Research & Insights

Do low-income students have the same access to effective teachers as their more affluent peers? A new study from Eric Isenberg and his colleagues at Mathematica Policy Research and the Brookings Institution examines this question and draws a surprising conclusion.

In 26 urban districts, the researchers set out to measure the “effective teaching gap,” that is, the discrepancy in students’ access to effective teachers depending on their socioeconomic status (SES). Teacher effectiveness was determined entirely by value-added scores.

Applying this framework, if all students had equal access to effective teachers, the effective teaching gap would be zero. The reality wasn’t that far off the mark and in contrast with the findings from a fair amount of other research. Many (though not all) of the districts turned out to be providing all students, no matter what their SES, roughly the same access to effective teachers.

To the districts’ credit, these results reflected a good deal of work on their part. On average, districts in the study had implemented around five common strategies aimed at promoting the equitable distribution of teachers across schools. They include comprehensive teacher induction, highly selective alternative routes to teaching, targeted use of bonuses, performance pay, and early hiring timelines in high-need schools.

What makes this study so different from others that have pursued similar questions and gotten different results? For one, there were clear methodological differences, particularly when it comes to calculating a teacher’s value-added score. Isenberg’s study controls for peer effects (a measure of how the characteristics of a class as a whole might affect the performance of any one student); however, many other studies do not incorporate peer effects in value-added measures. In addition, this study focused solely on the distribution of teachers within each district, not between them. It did not attempt to answer the question as to an effective teaching gap between urban districts and the suburban districts that border them.

For a detailed discussion of the conflicting literature related to the effective teaching gap, we recommend this CALDER brief, which describes the debate around peer effects, differences in research conducted across vs. within districts, and other methodological points.

Some people ask if the quality of teacher preparation matters at all.

Teacher quality researchers made plenty of provocative headlines in 2016. They identified trends to monitor, new tips for the trade, and a few wins worth celebrating. Here are the papers we think are the 2016 standouts.

1. Great teachers beget more great teachers
Papay, J., Taylor, E., Tyler, J., & Laski, M. (2016). Learning job skills from colleagues at work: Evidence from a field experiment using teacher performance data.

In one of our favorite experiments of the year from researchers at Brown University and Harvard Graduate School of Education, researchers helped schools identify a teacher who was struggling in a particular area and matched that teacher with someone who excelled in that particular area. Then, they left them to their own devices. No training. No oversight. The paired teachers found a way to work with one another to address deficiencies and grow professionally. In the end, gains for the low-performing teacher were large and persistent from year to year.

2 and 3. And the theme continues: When it comes to professional development, small may be better
Jackson, K. & Makarin, A. (2016). Simplifying teaching: A field experiment with online “off-the-shelf” lessons.

Okonofau, J.A., Paunesku, D., & Walton, G.M. (2016). Brief intervention to encourage empathetic discipline cuts suspension rates in half among adolescents. 

In a similar vein to the simple paired teaching strategy are two other papers, each helping us to better understand how professional development could be improved. They each involved inexpensive, light touch interventions that led to great gains. In the first, from the prolific Kirabo Jackson with his Northwestern colleague Alexey Makarin, teacher performance dramatically improved after they were given access to a library of high-quality, low-cost lesson plans in mathematics, as well as a few emails to remind them to use them. Teachers who also received access but no reminder emails did not use the plans and their performance did not improve. In the second paper, from Stanford researchers, suspension rates plummeted in classrooms taught by teachers who had participated in only a 45-minute online session, in which teachers were prompted to thinking about how to build more positive student-teacher relationships.

Both studies serve as a reminder that when big change is needed, every small step counts.

4. Districts must find better ways to address teachers’ unintended racial bias
Grisson, J.A., & Redding, C. (2016). Discretion and disproportionality: Explaining the underrepresentation of high-achieving students of color in gifted programs.

2016 began with a sobering finding: even when black students have the same high test scores as white students, they are much less likely to be enrolled in a gifted education program. The main culprit is the identification process, relying heavily on teacher recommendations. As we explore in this paper, jointly authored with Brookings researcher Michael Hansen, districts will not be able to hire their way out of this problem, recruiting more teachers of color. The solutions must also include better training of faculty and safeguards to help teachers recognize and overcome their biases.

5. Groundbreaking policy and smart talent management continue to make the District of Columbia a district to emulate
Adnot, M., Dee, T., Katz, V., & Wyckoff, J. (2016). Teacher turnover, teacher quality, and student achievement in DCPS.

