the achievement records.  If not, shouldn’t that be a standard part of the hiring process going forward?  Plenty of people can talk the talk in an interview, but the district’s students deserve leaders who have a track record of delivering results. 

Experience as Assistant Principal in Other Districts

Five of the hires from outside the district do not have experience as principals but have worked as assistant principals.  Evaluating the quality of assistant principals with CCRPI data is more difficult than evaluating principals because APs are not responsible for the overall direction of the schools at which they work.

Instead, one of the most relevant objective measures for these employees would be their employment history.  Of the five principals hired, all show cause for concern.

Four of the hires—Josie Love (Carver Technology), Tracie Astin (South Atlanta CAD), Amia Burnette (Bethune), and Laryn Nelson (Young)—come to APS from short stints at schools in Fulton County.  All of them have spent less than two years in their current roles.  Both Amia Burnette and Laryn Nelson have been moved around the county and most recently were moved from high school to elementary assistant principal roles.  Such moves are generally considered demotions. 

A Fulton administrator asked about the quality of candidates remarked that rejects from their district often land jobs with APS. 

One of the five, Ms. Thalise Perry has a more stable record of employment; however, according to those who interviewed Ms. Perry, the resume she provided contained several typos and she rambled rather than answering questions coherently. 

When APS hires staff from other districts, it needs to be aware of the lemon dance.  The district cannot continue to be the dumping grounds for leaders who fail to pass the leadership bar set by Fulton County and other districts in the area.  It needs to seek out leaders on upward trajectories, not settle for people looking for a new gig before their current employer asks them to leave. 

Even more concerning is the fact that parents recognized the lack of quality applicants but their voices were ignored.  Despite formally requesting a new slate of candidates, parents in the Mays Cluster were given the cold shoulder by the outgoing administration.  Three of the candidates they interviewed were hired to lead APS schools over objections about applicant quality. 

Hired From Within APS

As mentioned above, nine of the new hires are promotions from within APS or permanent appointments of interim principals.  Several of the placements seem promising.  Stimpson & Watkins both come from strong academic backgrounds.  Dr. Bockman was transformational during her time at Coan Middle School.  Mr. Douglass is reportedly respected by teachers and students at North Atlanta High School while Ms. Powell has been recognized for her efforts to raise graduation rates at Therrell.   

Still, APS doesn’t have enough quality leaders internally to staff all of its schools.  Instead, it is essential that the HR department also be capable of attracting outside talent to the system. 

With the cycle just completed, that didn’t seem to be the case.  Rather than rummaging the country for top applicants, APS hired the folks who landed in its lap.  Nine candidates were already working in the district, while 10 came from two adjacent districts.  Many of them have troubling performance histories, academic backgrounds, and employment records.  Though they may defy the odds and achieve at high levels, it is time for APS to choose leaders with the odds in their favor.

Still Waiting on the New Day

Four weeks before the new Superintendent’s arrival, there is still cause for hope.  Carstarphen has established new HR leadership with Pamela Hall and Skye Duckett coming from Austin.  Though Carstarphen approved all of the 19 principal hires, the process and recruitment were largely driven by the outgoing administration.  There is potential that Carstarphen’s HR team will move the department beyond its role of checking boxes into a role of recruiting and thoroughly vetting candidates to ensure quality hires.

Some evidence suggests that Ms. Hall and Ms. Duckett will have higher expectations of applicants seeking jobs with the district, particularly on the dimension of academic preparation.  Over the seven months ended March 31, 2014, APS hired a total of 202 teachers while the Austin Independent School District hired 212. 

A review of the institutions attended by each of those hires reveals that the most common Tier for the educational background of APS teacher hires was Tier 3, with almost twice as many new teachers coming from that Tier when compared to Austin.  In Austin, 70% of hires came from Tier 1 or Tier 2 institutions.

Though APS was able to hire 18% of its teachers from Tier 1 universities, most of those hires came through Teach For America.  In a dumbfounding move, the current administration decided to nix its relationship with Teach For America for the upcoming year, cutting off the district’s primary source of staff educated at the nation’s top institutions (Update: I understand that the district has decided not to pursue this plan and will be hiring TFA corps members this year).  Detailed hiring data can be accessed here

image

The APS of today seems to be the APS of yesterday.  Perhaps changes are being planned for tomorrow.


