The Education Gadfly Weekly

In 1970, the celebrated economist Albert O. Hirschman published Exit, Voice, and Loyalty. A few years ago, the Hoover Institution’s Williamson Evers explained its argument on the Education Next blog:

Hirschman discusses how individuals react when services they rely on deteriorate. The basic responses available to us are “exit” and “voice,” Hirschman points out, where exit means turning to a different provider or leaving the area, and voice means political participation.

We tend to think of these responses as stark alternatives. Hirschman, as a social scientist, wanted us to consider the interplay between them. Exit usually has lower costs than voice for the individual. With exit, you can avoid the long slog of politics and simply turn to someone else or move somewhere else.

But there is a limiting case: Exit can have high costs when individuals are loyal to institutions—thus the third component in Hirschman’s trio of exit, voice, and loyalty.

I’ve been thinking about exit, voice, and loyalty lately, and how they pertain to parents of school-age children, myself included.

Those of us in education reform have generally viewed parents as either choosers or helpers—in terms of exit or loyalty. Under the former rubric: If kids are stuck in failing or mediocre or “bad-fit” schools, we parents should be able to vote with our feet, “exit” the school, and go someplace else. That is the great promise of school choice—it gives parents real power and thus makes the needs of our kids a priority while providing beneficial competitive effects for everyone else.

Under the latter rubric comes the role of parents as loyal helpers in the work of school improvement and educational betterment. This is important to many educators and some reformers and is surely a form of “loyalty.” There’s “parental engagement,” i.e. getting parents to excel as their children’s first teacher, for example by reading to them every night. There’s helping them with their homework, etc. And then there’s parents helping to improve schools by raising money, volunteering at events, pushing public officials to make education funding a priority, and more. “Loyal” parents that we are, we generally comply.

But what about giving more parents greater “voice”—and heeding it? Has anyone sincerely tried that as part of a reform strategy? Some of us talk about “parent power,” but that mostly refers back to school choice, i.e., exit. There are an increasing number of grass-roots organizing efforts in ed reform, but again most of them, like those led by Innovate Public Schools, are about pushing for systemic change or more school choice options.

What about parents who are committed to staying in our chosen school—typically the traditional public school in our neighborhood—but want to help it get better? Is empowering parents to have a real say in the affairs of their kids’ schools a promising strategy for school improvement? On issues big and small, from smarter approaches, to “differentiating instruction” so advanced kids aren’t bored and struggling kids aren’t left behind, to figuring out how to make recess less Lord of the Flies, to working on getting youngsters more exercise and healthier school meals, and on and on?

Don’t parents already have a voice?

I suspect many people assume that’s already happening, especially in affluent schools. Don’t pushy, prosperous parents pressure principals all the time to fix this or that about their schools? Isn’t that what PTAs are all about?

That’s certainly what I assumed before I become a pushy and (somewhat) prosperous parent myself. But, now that I am, I’m just not seeing it. And, as I’ve asked around and looked for evidence, I don’t see many examples out there. (Maybe Parents for Public Schools comes closest.) PTAs appear to be almost entirely in “helper” mode; structures to meaningfully engage parents in school decisions are almost always lacking. And when parents do advocate to make change it’s usually on behalf of their own children—to get additional services, or a different teacher, or some accommodation. When it comes to improving the school, though, we have a collective action problem, and a balance of power problem. (The schools have all the power, and the parents have none.)

I see this in my own kids’ neighborhood public school. Not once in my five years at the school have I filled out a survey or been asked my opinion about how the school might get better. As far as I know, there’s not even a suggestion box. It’s hard to send a clearer message than that.

None of this surprises Tim Daly, founder of EdNavigator. “We don’t want to admit that schools have a different agenda from parents,” he told me. We parents may have ideas on how to improve the educational experience at our school; a principal has to worry about other things as well, including keeping supervisors happy and living within contractual obligations. “Parents are not always right. Parents are also not always wrong. It’s that schools and parents frequently don’t have the same interests.” And schools are not generally open, Daly said, to giving parents a role as “co-strategists.” After all, educators are the experts.