Over the last decade, District of Columbia Public Schools (DCPS) has charted new territory in teacher recruitment, retention, and management–and they have faced plenty of pushback and skepticism along the way. But this year the Academy has weighed in with a paper from University of Virginia researchers, surfacing proof that one of their biggest bets is paying off. By letting low-performers go, doubling down on retention efforts, and getting smart about recruitment, DCPS has increased the quality of their teacher workforce substantially. The result is real growth in student learning, surpassing what’s been measured in any other urban district.

6. On the need to get more intentional about student teaching
Goldhaber, D., Krieg, J.M., & Theobald, R. (2016). Does the match matter? Exploring whether student teaching experiences affect teacher effectiveness and attrition.

Our top list wouldn’t be complete without a paper from Dan Goldhaber and company. This year’s top Goldhaber paper looks at student teaching, a much unstudied topic that the researcher and NCTQ both agree deserves a lot more attention. This Washington state study finds that teachers are more effective if they have completed their student teaching in a school that was demographically similar to the school where they would ultimately work. Not only that, but teachers are also more likely to remain in the profession if they student taught in a school that had low teacher turnover. Two important insights for teacher prep programs to ponder.

Purpose and Changes in the Teacher Prep Review: Undergraduate Elementary

Our qualifications and why we do this

Today, NCTQ is excited to share the latest results in our ongoing analysis of teacher prep programs, this time focusing on what we’ve learned from 875 undergraduate elementary programs.This is our first set of grades since 2014, and our research discovered that programs have made some gains. We’re far from out of the woods, but at least we can see a bit of light through the trees.

The top undergraduate elementary programs in 2016 (all scoring in the 99th percentile) are:
Purdue University (IN)
Louisiana Tech University (LA)
Texas A&M University (TX)
Taylor University (IN)
University of Alaska Fairbanks (AK)
CUNY – Hunter College (NY)
University of Houston (TX)
Arizona State University (AZ)
University of Arkansas (AR)
University of Mississippi (MS)
University of Nebraska – Lincoln (NE)

Unlike previous Reviews which grouped together all different types of teacher prep programs at once–publishing data on over 2,400 programs!–this report is smaller in scope, looking only at undergraduate programs preparing elementary school teachers. (We’ll cover the other kinds of programs in future reports, each spread out by about six months, with the next one on undergraduate secondary due out in Spring 2017.) For more information on changes, see our Profile, one in a series of short blogs about different aspects of teacher prep and the TPR.

We are pleased to report some genuine overall progress by programs on the evidence-based criteria we examine. The big news is that more programs are adhering to evidence when teaching elementary teachers how to teach reading. In 2016, 39 percent of programs teach evidence-based approaches to early reading, up from 29 percent in 2014.

While most programs are still not selective enough (with only 26 percent limiting selection to the top half of college-goers), there was some progress. More programs that are housed in institutions lacking strong admissions requirements have stepped up, setting their own relatively high admissions standards (at least a 3.0 GPA for admission)–up from 44 in 2014 to 71 today.

Importantly, half of these selective programs are also relatively diverse when compared to the institution as a whole or to the state’s teacher workforce. These 113 programs demonstrate that teacher prep programs can be both diverse and selective.

Unfortunately, in light of the recent PISA results, the news on mathematics preparation is gloomy. Just 13 percent of programs cover the essential math that other nations expect their elementary teachers to have mastered.

The findings are even worse on content preparation–which is so important for states that have adopted the Common Core or similar standards–with just 5 percent of programs requiring aspiring teachers to be exposed to the literature, history, geography, and science found in the elementary curriculum.

For more information see our Landscape report and our Teacher Prep Review website.

This report not only will help principals and human resource officers looking for where to find teachers with excellent training, high school guidance counselors will also find it helpful in advising students interested in becoming teachers to the best programs. It will help guide teacher prep programs in their own efforts to do better, and we hope it will inspire university officials to upgrade the quality of their teacher prep programs and even lead states to reconsider the oversight they provide.

We urge our readers to share this report with university officials and teacher prep leaders, encouraging them to use this as a blueprint for change. Districts also can use their leverage by recruiting first at top rated programs, giving their graduates first pick of jobs, and insisting that the programs make reforms if they want to continue to send their student teachers to the district. State officials can recommend that programs voluntarily adopt more research-proven methods and content or risk more mandates from above.

Thank you for joining us in our efforts to raise the quality of teacher prep programs so that all children can be taught by more effective teachers.

If there’s one thing that research has shown us time and again, it’s that being a brand new teacher is hard—and being one of their first students is not all that ideal either. But what happens when an experienced teacher switches to a new grade level or subject? Does she start all over again? This shift, called “churn,” might not sound like a big deal, but as with any change in professional responsibilities, it does come with a learning curve and some predictable consequences.
In a new study looking at data from New York City over a 35-year period, Allison Attebury (UC Boulder), Susanna Loeb (Stanford), and Jim Wyckoff (UVA) find that a student taught by a churned teacher faces a moderate disadvantage. But because 25 percent of the city’s teachers churn to a new grade, subject, or both within their school every year, that moderate difference can really add up to a significant learning loss over several years.
Little Johnny encounters 4-5 churned teachers in grades 3-8, but just one brand new teacher.