[i] Though KIPP Strive Primary and KIPP Vision Primary do not yet participate in the CRCT, their leaders have been included based upon performance on MAP assessments and the network’s track record with each of its other schools.

[ii] Tiers 1, 2, and 3 were developed using data released in the White House College Scorecard.  Tier 1 institutions graduate at least 75% of students who enroll in the institution.  Tier 2 institutions graduate at least 45% of students who enroll.  Tier 3 institutions graduate fewer than 45% of students who enroll.

01

May

Economics’ Manifest Destiny: Gary Becker and the Conquest of New Realms

Note: The following post is a paper I wrote this semester for a History of Economic Thought course.  We were asked to write about the economic contribution of a Nobel Laureate.  Though it is not directly related to the education topics I normally write about, after getting the paper back today I felt sad to think I had put effort into something that would likely never be read again.  So, I am posting it to the blog. 

By Jarod Apperson

Much like an aggressive first date, Gary Becker extended the invisible hand of economics into a host of uncharted territories.  From early theories framing racial discrimination through an economic lens to later work on drug addiction, most of Becker’s writings serve to expand the scope of economic theory beyond its original home in the world of production, demand and trade.  Today, believers in the supremacy of rational individual choice need no longer limit their faith to the banality of labor negotiations and the like, for Becker’s body of work develops a host of economic theories rationalizing the decisions individuals make through all aspects of life.  Over a prolific six-decade career, Becker’s contribution served to further establish the assumptions underlying neoclassical economic theory as natural laws, governing not just trade but all choices, laws which persist independently of context, culture, and time.  

Biography

Becker’s own telling of his personal history, as published in Les Prix Nobel 1992, betrays a faith in the malleability of life and the extent to which its course is dependent on an individual’s personal choices.  He chooses to highlight the “up-by-the-bootstraps” anecdotes commonly shared by tea-party members and other neoclassical enthusiasts as evidence that their current circumstance was mostly earned through their own grit and perseverance.  Present are references to the fact that his parents both dropped out of school in the eighth grade while he went on to earn a Ph.D. from one of the nation’s most respected economics programs (Becker 1992). However, scant recognition is made of the comparative early advantages afforded Becker by a literate father who owned his own business, a family that could afford to live in a respectable school district, and the famously successful New York Jewish community to which the Beckers belonged, providing consistent examples of achievement that could be modeled.  Rather than a life largely determined by these circumstances with marginal augmentations the result of his decisions, Becker understands his personal history as largely determined by a series of his own actions, a perception consistent with the neoclassical theory underlying much of his economic contribution.   

Just fourteen months after the 1929 Black Tuesday stock market crash ushered in the Great Depression, Becker was delivered to two immigrants, an eastern European mother and a Canadian father.  Though born in the small town of Pottsville, Pennsylvania, Becker spent the majority of his childhood in Brooklyn, New York where his family moved before he reached school age. Becker’s early education failed to garner his full attention.  Instead, the young boy’s interests leaned more toward athletics.  Meanwhile, Becker’s father was an avid reader of newspapers on politics and other current events.  As his vision began to erode, he enlisted Gary to read articles aloud.  This experience laid the foundation for Becker to build his understanding of social interactions.  By the time he reached high school, Becker had developed a more keen interest in academics and was particularly drawn to the study of mathematics, having joined the math team at age 16 (Becker 1992).

Becker went on to attend Princeton University before earning his Ph.D. in Economics from the University of Chicago where he studied under Milton Friedman.  Upon completing his dissertation, Becker remained at Chicago for three years before moving to Columbia University.  While at Columbia, Becker earned the John Bates Clark Medal.  Following a distasteful observation of the 1968 student riots at Columbia, Becker decided to return to Chicago where he remains as a professor today. 

Becker married the mother of his two daughters while completing his doctorate.  Following her untimely death in 1970, he spent a decade single before marrying Guity Nashat, a historian of the Middle East, whom he continues to be partnered with today.  Guity eventually convinced Becker to begin writing a monthly column for Business Week that he credits with helping him develop the capacity to refine his ideas and explain them to non-academic audiences (Becker 1992).

Major Contributions to Economic Theory

In 1992, Becker was awarded the Nobel Prize in Economic Sciences for his work in the fields of discrimination, crime, family behavior, and human capital (Becker 1993).  The first three each represented a new realm into which economic foundations could be applied to develop theoretical frameworks for understanding individual choices.  The last represented an augmentation of theory underpinning the economics of labor.  The remainder of this paper will focus on the contributions Becker made to these four fields.