That’s all true, of course, but surely there’s no harm in asking parents for feedback, input, and suggestions. It’s not a crazy idea that parents’ views might have some value. Some states (including Maryland!) have even considered including the results from parent surveys in their school accountability systems. And parent comments are a key part of the popularity of the GreatSchools.org site. Furthermore, at a time when many reformers are accused of being outsiders and not in touch with “the community,” why not extend an olive branch and indicate that parents’ ideas and concerns are taken seriously?

The honest reason why not, of course, is because it could make the principal’s life more difficult. Parents might want changes that teachers resist or are legitimately beyond a school leaders’ control—especially in big, centralized, bureaucratized systems (hello, Montgomery County!) where most key decisions are made at the school board or central office levels. And we can’t assume that all parents are going to want the same changes—which might be why parents in traditional public schools are less satisfied with engagement efforts than those in charter and private ones.

In some parts of the country, parents haven’t waited to be asked for their input—they’ve grabbed their pitchforks and forced school or district leaders to pay them heed. Parent activist and blogger Erika Sanzi has seen several such efforts up close. “A small brave group is all it takes,” she says, “armed with data and information and examples of how they do things in places that excel. A group of informed parents that doesn't only talk about their own kids but is fiercely on the side of improving the school/district for ALL kids.” And if the principal doesn’t listen? “Committed parents can be very influential when it comes to replacing lousy leaders,” she argues, “because once you get a strong leader into place—and they hopefully build their own leadership team— things begin to improve.”

As inspirational as such parents and their stories are, they seem to be the rare exceptions. What we need is a strategy that can work at scale.

Parent advisory councils

What might work to give parents a real voice in their schools? The most radical approach ever tried in the modern U.S.A. was probably Chicago’s experiment with “Local School Councils” in the early 90s. These councils, comprised of parents, community members, and educators, were given real power over the day-to-day affairs of their schools, even including the selection and oversight of the principal. This didn’t go well in some low-income communities, where it was harder to develop a coherent plan for school improvement, and where cronyism combined with racial politics stymied efforts to make real change. But in more affluent parts of the city there was some evidence of meaningful progress, both in terms of student learning and in community buy-in.

Those local school councils have been stripped of much of their power, but some charter schools and private schools still put parents on their boards, which is an avenue for parent voice and engagement. Another alternative is via parental advisory committees. Take the Namaste Charter School, also in Chicago. Maureen Kelleher, another Education Post blogger, told me about her experience with its “bilingual advisory committee,” launched by the teacher who leads the school’s dual-language immersion program. In Kelleher’s words:

It quickly became the most visible and active parent-school connection, building community, designing meetings where teachers went in deep with parents about the instruction taking place. We had a great session where teachers shared classroom video and led mini-exercises so parents could get a real feel for how dual-language was working in the classrooms. We also became a place where parents could get questions answered about much more than just the dual-language program.

Perhaps most importantly, our council became a go-to for recruiting parents to help with the most recent principal search. We also went to the board to talk about the principal search and the need to find and keep a strong school leader. I think we have her now—and she continues to rely on us.

Another example comes from KIPP Comienza Community Prep in Los Angeles. That schools’ Family Leadership Council “meets regularly with school leadership to represent the voice and needs of parents.” They set goals together, collaborate on developing a strategy to meet them, and team up when it’s necessary to advocate for the school and its students.

I’m a long way from L.A., but a Family Leadership Council sounds great to this suburban public school dad. Maybe it could be connected to the PTA, or maybe it would work better independently. The creation of such councils would probably need to be mandated by local school boards, at least in the district sector. But it strikes me that this is the sort of civil society invention needed in every public, private, and charter school in America.

Parents are an untapped resource for reform. It’s time to give them—to give us—a voice.

In a recent blog post, University of Virginia cognitive scientist Dan Willingham posits three possible types of personalization in personalized learning—children learning at their own speed, pedagogical tailoring, and individualized content. I have sought out all of these variations for my children over the years and, as Willingham notes, they are not mutually exclusive. But neither are they equally important. Let me make the case, as a father of two high school girls, that personalized pacing is a must-have, personalized pedagogy is a nice-to-have, and personalized content is largely to be avoided, at least until the end of the K-12 experience.

Personalized pacing and pedagogy

My children’s experience at The Metro School, a 6-12 STEM-focused early college school in Columbus, shows that students learning at their own speed is the prime mover of successful personalized learning (PL).