As we explored in last month’s TQB, it seems plausible that there could be an upside to shifting teachers to new grades within a school. Maybe Mrs. Jones does better with older students than younger, and so a switch in this instance could help her find a better teaching fit. But unfortunately, this study did not support that hypothesis as a guiding principle. Instead there’s no clear evidence that churned teachers end up performing much better in new positions.
Nevertheless, if we’re not speaking about within-school churn but transferring schools, it may be another story. Susanna Loeb, one of the study’s co-authors, recently published another study with Min Sun (University of Washington) and Jason Grissom (Vanderbilt) that does identify some benefit to moving teachers around. Experienced and relatively effective math teachers who transfer from one school to another create a “spillover” effect—meaning math scores go up for the students taught by the other teachers in that grade level. This finding suggests that there are some ways in which administrators can rethink teaching assignments to improve student outcomes—as long as these shifts are made sparingly and strategically.

Every consumer knows how the law of supply and demand affects prices. When demand is higher than supply, the price goes up. As suppliers increase production to match the demand, the price goes down.
I’ve frequently lamented how school districts try to exempt themselves from this law. When districts are reluctant to raise the salary to teachers in fields with more demand than supply, such as ELL, special education, and secondary math and science, it is no surprise when they fail to find qualified candidates with these much-needed skills.
Teacher unions and other opponents of salary differentiation have to exaggerate the breadth of this teacher shortage into claims of a general teacher shortage so they can then justify demanding general policy changes. Of course, raising the salary for all teachers, even those in fields with plenty of applicants, is an expensive way of solving a limited problem.
Still, even when districts ignore supply and demand, others don’t. College students, for example. News about teacher hiring and firing does influence undergraduates when they’re trying to decide on a major. A few years ago, during the Great Recession, schools dismissed 220,000 teachers–most of them relatively new hires. People with jobs held onto them tightly, reducing the replacement rate. It’s no surprise that college students decided against a teaching career, resulting in a drop in enrollment in teacher education of 36 percent between 2009-10 and 2013-14.
Today’s college students are hearing a very different story of teacher shortages and plentiful hiring. There’s now some new evidence that this is affecting their choices exactly the way the law of supply and demand would predict.
For instance, the California Commission on Teacher Credentialing found that state enrollment in teacher education increased nearly 10 percent between 2013-2014 and 2014-2015. In Indiana, the number of newly licensed teachers increased 18 percent between 2014-2015 and 2015-2016. It turns out that publicity around teacher shortages actually makes them less likely to occur.
The bad news is that we still won’t do what it takes to address those persistent shortages of teachers who can fill certain subject areas or work in a rural or high urban school. That’s because both higher ed institutions and school districts continue to break the law of supply and demand and should be cause for targeted recruitment strategies, not general alarm.
For instance, higher education programs continue to prepare about twice as many new elementary teachers as are needed to fill openings. They should steer some of these candidates to fields like special education where supply is lower than demand.
There’s no need to lower standards at teacher preparation programs to bring in more teachers. Leave it up to the law of supply and demand to rectify that problem. Instead, states, higher education, and districts should solve the actual, more limited problem of shortages in specific fields by directing candidates away from over-enrolled elementary programs and increasing compensation in specialized areas.
As civics teachers tell their students, everyone is supposed to follow the law. 

According to the National
Center for Education Statistics, in the 2011-2012 school year, average
class size across all public schools in the United States was 21.6 students in
elementary school, 25.5 students in middle school, and 24.2 students in high
school. This month, Trendline looks at class size restrictions and what
happens when class size limits are exceeded.

Passing a good law and implementing it with fidelity are two very different tasks.
Earlier this summer, Indiana University researchers took a
thorough look at that state’s ground game on teacher evaluations—and identified quite a few missed opportunities.

Like many states, Indiana passed a law that ushered in evaluation reform (the implementation timeline is
here), being one of the first states out of the gate. Most notably, the law required districts to use student
achievement to evaluate teachers starting in 2014. Ultimately, each district
was permitted to develop its own evaluation system—provided it met certain
criteria established by the state.