Discrimination

Prior to the 1957 publication of Becker’s The Economics of Discrimination, racism was not widely seen as the sort of topic appropriate for economic analysis.  To an observer of the world, the choice of whether to discriminate or not might seem binary.  Businesses either set race-based criteria for the customers they would serve or they did not.  Neighborhoods were either segregated or they were not.  Schools either had rules prohibiting attendance by blacks or they did not.  The framing of these decisions began to change with the publication of Becker’s first book, which was an expansion of his University of Chicago dissertation and extended marginal analysis to racism. 

In understanding and evaluating Becker’s theory, it is relevant to consider the personal context from which he approached the topic.  During the years of Becker’s childhood, Brooklyn was home to immigrants from a variety of backgrounds; however, the communities were almost exclusively European, and 97.2% of the county’s residents were white according to the 1930 census.[1]  The lack of racial diversity in Becker’s immediate surroundings provides an interesting backdrop to his work on discrimination, suggesting that, despite any empathy he may have genuinely felt for those being analyzed, the theories are likely more informed by deductive conjecture than induced by personal relationships or observations from his own experience. 

Rather than establishing racism as a binary choice, Becker theorized that racism existed on a continuum.  As such, individuals harboring racist attitudes require compensating differentials in order to conduct business or otherwise interact with groups they would prefer to avoid (Becker 1957).  These individuals can then maximize their total well being by considering the race-adjusted marginal utility of a choice.  If the differential compensation were high enough, individuals would choose to conduct business despite their racist attitudes.  For example, one hiring manager might be considered a 25-cent racist, while another more bigoted manager, could be viewed as a 75-cent racist, requiring employees of a certain race to work for a wage at least 75 cents less than other similarly skilled workers in order to compensate the manager for his “distaste” for integrated working conditions.

While this theory has some appeal and Becker was able to develop an interpretation of empirical wage data in a way consistent with the notion, it also seems to have a significant weakness when viewed in the historical context of the time Becker was writing.  Princeton University historian Kevin Kruse’s book White Flight: Atlanta and the Making of Modern Conservatism chronicles the process of racial desegregation in the 1950’s and 60’s.  One finding that stands out from his work is the extremity of outcomes.  For example, Collier Heights in west Atlanta went from 100% white to 100% black within a few months (Kruse 2005).  At the same time, housing prices plummeted.  This outcome was not limited to a single neighborhood, but instead was a pervasive occurrence across the South.  Were Becker’s theory accurate, one would expect new whites who required lower compensating differentials to have been attracted to the neighborhoods as prices fell.  However, the reality is that such interest never materialized, and now 70 years later the neighborhood remains almost 100% black.  While some bigots may exist on the continuum described by Becker’s work, the empirical evidence from the period instead seems to expose Becker’s model as the exception with fully elastic demand as the rule.

Whether the work holds up under scrutiny or not, Becker’s first attempt at publication successfully shifted the conversation around discrimination and expanded the creed of rational choice to a new realm, an accomplishment that in itself is remarkable.  Much like Becker’s work on discrimination sought to develop a rational lens through which the decisions of racism could be understood, his later work on crime and punishment attempted to rationalize the choices of criminals.

Crime

            Imagine a tall, muscular man who has placed a gun to the head of over 200 individuals and, with the threat of death, demanded their belongings.  One Sunday morning, after steeping himself a cup of Earl Grey, he sits down to read The New Yorker, as is his weekend ritual.  Coming to an article filed under “Annals of Crime,” he learns that many states, including his own, have raised both the punishment for committing robbery and the likelihood of being caught.  After opening an Excel spreadsheet entitled “Tradeoffs and Benefits to My Criminal Activity” and making the appropriate adjustments for new information gleaned from the article, he learns that the expected, risk-adjusted return on his career path is now negative.  Aghast, he decides to set aside all criminal activity and applies for a job as a cashier at McDonalds, where he lawfully spends the remainder of his existence.  Such is the theoretical life of a criminal in Becker’s conception.