Metro’s model generally compresses what would be year-long courses in traditional schools into one semester. Course material is divided into discrete units and subunits, with each having clear goals for students and teachers and clearly connecting to the next. It moves fast, the expectations are high[1], and there is little downtime. Students’ progress is assessed regularly along the way, usually by teacher-created tests, giving teachers clear feedback about the challenges students are experiencing. Difficult-to-grasp concepts can be reinforced somewhat in future lessons, but Metro students are generally required to remediate their previous work while continuing with new material. More on this in a moment.

Willingham and others suggest that such a model sounds messy for teacher and student, but it is here that pedagogical tailoring comes in. Approaches to remediation must be varied: student-driven efforts to find and correct errors (test corrections, report revision, etc.), the same material taught in a different way by another teacher or a tutor, watching a Khan Academy video and doing sample problems, etc. Remediation can take place as homework, in a study hall, or after school in office hours. Whatever works to make the material understandable to students must be available. At Metro, pedagogical tailoring is not a distinct version of PL but just another tool in the same box.

Even with pedagogically tailored remediation baked in, however, fast-paced PL won’t work perfectly for every student in every subject, as my children can attest. But that is not a reason to reject the approach. Consider: The students who pass an accelerated class at the end of the semester move on (to another class, another teacher)[2], leaving, say, the sixth grade math teacher with only those students who still need help. By stating clearly what students must know and be able to do to succeed in the next math course and by employing ongoing assessment and remediation during the regular course to determine specific proficiency gaps, the teacher can concentrate her post-course remediation where it is needed. A rigorous, focused “recovery week” should be enough to boost most students’ skills up where they need to be. And even if a student is so far from proficiency that he would benefit from repeating the entire course, it’s still only one additional semester—ultimately the same amount of seat time as in a traditional sixth grade math class. The clear goals and unit/subunit breakdown allow teacher, student, administrator, and parent to know exactly where the child needs to concentrate her efforts in order to become proficient.

Contrast this approach with most traditional schools in which a student earns a “D” at the end of a full year of sixth grade math. If such a school would dare to hold him back, he would repeat the entire course again over another year. Perhaps there would be additional tutoring, perhaps not. If he takes the course again and finishes with a “C,” he has improved, but he has taken two years to do so and he still may not have proficiency enough to be successful in the next math course for which he is already “late.”

Of course, this doesn’t just apply to math at Metro. It also works for English, social studies, and science, too. Assessing and addressing students’ proficiency gaps from beginning to end of every course and having remediation time and methods immediately available provide the needed backstop to keep students moving forward even at high speed and even if they have incomplete mastery along the way.

As you can imagine, there is a lot of prep work involved in building a PL education model. All teachers in a department must be on the same page as to what their incoming students must know and be able to do to succeed in their courses; assessments must accurately define the required level of proficiency all along the continuum; and administrators must make sure that remediation practices are thorough and rigorous while causing the least interference with overall forward progress. But once that infrastructure is in place, PL looks pretty much like regular teaching and learning. Students and parents must understand the acceleration and remediation processes, but the basic pedagogy is wholly recognizable. The various lessons and approaches need not be invented anew every year. New teachers will need to get up to speed and additional approaches can always be workshopped and added to the rotation, but if you've got three or four reliable ways to teach sixth grade math or freshman biology concepts, that's probably all you'll ever need.

Personalized content: Proceed with caution

As for the “individualized content” concept, in my experience, its usefulness is far more limited than the others. My daughters attended a Montessori school from age three through the end of first grade. Montessori education is a tried and tested version of the individualized content concept. Children are free to choose their activities from the classroom materials and those activities are often multi-disciplinary. It allows students to spend a lot of time on subjects and activities of interest (making calendars was a huge favorite for both of my girls) but it also allows for procrastination and outright avoidance of topics of lesser interest, as Willingham and others fear. That’s where my family hit the wall with Montessori education—math avoidance. For me and my wife, that was a no-go: We were okay with our kids picking whatever they wanted from the buffet, but they still needed to eat their vegetables, too. When preschool was over and the stakes got higher, we left for a more traditional private school where math was never optional.