That
provision resulted in 271 evaluation plans which are all over the map in terms
of quality. (There are roughly
295 districts in the state, but not all of
them submitted evaluation plans that were reviewable by the researchers.) The
majority fail to include key components that the researchers assert would be
elements of any high-quality evaluation—and indeed it’s hard to argue with
their logic. For example, only 15 percent of districts link a teacher’s
evaluation with their professional development, and less than 25 percent of
districts require evaluators to hold pre- and post-observation conferences with
teachers, giving them the feedback they need and deserve.

In
addition, many districts struggled to articulate policies that would have built
teacher trust in the evaluation process. Fewer than half of the districts
require evaluator training and certification, and fewer than a third of the
districts convene an evaluation oversight committee. Of those districts that
have an oversight committee, few districts have put a process in place for the
leadership team to meet regularly and resolve ongoing issues.

Change
is hard. If we want to make
evaluation reform stick, districts need to not only lay
the foundations of a fair and reliable system, but also include practices and
policies that make evaluations meaningful for, and trusted by, teachers.

It’s not surprising that teacher educators feel defensive these
days. They frequently face accusations for their allegedly inferior quality
programs. Non-traditional training programs assert they can produce better
teachers in a few months than teacher preparation programs can in two years. Looking
at the existing research will not provide evidence to the contrary. Further, surveys
of teachers routinely reveal complaints of inadequate preparation. States have
made a lot of progress over the last five years ramping up accountability
measures, with 44 states making significant policy changes designed to make it
harder to get into a teacher preparation program or get out of one. (We’re not
sure if these new regs will deliver however.)
Our own Teacher Prep Review certainly hasn’t helped improve
teacher prep’s reputation. Harsh, it certainly is. A plot to destroy
traditional teacher prep, it is not.
Teacher preparation’s latest “critic” is the federal
government with its recently released Teacher Preparation
Regulations.
Though I’m not sure anyone will believe me, I do sympathize.
Nobody wants to be told they’re not doing a satisfactory job or that their
efforts are worthless. This is especially demoralizing for teacher educators
who didn’t join the profession for its money or prestige, but to build a better
future by helping the teachers who educate our nation’s children. Also, there
are some high-quality programs and professors who justifiably resent blanket
criticism of their entire field.
Still, there’s a right way and a wrong way to respond to
criticism. For example, when The New York Times published an editorial backing the new federal regulations as a way of helping teachers
and emulating first-in-the-world nations like Finland,letters to the editor attacked the paper for assuming that teacher education was
“mediocre and underperforming,” while criticizing the research methods in
NCTQ’s Teacher Prep Review.
Responses like this simply further establish the belief among many
in the general public that teacher education needs to be replaced since the
people who run it won’t even acknowledge the problems, let alone take action to
fix them.
Claiming that NCTQ’s study isn’t valid, for example, because most
programs chose not to cooperate (as one letter writer asserted) ignores the
extent of our massive data collection efforts—including originally being forced
to go to court in nine states to win the right to look at course syllabi,
paying out $250,000 in open records fees, and dedicating teams to reach out to
professors and students for materials absent the cooperation of programs. It
also challenges any outside watchdog review. Would anyone claim that Consumers’ Reports ratings are invalid
because the independent group does not involve the manufacturers? One cannot
simply claim non-cooperation as a delegitimizing factor—essentially handing
institutions a heckler’s veto.
And of course, no one connected to higher education should try the
argument from anecdote fallacy—but again that is the most common defense.
Asserting, “That’s not true at my school,” is not only subjective, but it also
attempts to argue from a single data point. If everyone else is wrong and the
teaching field really does not need improving, than there should be no problem
collecting a plethora of valid evidence in support of such a view. Yet that
evidence, while often promised, has never been delivered.
In any case, we do not want our ratings to be used to bash teacher
preparation programs. Instead we want our ratings to help programs improve. We
want teacher preparation programs to incorporate research-proven methods so
their graduates can be more effective in the classroom from their first day.
I sympathize with teacher educators who have
devoted their lives to this low paying, frequently attacked profession. But in the
current defensive posture, real problems remain neglected.
Teacher educators and their critics both want the
same thing—a better education for tomorrow’s teachers.

If you look closely, they’re actually not hard
to find: inexpensive professional development opportunities for teachers that
actually do make a real difference in student learning. Earlier this year, we
covered an experiment that demonstrated that teachers are remarkably more
empathic towards their students after completing just a brief 70-minute exercise. We also learned how pairing a highly skilled teacher with one
who is struggling, even absent a formal curriculum, turns out to be more
effective than a lot of higher priced PD models.

Now, we’re offered new evidence that spending
very little money giving teachers access to high-quality math lessons can yield
big pay offs.