Delivering his Nobel Lecture, Becker reflected on the motivation that led him to pursue the field of criminal choices, citing a mid-century academic climate where behavior of criminals was largely attributed to factors outside the criminal’s control.  He “was not sympathetic to the assumption that criminals had radically different motivations from everyone else” (Becker 1993).  With an intent to counter this paradigm, Becker published “Crime and Punishment: An Economic Approach” in 1968.  The paper examined a wide range of criminal activity from vehicle theft to the violation of anti-trust laws, and it examined appropriate responses.  Again, the impetus for the work is relevant.  Becker had a personally held belief—not based on empirical evidence—that criminals were rationally motivated and he chose to develop a theory formalizing that faith out of frustration that an alternate perspective dominated the prevailing academic discourse.

Broadly, Becker’s work suggests that the penalties levied for crimes should reflect the damages resulting from the action with additional consideration given to the cost of enforcement and the possibility of shifting some of this cost to criminals themselves through higher penalties (Becker 1968).  In this effort, the argument succeeds when readers accept the basic premise that criminal choices flow from rational evaluations of risk and reward, but it falls short should the reader find this foundation implausible.  The approach appears informative for civil damages, where judges must determine how much to award the injured party in the case of patent infringement or negligence as these parties are often highly educated and it can be shown that significant amounts of thought have gone into their business decisions.  However, empirical evidence calls into question whether the theory can realistically be extended to wider criminal activity.  For example, work by Becker’s colleague Steven Levitt and Columbia University’s Sudhir Venkatesh evaluated a Chicago gang and showed that the typical member earned just $3.30 per hour, significantly less than the minimum wage (Levitt and Venkatesh 2000).  Not rocking the boat, the two authors explained their finding as confirmation that people work diligently toward the possibility of large payoffs, whether those payoffs are likely or not.  Yet, less zealous neoclassical adherents may consider it equally compelling evidence that criminal choices are often irrational, calling into question the broad applicability of Becker’s theory. 

Family Behavior

Continuing his expedition through life’s major decisions, Becker next tackled marriage and the family unit.  He published A Treatise on the Family one year after the consummation of his second marriage.  Forestalling any possibility this event might cause some to misconstrue him as a hopeless romantic, Becker’s contribution to the evaluation of family was “the assumption that when men and women decide to marry, or have children, or divorce, they attempt to raise their welfare by comparing benefits and costs” (Becker 1993).  The astute reader will notice the similarity between this work and Becker’s contribution to crime.  Essentially, it is yet another extension of the idea that individuals make life decisions by determining the actions that will maximize their expected utility.

            One unique aspect of Becker’s work on the family is the incorporation of altruism into utility functions.  Family members often make choices that are beneficial to another member with no observable value flowing to the individual taking action. A husband opening the car door for his wife might seem irrational under a standard utility function unless he somehow derives personal satisfaction from the process.  Recognizing this weakness in the theory, Becker developed a utility model which incorporated a measure of altruism (Becker and Barro 1986).  Consider the following utility function:

U0v(C0) + αU1

In this model developed by Becker, the utility function of person zero is said to incorporate altruism because her/his wellbeing is dependent not only on consumption but also on the utility of person one.  The α represents the extent to which the individual is altruistic.  So someone like Mother Theresa may have an α above one, indicating that she values the utility of others more than she values utility from her own consumption while Justin Bieber would likely have an α closer to zero. 

            Starting from this interesting foundation, Becker developed a theory wherein families choose the optimal number of children by maximizing such an equation.  Conceptually, the idea is compelling, and it helps explain why fertility rates vary across countries depending on the level of economic opportunity.  However, despite Becker’s claims to the contrary, the model’s construction seems to still be focused on rational self-interest, missing what some might call the whole point of altruism.  Yes, the utility of person zero is dependent on the utility of person one, but that alone doesn’t make person zero altruistic.  Instead, he/she could be fully selfish, yet recognize that his/her personal utility improved when the needs of others were met.  In contrast, the historical and philosophical concept of altruism connotes selflessness.  With this in mind, it seems that Becker’s contribution to family choices explained some facets of fertility observed in the real world, but the theoretical framework employed to achieve this end conflates standard concepts of selfishness and altruism.

Human Capital

            The final area of contribution explored in this paper differs from the previous three.  While each of those expanded economic theory along the extensive margin and entered what were historically considered the province of other social sciences, Becker’s contribution to human capital more intensively tilled the fertile ground of labor market theory.  Prior to Becker’s contribution, most labor market theory rested on a foundation of homogeneous agents.  With the publication of “Investment in Human Capital: A Theoretical Analysis,” Becker posited that the theory could be amended to introduce heterogeneity of workers through the incorporation of human capital (Becker 1962).