Interestingly, as our daughters are currently finishing up their high school requirements early (thanks to the acceleration described above), the opportunity to reintroduce individualized content has arisen: higher level elective classes, college credit-bearing classes, student-interest driven research, community service projects, student-led theatrical productions. Self-tailored and largely self-directed work, with an advisor to oversee inputs and outputs, this content is less traditional and more an exploration of future paths they could take in college and in life. Having achieved the “goal” of high school (i.e.—graduation) early, the stakes are lower for them than in preschool–the vegetables having been consumed, as it were–my daughters can take better advantage of the last two years of high school in fully personalized learning.

To summarize: the basis for successful PL is an accelerated course model with solid structure and frequent assessment of mastery. Most students can learn material successfully at a faster pace than in a traditional school. Those who need more time should get it, along with additional teaching methods based on the students’ individual needs. Once basic education goals have been met, individualized content can be deployed to take the place of readin’, writin’, ‘rithmetic, and the rest.

A note about technology

In my experience, technology is an adjunct to personalized learning, not the center of it. Perhaps it’s not surprising that tech billionaires interested in education would lean heavily toward this as the prime driver of personalization, but these tools are only as good as the knowledge of the user. Giving kids access to Quizlet, Prezi, Google docs, and Grammarly is valuable and may help some students engage more with the material, but software cannot educate a student in isolation. Coding is an important course that kids can take, but it’s not a way to learn English or American history. The ways that, say, literature have been traditionally taught are still rock solid—vocabulary, reading, analysis, writing papers—even at an accelerated pace. On the teacher side, Curriculet and Newsela are good reading applications and Schoology is a great platform for tracking gaps in proficiency—but it still takes great teaching to actually fill those gaps. My children can attest that laptops and tablets are valuable and one-to-one technology programs in schools should certainly be expanded, as should home internet access. But the personalization part of PL is inside the student, not inside the machine. “Kids on computers” is what you may see from outside the classroom window; inside, you will see that kids are learning the material at their own pace and in the ways that make sense to them.

My children have been inside those classrooms. It works.

***

Professor Willingham ends his piece by asking, “is personalized learning worth pursuing?” and suggests that more research on the topic is required to answer that question. Chan/Zuckerberg seems to agree and is proposing a rigorous R&D plan to engineer PL for the future. That’s fine—and those of us at the Thomas B. Fordham Institute would be happy to participate. But personalized learning has already proven its worth to me and many others. My children have thrived and I have seen the model I described work for students who were grade levels behind in math or English as well as d for students who would be considered gifted. I love this model and am gratified that others in education reform are starting to discover it. My children are living examples that PL works and allows students to achieve quantifiable growth and proficiency in subjects across the board.

[1] Metro employs “mastery grading”, in which students must reach at least 90% on most assignments or their work is “incomplete” and must be remediated up to 90%. This is not a requirement for successful PL, but is a good fit for acceleration of this type. With seat time eliminated as a determination of course completion, the proficiency bar can be set as high as we dare.

[2] Metro holds an intersession between semesters to facilitate remediation and elective courses, but the details are too complicated to go into here. Suffice it to say that deviations from a traditional school calendar are clearly required for an accelerated course model.

On this week's podcast, special guest Elisa Villanueva Beard, CEO of Teach For America, joins Mike Petrilli and Alyssa Schwenk to discuss whether TFA has moved to the left politically and educationally, and why that might be a problem. During the Research Minute, Amber Northern examines the prevalence of good American jobs that don’t require a bachelor’s degree.

Amber’s Research Minute

Anthony Carnevale et al., “Good Jobs that Pay without a BA,” Georgetown University Center on Education and the Workforce (November 2017).

Ever since the federal government mandated annual standardized testing two decades ago, test preparation, i.e., instructional time spent preparing students for tests, has been hotly debated. Critics argue that it negatively affects teaching and learning by focusing instruction on rote and procedure over more complex content, while proponents contend that test prep can improve instruction if the tests themselves, and the academic standards that they assess, are rigorous and high-quality.

Oddly, there’s little research to substantiate the claims of either side. So let us welcome a recent study on these issues by David Blazar of the University of Maryland and Cynthia Pollard of Harvard’s Graduate School of Education.