In a white paper from the National Bureau of
Economic Research, Kirabo Jackson and Alexey Makarin describe a nice little
experiment in which 119 math teachers were given access to a library of
inquiry-based lesson plans that currently runs at a cost of $320 per teacher
for the year. Teachers also received a few occasional email reminders about the
availability of the resources and were reminded that they could collaborate on Edmodo with other teachers
also using the lessons. That was it.

The results were remarkable. By the end of the
year, the growth in student achievement among these teachers was on par with
far more costly efforts, such as reducing class size by 15 percent or replacing
an average-quality teacher with a great one.

A couple of interesting points: 

First, it was the lowest performing teachers who
saw the greatest gains.

Second, there were some teachers in the
experiment who didn’t get the email reminders and, with that group, there were
no gains at all.

There are also some important caveats to this small study, including
a significant decrease in lesson downloads and usage over time, suggesting that
teachers did not themselves appreciate their value enough to implement more of
them.

You can check out the lessons used in the study on the Mathalicious website. 

For a long time, we’ve
heard about the damage done by teacher turnover. Often, the thinking is that
schools struggle to replace the teachers who leave with replacements who
perform at least as well or better.

And while it is true
that high turnover can really hurt a school, it turns out that the damage is
largely caused by the ripple effect of a teacher leaving, mainly the amount of
grade switching that occurs. A new working paper
from Eric Hanushek, Steven Rivkin, and Jeffrey Schiman measures the broader
impact of teacher turnover in a large school district in Texas.

When a teacher leaves a
school, the remaining teachers often have the opportunity to move around,
selecting what might be considered ‘better’ grades and leaving the more
‘difficult’ grades to be taught by newer teachers. Not only do the new teachers
have to deal with the more difficult assignment, but the teachers who switch
grades, as some other recent research
has found, are also less effective, at least in their first year teaching the
new grade.

Not as new but always
worth restating:  Hanushek et al. again
find that the weaker teachers were more likely to leave their schools than
strong teachers, which means that unfocused efforts to reduce turnover are not
necessarily in the best interest of schools. 

At long last the US
Department of Education released its teacher
prep regulations this week, prompting me to check my
calendar. I had to look up the date of my first meeting with Department
officials on the subject. Back then, I’d been willing to bet anyone that the
Department would never finish its regulations – but nearly seven years later,
this is a bet I’ve finally lost.

Were the regs worth the
wait? On balance, yes–not just because they represent a big improvement over
current Title II reporting requirements, but also because they reinforce the
work many of the rest of us are doing on teacher prep.

What is most appealing
about these regulations is more data and more transparency. States will need to
annually survey principals and first-year teachers on the quality of the novice
teachers’ preparation programs. While many programs already engage in this
practice, this requirement essentially requires a common survey, allowing for the first time real comparison among
programs.

There will also be a
slew of data generated about teacher supply, employment, and retention rates,
perhaps putting to an end the current reliance on conjecture and anecdotes to
predict when and where there’s about to be a teacher shortage.

And in what may come as
a surprise to some, I don’t disagree with the decision to omit a requirement
that states examine the test scores of students taught by prep programs’
graduates.  While the Department may have
just thrown up its hands at the amount of resistance to test scores, the use of
value-added measures to assess program quality is in fact fraught with
methodological difficulties, especially for smaller prep programs. To get
enough data points to reach a sound judgment of program quality, it’s often
necessary to collect teachers’ performance data for five, or even more, years after
graduation. That hardly seems fair to programs. The fact is that value-added
measures only produce meaningful results for the few programs turning out big
numbers of graduates who go on to teach tested subjects each year, and in some
cases, the programs whose graduates’ performance is a clear outlier. 

A great substitute for
test score data could be either candidates’ pass rates on licensure tests,
including the percentage of candidates passing these tests on their first
attempt, as well as surveys of the students in teachers’ first classrooms. Pass
rates on licensing tests might have been something the Department insisted
upon–though it’s been down that road before to no avail both in the 1996 and
2008 HEA reauthorizations.

Importantly, these regs arrive at a singularly
opportune moment–when the winds of change are blowing from every direction,
including from within. It’s certainly not just NCTQ raising the ruckus. The
last five years has elicited unprecedented activity, with no fewer than 44
states passing significant teacher prep regulations. And what may be the
biggest disruptor of all is the 30 percent drop in enrollment in teacher prep
programs–for reasons that are anyone’s guess, but which surely include the
poor reputation of teacher prep. As anyone knows who has managed a budget,
institutions are more likely to consider making changes when confronted with
fiscal pressures.