            Becker’s early work on human capital primarily focused on investment in the training of a workforce.  He identified several facts about the real world which did not seem to be consistent with the theoretical models accepted at the time.  Two facts were: earnings typically increase with age, and unemployment rates are negatively correlated with the level of skill.  Unless heterogeneity of marginal productivity could be established, the variation observed in wages and likelihood of experiencing unemployment might seem to contradict John Bates Clark’s work showing that under a competitive, capitalistic structure wages are determined by the marginal productivity of labor (Becker 1962).  Expanding Clark’s theory, Becker showed that when workers have accumulated greater levels of human capital they become more valuable to firms because they produce more than other workers.  Thus, in order to maximize profitability, firms should augment the standard MPL = W equation, replacing it instead with MPL + G = W + C, where G is the return to the firm from training and C is the opportunity cost and outlays from the worker in order to become trained.

            While Becker’s contribution to human capital opened the door to better understanding some of the factors underlying disparate wages that workers in the real world experience, it seems to fall short in its focus on adulthood as the primary time when human capital is developed.  Evidence shows that by the time individuals reach adulthood and are able to take personal responsibility for their choices, human capital may be largely immovable.  In 2012, Paul Tough published the book How Children Succeed which reviewed a wide array of research on the development of skills in young children.  One finding he illustrated was the extent to which traumatic experiences in early childhood can have long-lasting impacts on brain chemistry.  Specifically, the book cited a study by McGill University Neuroscientist, Michael Meaney, which monitored rat pups and showed that parental responsiveness to trauma, or lack thereof, altered brain chemistry (Tough 2012).  Similarly, children who are raised in stressful environments, and are not appropriately attended to by parents, have higher levels of cortisol which significantly inhibits their capacity to focus, a skill essential for developing human capital. 

While firms may be able to maximize productivity by investing in human capital at levels consistent with Becker’s calculations, the larger question seems to be whether or not society is properly endowing future generations with the basic set of skills they need to develop more complex human capital once they reach the workplace.  If outcomes are largely predetermined by the environment in which children are raised, human capital is no longer a rational choice within their control, and even if Clark’s MPL = W equation can be saved, it could scarcely be considered equitable.  

Conclusion

Over an illustrious career, Gary Becker contributed to a wide variety of topics often expanding the borders of what would have previously been considered appropriate work for an economist.  Some of his most influential publications applied the economic principle of rational individual choice to topics previously dominated by sociologists and psychologists.  A missionary for the economic faith, Becker colonized the fields of discrimination, crime, and family choice, converting many to the belief that behavior is primarily motivated by a rational weighing of costs and benefits.  While the jury is still out on this theory’s validity, the social sciences have certainly become infatuated with the concept, and hundreds of economists have migrated to the academic colonies founded by Becker.  Time will tell whether they have settled in Jamestown or Plymouth.

 

References

  1. Becker, Gary S. and Barro, Robert J. “Altruism and the Economic Theory of Fertility.” Population and Development Review, 1986, 12(Supplement), pp. 69-76.
  2. Becker, Gary S. “Crime and Punishment: An Economic Approach.” Journal of Political Economy, 1968, 76(2), pp. 169-217.
  3. Becker, Gary S. The Economics of Discrimination. University of Chicago Press, 1957.
  4. Becker, Gary S. “Investment in Human Capital: A Theoretical Analysis.” Journal of Political Economy, 1962, 70(5), pp. 9-49.
  5. Becker, Gary S. Editor Tore Frängsmyr. Les Prix Nobel. The Nobel Prizes 1992. Stockholm, 1993.
  6. Becker, Gary S. “Nobel Lecture: The Economic Way of Looking at Behavior.” Journal of Political Economy, 1993, 101(3), pp. 385-409.
  7. Becker, Gary S. A Treatise on the Family. Cambridge, Mass.: Harvard University Press, 1981.
  8. Levitt, Steven D. and Venkatesh, Sudhir Alladi. “An Economic Analysis of A Drug Selling Gang’s Finances.” Quarterly Journal of Economics, 2000, pp. 755-789.
  9. Kruse, Kevin M. White Flight: Atlanta and the Making of Modern Conservatism. Princeton University Press, 2005.
  10. Tough, Paul. How Children Succeed: Grit, Curiosity, and the Hiden Power of Character. Mariner Books, 2012.