Using data previously collected by the National Center for Teacher Effectiveness (NCTE), Blazar and Pollard analyzed two separate measures of test preparation to answer two questions. First, does test prep lead to lower-quality instruction? And, second, does it make any difference if teachers are teaching to a more cognitively demanding test? 

The researchers used teacher surveys as well as transcripts from videotaped math lessons to determine how frequently teachers engage in test prep, and what types. The survey yielded self-reported data on how often teachers engaged in five common forms of test prep activities, including focusing instruction on students just below a given performance level on a state test and using standardized test items in their instruction. Blazar and Pollard also scoured transcripts of videotaped lessons to determine whether test preparation was a major focus of instruction. They then compared video and survey findings to teachers’ observation scores, as measured by the cognitive demand of math content provided to students, interaction with students, and the accuracy of content delivered.

After controlling for the background characteristics of teachers, students, schools, and districts, the researchers found that “test preparation is a significant and negative predictor” of instructional quality—though they stress that its negative effect is fairly modest and likely overstated in the current discourse. Somewhat surprisingly, when comparing the incidence of test prep to the rigor of respective state tests, they also found “little support for the moderating role of test rigor.”

The authors flag several major cautions. Most importantly, the data come from a relatively small, non-nationally representative sample of fourth and fifth grade mathematics teachers in Massachusetts, Georgia, and Washington, D.C., so it’s uncertain whether the findings can be generalized to other grades, content areas, and parts of the country. Data were also collected from 2010–13, prior to the administration of Common Core–aligned assessments such as PARCC and Smarter Balanced, so it’s unclear how test prep for “next-generation” assessments (which Fordham reviewed in 2015 and found to be generally high-quality) might affect instructional quality.  

Given all of the criticism and fearmongering around test prep in recent decades, surprisingly little research has investigated the relationship between test preparation and instructional quality. This study sheds a bit more light on this heated issue, and underscores the many complexities surrounding instruction, assessment, and student learning. But we clearly need more light to illuminate this tunnel.

SOURCE: David Blazar and Cynthia Pollard, “Does Test Preparation Mean Low-Quality Instruction?,” Educational Researcher (October 2017).

The gap in vocabulary for children growing up in poor households compared to their higher-income peers is well documented in research, especially for the youngest students just entering school. But shouldn’t the start of formal education begin to mitigate that gap? Research has shown that, unfortunately, initial gaps tend to persist, leading to a steep uphill climb by the time students are “reading to learn” in fourth grade and up. A group of researchers from the University of Texas at Dallas and San Diego State University recently studied whether the pernicious effects of socioeconomic status (SES) might negatively affect not only base vocabulary size but also the typical processes of word learning, which would serve to increase a child’s vocabulary going forward.

They recruited a group of 68 students ages 8 to 15 to take part in an experiment that required participants to use the surrounding text to identify the meaning of an unknown word. Each exercise included three sentences, all with a made-up word at the end. For example: “Pour some water in my raub.” This was the last of a three-sentence triplet designed to lead a reader to that “raub” means “cup.” All of the words comprising the context were high-frequency words typically acquired by age 8, but half of the triplet sequences were geared to give less obvious build up to the correct answer. The goal was to identify whether vocabulary, reading skills (both decoding and reading comprehension), working memory, and/or SES status correlated with success in identifying the unknown words. Through this experiment, researchers found that SES was negatively related to success in this task. Poor students did significantly worse than did their higher-SES peers. Reading skills and working memory were not found to be related to success. Some caveats related to this study include its small n-size and a high proportion of bilingual participants with varying levels of exposure to English.

This study provides support for the so-called Matthew Effect, which posits that students who enter school with a large vocabulary have an ongoing advantage over students who do not. The usual response to this is call for expanded pre-K and richer content in the earlier grades, but the students being studied here are well past that age. Luckily for them, there is plenty of research showing successful vocabulary-building interventions for middle schoolers who arrived with deficits. This study includes no data on the educational history and experience of any of the participants. Any extensions of this research should take into account the amount and quality of the education children have received because good schools matter, especially for students growing up in poverty.

SOURCE: Mandy J. Maguire, et. al., “Vocabulary knowledge mediates the link between socioeconomic status and word learning in grade school,” Journal of Experimental Child Psychology (February 2018).