The only aspect of the
regulations that is absolutely without merit is the Department’s decision to
drop its requirement, present in previous drafts, that programs must raise
their admissions standards. It’s dropped in the final version because
institutions made a lot of noise about the impact that raising standards will
have on diversity. Not only is this common complaint denigrating to African
American and Hispanic students—implying that a teaching career is only
available to them if standards are kept intolerably low—but the consequence of
an open door policy sounds a death knell for programs’ ability to raise the rigor
and quality of instruction. It perpetuates the low status of the education
major on college campuses. The Department defends its decision by stating that
its regulations set a high standard for program exit, but the details seem to
imply that a candidate only need to pass the edTPA or a similar assessment
which, in some states, is reporting about a 98 percent pass rate. 

It’s not that the final version of these regulations
doesn’t bear the marks of heavy compromise; but on balance, the federal
regulations are generally sensible and respectful of the parameters of federal
authority – and provide a much-needed opportunity to illuminate how prep
programs’ graduates fare in the classroom.

I can safely say that this is one bet I’m happy to
lose. 

This month, we take a look at how much leave time teachers receive each year and what policies school districts use to incentivize higher teacher attendance, from allowing sick and personal leave to carryover from year to year to paying teachers for unused leave when they retire.

Licensure tests:
depending on who you ask, they’re either an important check on a prospective
teacher’s knowledge before entering the profession or a burdensome requirement
that keeps good candidates out of the classroom.

As we discussed
earlier this month, holding teacher prep programs accountable for the
performance of their graduates is no easy task. The data is often scant and
researchers usually can’t distinguish any standouts in a sea of mediocre or weak
programs. 

That’s why we are
pretty enthusiastic about a new study
from Matthew Ronfeldt and Shanyce Campbell of the University of Michigan. Previous
studies looked only to one data source—graduates’ value-added scores—to determine the
strength of program graduates. These
two researchers use multiple measures involving, first, teacher observation
scores and, second, value-added scores. They unearth clear evidence that others
have not: not all programs are created equal.

In the sample of 118
programs, 21 surface for graduating teachers who consistently have either
higher observation scores than most other programs, or, conversely, consistently
lower scores. 

The waters do get
muddied a bit when folding back in the value-added measures. Not surprisingly,
programs that did really well or really badly on observation scores didn’t
always have similar results on value-added. In fact only about 40 percent of
the programs produced observation and value added scores that were similarly
positive or negative.

Nevertheless, if a
policymaker were to assess program quality by looking only at the overlapping
data, it seems safe to conclude that there are programs clearly succeeding or
failing–producing teachers who consistently get both great evaluations and great test score results or the
reverse. 

When all was said and done,
there were #25 standout programs in the state, but as is the frustrating custom
of academic research, these programs were not identified.  

These promising results
reinforce our interest in multiple measures for evaluating program quality. One
such additional measure could be provided by TPI-US, essentially a
comprehensive on-site inspection process imported from the United Kingdom. In its
assessment process, teams of four trained education professionals visit prep
programs to collect evidence on program quality as well as to provide actionable
feedback. They observe student teachers and course instructors, examine data on
candidate performance, and conduct interviews with key stakeholders, including
graduates and leaders at the schools that hire them—all of which could serve as
yet another source of data on a program’s quality. 