[1] United States Census, 1930.  As retrieved from SocialExplorer.com on April 10, 2014.

25

Apr

Bending the Curve: Why CCRPI Misleads Educators and Parents

More than a year after students placed their pencils on the desk and closed their 2013 CRCT exam booklets, the Georgia Department of Education released its College and Career Ready Performance Index (CCRPI), which is intended to inform schools how they fare relative to others in the state.

While the measure is certainly an improvement on the rudimentary Adequate Yearly Progress benchmark developed under No Child Left Behind, it can hardly be called a success.  It continues to miss the most important element of effective evaluations: expected performance.

Few would be shocked to learn that students who live in wealthy neighborhoods and have educated parents on average score higher on the CRCT than students who live in poor neighborhoods and have uneducated parents.  Yet, the majority of CCRPI points are awarded with no consideration of expected achievement.  Therefore, the measure is intrinsically biased.

This is problematic on both ends of the spectrum.  Schools with low-poverty populations are uniformly receiving high scores while schools with high-poverty populations are consistently receiving low scores.  Both are missing out on useful feedback.  If the intention of the measure is to let school and district leaders know where they stand, achievement should be measured relative to peer schools. 

In the interactive graph below, I have taken the CCRPI scores of all elementary and middle schools in the state and plotted them against a percentile rank for free-and-reduced lunch population (FRL Percentile).  By clicking the “type” and “district” options on the left, users can highlight the performance of specific districts and also see how local and state charter schools perform. The top 20 districts (by size) are listed first, then the remainder are listed alphabetically.  Users can also hover over the circles to see info about the school.

The trend line represents the score one would expect a school to achieve based on its FRL population.  Schools above the line are overachievers while those below the line are underachievers.

(additional comments below the graph) 

The majority of districts in the state look a lot like Fulton County.  They are average performers, and most of their schools are relatively close to the line, with some above and some below.  In the schools with low poverty, CCRPI scores are high while the districts achieve low scores at their schools with higher levels of poverty.  In the metro area, Atlanta Public Schools, Cobb County, Forsyth County, Decatur City, and Henry County all fall in this category.  Fulton results are highlighted below.

Fulton

One district in the metro stands heads and tails above the rest, and that district is Gwinnett County.  As shown in the graph below, virtually all schools in the county outperform what would be expected based on their FRL rates.  Even at Corley Elementary School where 95% of students are low-income and almost all are Hispanic/Black achievement is stellar.  Gwinnett County is “bending the curve,” a strong signal that something operationally is working right.  We should be learning from the district and highlighting their success, but because the data distributed by the state fails to adequately account for expected performance, that message is lost in the headlines.

Gwinnett

Two districts in the metro are solidly underachievers.  Both Dekalb County and Cherokee county see most of their schools fall below the line.  Again, this message is muddled in the CCRPI releases.  Dekalb is recognized as a low achiever because it has high poverty, but Cherokee is thought to do fine.  It’s only when one benchmarks the county’s schools to others with similar FRL rates that the message becomes clear: Cherokee has some serious work to do.

Cherokee

In order to provide meaningful information to districts, principals, and parents, it is time for the state to again adjust its CCRPI measure, adequately accounting for expected performance given observable characteristics of the student population served.  Without these adjustments, the measure will continue to mislead the public, fail to highlight meaningful success stories, and miss out on the improvement that could be achieved by giving school leaders realistic feedback.

25

Jun

CRCT 2013 - APS Math Scores Rise

This morning, the Georgia Department of Education (DOE) issued its first release of CRCT data from the 2013 testing cycle.  The data released today includes system-level scores, and the DOE states that school-level information will be made available on or before July 10.

According to a press release accompanying the data, scores were up across the state in most grades and content areas.

Since the difficulty of exams can vary year-to-year, I am typically less interested in the raw scores and more interested in how districts perform relative to each other.  My initial review of the data has some positive news for APS: scores for the system’s students rose relative to the rest of the state across all grades.  The greatest gains came in math, while english language arts (ELA) scores declined slightly.