Here’s something I’ve been struggling to understand of late. What makes the prospect of a teacher shortage such an immediately compelling narrative, capable of spreading with all the speed of a brush fire?
With almost no real data–because neither states nor the federal government collect the data that’s really needed to pronounce the onset of a teacher shortage–we witness the press, school districts, state school boards, and even the US Congress all concluding we are in the throes of a full blown national crisis.  
At the root of this crisis was a random New York Times news story published two summers ago in which eight school districts reported in August that they were having a tough time filling positions (though all but two ultimately started the year just fine). Whoosh! Overnight the teacher shortage became real.  
That early rumbling was then steadily fed by news stories that teacher preparation programs were facing unprecedented enrollment drops. Whoosh! Nobody thought it important to mention that teacher preparation programs had for years been graduating twice as many teachers as are needed.  
Get this. Over the last 30 years, programs have graduated between 175,000 and 300,000 teachers each year, yet consistently school districts have only hired somewhere between 60,000 to 140,000, with about 95,000 being the most recent number. 
The blaze reached new heights last week with a new report from Learning Policy Institute, producing a scary chart that shows an ever widening gap between teacher supply and demand over the next nine years. While I would not characterize LPI’s supply and demand projections as irresponsible or even without some merit, they were predicated on glass-half-empty assumptions that economist Dan Goldhaber rightfully questioned in this 74 editorial. (It’s also worth mentioning Mike Antonucci‘s reminder of the report’s déjà-vu qualities given that LPI is led by Linda Darling-Hammond.)  
Let’s make this simple. Simply by tweaking just one of the assumptions made by LPI, the results are altogether different. For example, if we project that the class size average of student to teacher is 16.1 to 1 (which, importantly, it is currently) rather than LPI’s estimate of 15.3 to 1, voila! The shortage disappears entirely.
Anyone basing predictions on the available data needs to be transparent about the limitations and assumptions baked into the analysis-which I would argue LPI was not- and be clear that if anything changes (such as class size, or slower than expected student population growth, or a renewed interest in teaching following the end of the Great Recession), their predictions will shift dramatically.
The systems to report whether districts are facing a shortage exist in a small number of states (data is recent but not in real time) and not at the federal level at all. Contrast this to the reporting on the health of the American economy, which is done routinely with mountains of real-time data at the federal, state, and business levels. Even then, no one would declare a depression based on years-old data and rough hiring projections a decade from now. But somehow doing exactly this is ok for teacher shortages.
What I find so frustrating about all of this is that we do actually have a long standing, huge problem with teacher supply and demand – one that not only gets lost in the current rhetoric but that is, believe it or not, actually ill served by what could be a drummed up crisis.  
Let me explain this apparent contradiction in my logic.
For 30 years nearly every district in the nation has struggled to find enough secondary science and math teachers. Also and for just as long, rural and urban districts have been unable to tap into a reliable and stable source of new teachers, putting band aids like Teach For America on the problem.
One of the answers is to pay such teachers more than other teachers are paid, but most districts continue to reject that solution because it is untenable with their unions. For STEM teachers we could ramp up the availability of part-time teaching positions, but again few districts and states embrace this option–also because unions worry that districts will begin replacing full-time employees and their costly benefits with part-timers.
For even longer than those shortages have been so problematic, school districts have been awash with applicants for elementary teaching positions. That’s because teacher prep programs don’t see it as their job to tell their incoming candidates that they can’t all major in elementary ed, that they’ll need to consider another teaching field like special ed or ELL where there is real need.  The problem is that higher ed accepts no responsibility for aligning teacher production with district demand. Given that those teacher prep programs can’t operate without state approval, states could conceivably impose limits on production in some areas.

It’s
been 30 years since states first began experimenting with alternative
certification (AC) pathways for teachers, and while these routes have become
firmly entrenched in many districts’ talent strategies, the debate over their
value continues.

In
a recent study
in the American Educational Research
Journal, Christopher Redding and Thomas M. Smith contribute some new
evidence around two long contended points—namely, whether alternatively
certified teachers are prepared for the classroom and whether they’re likely to
stick around.

Preparedness. Previous research has established that there is no
clear answer to the question of whether teachers from AC routes are better
prepared or more effective than their traditionally certified peers. As our own review
of non-traditional teacher prep shows, there are some high-quality alternative
preparers of teachers, and there are just as many, if not more, ineffective
alternative preparation options.

Analyzing
data from the government’s Schools and Staffing Survey, Redding and Smith turn
up an interesting new finding on trends in AC. In the 1999-2000 school year, 23
percent of alternatively certified teachers entered the profession with no
practice teaching, compared to 8 percent of teachers entering from a
traditional prep program. By 2011-2012, the proportion of AC teachers with no
teaching experience had grown to 40 percent. Why? 

Retention. If you ask most people about the problems with
alternative routes, the number-one gripe is usually that AC teachers leave the
profession more quickly than traditional candidates. Redding and Smith show
that this is the case, but that it wasn’t
always so. In the 1999-2000 school year, there was little difference in the
retention rates between early career AC and traditionally prepared teachers. By
2007-2008, however, the predicted turnover rate for AC teachers was 10
percentage points higher than that of traditionally trained candidates, even
when controlling for school environment.  

Redding
and Smith’s work serves as another reminder that the quality of alternative
certification programs matters—which is something we’ve been saying for a long
time. Moreover, we’d do well to remember that most alternative certification
programs are expensive to districts, candidates, and communities alike. With
around a quarter of early career teachers now entering through AC pathways, the
need to measure the returns on this investment is greater than ever.

Is it too much to ask that professionals stay abreast of the research? The authors of a recently published study, ”
The Dyslexia Dilemma
,” don’t think it is and the extended
title of the study suggests the reasons why without mincing words: “A History of Ignorance, Complacency and Resistance in Colleges of Education.”
^[1] The
study highlights the fact that the Science of Reading instruction is neither
studied nor taught in teacher prep programs.

For 20 percent of
children, reading is the most complicated, difficult endeavor they will face
probably until adulthood. Often these children who struggle to learn to
read are labeled “dyslexic.” The term has been medicalized into a neurological
syndrome across the board. The authors of this study, led by David Hurford,
contend that these children are simply not being properly taught. The authors’
dissatisfaction with the failure of teacher preparation programs to teach the
science of reading to aspiring teachers almost rises to the level of outrage—as
well it should.