Before jumping into the numbers, it is important to remember that averages give us a general sense of the direction districts are moving.  The don’t tell us much about the experiences of specific students or the performance of specific schools.  APS has historically had some schools ranked in the top 1% and others in the bottom 1%, a pattern which is likely to repeat when school-level information is released in the coming weeks.    

Lets start by looking at last year’s performance so that we have a good benchmark to compare to.  As shown in the chart below, APS students in Grades 3-8 ranked in the 19th percentile statewide last year.  This means that 19% of the state’s students attended school districts which scored lower than APS last year.  The remaining 81% of students in the state attended districts with average scores equal to or higher than APS.

2012 APS Percentile CRCT Scores

This year, APS saw modest gains across all grades, rising to the 23rd percentile.  While this performance is still well below average, the district is moving in a positive direction.  

Looking at some of the subject-level information, it is evident that the district’s most significant improvements came in Math, while ELA performance fell slightly, and other subjects rose slightly or remained flat.  

The percentile scores for each subject for the years 2012 and 2013 are included in the charts below.

2012:

2013:

20

May

Meeting With Supt. Erroll Davis

Last Wednesday, APS Superintendent Erroll Davis met with about 30 community members from around the city to discuss the FY2014 budget.

I thought that Davis did a good job of responding to the questions that were asked, and I appreciate him taking the time to sit down with community members.

A couple of the impressions that I left with were:

  • Davis feels that the district faces a lot of administrative challenges which take money to address, and he feels it is more important to direct money toward those initiatives than to keep class size low.
  • From Davis’s perspective, the central office and administrative spending are reasonable (or even favorable) compared to other urban districts.
  • From Davis’s perspective, a primary driver of administrative inefficiencies is having under-enrolled schools. 
  • There is a disconnect between the information the administration is relying on to make decisions (from the computer systems) and the practical experiences of those closer to the schools.
  • From Davis’s perspective, a slow pace of change/improvement is to be expected from a large organization, and he feels like they are moving in the right direction slowly.

I continue to think there are several ways that I see things differently from Supt. Davis, and I don’t think he is closer to coming around to my point of view or that information has been presented which would help bring me around to his point of view.  The things I continue to think are:

  • Even setting aside assistant principal classification, under-enrolled schools, and the pension liability issues which have been mentioned, the district spends significantly more money on the central office than other districts in the state, whether those districts have higher-needs students or lower-needs students.
  • Some of this is driven by inefficiencies in the hierarchy of the central administration.  A couple of examples include:
  • Karen Waldon’s department has its own budget team rather than relying on the finance department.
  • The reorganization of the curriculum and instruction department added lots of new positions.  Under Karen, there are 4 associate superintendents, 4 executive directors, and 4 coordinators, all of whom are highly compensated.  Yet, what I’ve heard from the experience of principals is that they feel C&I has become less responsive during Karen’s time at the helm.  I have heard reaching out to C&I described as sending messages into a black hole.  While this is certainly not reflective of all principal’s take, I have heard it now from multiple people in four different clusters.
  • The 4 leaders of the reform districts are highly paid (and most say they’re talented people), but they are not vested with authority, and therefore they do not contribute meaningfully.
  • While I agree that large organizations are hard to change, we can’t be resigned to slow improvements.  We may need to look closer at the possibility of moving faster to a decentralized model where clusters or quadrants of the city are responsible for decision making, and change can be implemented at a faster pace.
  • Even in the depths of a recession, we have significantly more money per student than other Georgia districts ever have in a booming economy.  It is challenging for me to reconcile that with a need for class sizes of 33.  It seems to me that if we were more strategic with our use of money and planned better, we could have smaller classes and still have funds left over to fund investments in the administration which Davis sees as needed.
  • The data Davis relies on to understand the finances of the organization is compiled with a lot of assumptions, which it seems to me make the numbers reflect something different than he (or any other user) would initially assume based on the summary descriptions of the expenses.
  • We have to plan holistically.  It is perfectly appropriate for certain funds to have restricted uses, but that shouldn’t be used as a reason for us not to plan ahead for how we will use those funds.  I’m not aware of any regulations which would prevent planning the use of Title I funds or other special funds.  I also don’t think the fact that discretion is provided to the principals should prevent this process from taking place.  The principals should be asked to work with the administration to put together a plan for next year as part of the budget cycle.

13

May

Learn about the Atlanta Public Schools budget!