In spite of decades of research and
legislation going back to the 1980s’ A Nation at Risk, the 1990’s America’s
Schools Act, the Goals 2000: Educate America Act, the 2001 National Reading
Panel Report, the 2002 No Child Left Behind Act, and the Common Core State
Standards initiative, reading achievement in the United States remains
stagnant. By NAEP measures, reading achievement remains at 1992 levels. Well
over 50 percent of children in grades 4, 8, and 12 do not read at a proficient
level. Even the attempted “end run” of a couple decades of teaching to the test
has not caused the scores to budge.

Failure to learn to read has dire consequences
reaching beyond the school years well into adulthood. The psychosocial issues
related to dyslexia include low self-esteem, depression, post-traumatic stress,
substance abuse, incarceration, poverty, social dysfunction, and more. Failure
to learn to read proficiently also constitutes a national economic
liability. 

Nonetheless, a survey of hundreds of teachers
revealed serious gaps in teachers’ knowledge of basic scientific findings. It’s
especially lacking when it comes to their need to understand the structural
phonology of language and its relationship to learning to read. Only 20 percent
in a sample of over 700 teachers could segment words into speech sounds, for
example. The teachers surveyed reported that they had never received formal
instruction in phonological processing.

NCTQ has documented this lack of instruction in the
reading courses taken by teachers. So too has Kelly Butler from the Barksdale
Institute in Mississippi, Milt Joshi from Texas A&M, and several others.
Not surprisingly, phonemic awareness is the single most absent topic in reading
syllabi. So let’s be clear about reading failure and teacher accountability:
teachers cannot teach what they themselves
have not been taught.

In the same way that teachers cannot teach what they
have not been taught, neither can college instructors. The difference is that
college instructors have a responsibility to be on the cusp of research. Both
the ignorance and culpability are systemic in colleges of education.

In addition to documenting the pervasive weaknesses
in reading found in most Ed.D. and Ph.D. programs, Hurford et al point to the
persistence of myth in teacher preparation, the most insidious of all being the
idea that learning to read is a natural, innate process—the myth that gave rise
to the scientifically discredited and abject failure of the “Whole Language”
approach to reading instruction. 

Many children who
come to school ready to read are labeled dyslexic. The etiology of their
dyslexia notwithstanding, they can be taught to read. Hurford
et al close with: “Children with dyslexia and reading difficulties are waiting
to be taught to read and the knowledge and skills necessary to do so exist. It
is essential that the Science of Reading become part of the vocabulary,
knowledge base and training within colleges of education.” Children who are
neurologically dyslexic or just struggling to learn to read will continue to
suffer until the benefits of scientific findings gleaned over decades of
research with tens of thousands of children and adults make their way into
college classrooms.

Bob Marino leads
the NCTQ review of reading coursework as part of the Teacher Prep Review and is
a former principal in Baltimore City Public Schools.

If
you want to judge the quality of a teacher prep program, one approach is
obvious: examine whether the teachers who graduate from that program actually
help students learn.

Easier
said than done. Many factors affect whether a teacher will succeed in the
classroom, and determining how training factors into that success requires more
data and statistical power than is typically available. That’s one of the
reasons that academic researchers, states, accreditation agencies, and NCTQ’s Teacher Prep
Review assess prep programs
mainly by measuring the program features known to influence teacher
quality—such as admissions standards and what teacher candidates are being
taught about reading instruction.

A
new study
from Paul T. von Hippel (UT-Austin) and his colleagues is among the most
promising of the few studies that have sought to measure the direct impact of
individual prep programs on student learning. Thanks to cooperation with the
Texas Education Agency, he was able to examine outcomes across thousands of
teachers and hundreds of thousands of students in the nation’s second-largest
state.

Unfortunately,
even with this much data, they turned up little new evidence that one teacher
prep program is better than another.

Given
the impressive load of data that von Hippel et al. had at their disposal, this
conclusion raises some red flags. The data set is unprecedented in size and
includes a diverse set of traditional and alternative prep programs. Given this
variety, it is reasonable to expect to be able to pinpoint at least a few
obviously high or low performers. Instead, we see more evidence that many of
the challenges in teacher prep likely exist across the board. It’s a finding
not unlike NCTQ’s own much different scan of the landscape in which 80 percent
of all teacher prep programs earned scores classifying them as weak or failing. 

The
study authors use these results to warn policymakers against using student
outcomes data to make decisions about program expansion and closure, reasoning
that the programs were all too similar for anyone to be sure they were singling
out the right ones. We’re hoping Texas policymakers will start asking a new
question: Why are programs continuing to
operate without clear and specific standards for what teachers should know and
be able to do?