This guest post is written by Craig D. Jerald of Break the Curve Consulting and author of “On Her Majesty’s School Inspection Service.”
Last week, Mike Petrilli at the Fordham Institute incited a welcome flurry of discussion about school inspections with a very thoughtful proposal in the Huffington Post. Citing my recent Education Sector paper on England’s* inspection system, Petrilli argued that it’s time to stop looking at numbers alone when evaluating American schools and begin to capture a broader range of evidence best collected through human observation in schools themselves.
Anne Hyslop of Education Sector and Robert Pondiscio of Core Knowledge each responded to Petrilli’s proposal with some degree of skepticism. While their responses raised a number of issues, those issues really boiled down to two major concerns: the cost and the potential corruptibility of school inspection systems. While such concerns must be treated seriously in American debates about whether to adopt school inspections, they are not insurmountable. After all, this is not some radical, untested idea. Dozens of other countries around the world have found feasible and reliable ways to inspect their schools.
Hyslop convincingly argues that it simply would not be feasible to recruit and pay enough inspectors to conduct annual inspections of every public school. Therefore, she proposes that inspections only be conducted in schools already judged low-performing, based on test scores and other quantitative indicators, in order to make a final determination of whether interventions are called for and to guide improvement efforts.
But Hyslop’s proposal misses an important point that Petrilli makes: Inspecting all schools would help identify ones that are achieving acceptable quantitative outcomes through unacceptable means, such as narrowing curricula or replacing good instruction with test-prep strategies. Inspections also can help assess whether schools are serving students’ interests in ways that go beyond helping the minimum number of students achieve acceptable test scores. Kate McGovern of Teach Plus provides an excellent example from her own experience teaching in the United Kingdom.
A better way to reduce the costs of inspections would be the following: First, adopt a general policy of inspecting schools every few years, perhaps beginning with a three-year cycle. Second, prioritize schools for inspection based on their quantitative outcomes and their previous inspection results, such that the subset of schools Hyslop rightly sees as most urgently in need of inspections would be moved to the front of the queue. Third, increase the time between inspections for schools that achieve one or two rounds of excellent inspection results, perhaps to five years. Finally, continue to monitor qualitative data annually so that an inspection can be “triggered” if any school’s outcomes show marked deterioration.
Based on my analysis of England’s system, the costs of such an approach would still be significant, both in terms of human capacity and financial demands. However, the United States has made school accountability the centerpiece of education policy for more than a decade, placing far more emphasis on accountability than most other countries. If those other countries can find the resources to hold schools accountable based on more than narrow quantitative data, why can’t the United States? It’s time to step being penny-wise and pound-foolish in funding the systems we use to hold public schools accountable in this country.
Second, potential corruptibility. Pondiscio argues that it would be far too easy for schools to fool inspectors by sprucing up bulletin boards and making other cosmetic changes during the days on which a school is inspected. But such risks also can be greatly mitigated in the design and implementation of an accountability system. First, give schools short notice of inspections so they are not tempted to spend weeks preparing a “dog and pony show.” When England began giving schools one or two days’ notice of inspection, instead of several weeks’ notice, everyone, including teachers, liked the new approach much better. Second, equip inspectors with a full complement of quantitative data on student outcomes before they visit a school. English inspectors do not simply walk blindly into a school and begin observing; they begin observing having reviewed a wide range of quantitative performance data, including student test scores and value-added growth measures.
Perhaps the best way to mitigate such risks is by ensuring that inspections are based on frameworks and rubrics that do not focus on simple surface features of schooling, but rather guide the collection of much more meaningful evidence. In England and other countries, inspectors do not evaluate a school simply by “walking its halls and sitting in its classrooms,” as Pondiscio suggests. Inspectors are trained to systematically observe classroom teaching against a detailed rubric in order to collect evidence against a set of meaningful standards. Research in the area of teacher evaluation suggests that, under such conditions, direct observation can yield reliable results that correlate with student learning outcomes. It is very difficult to “fake” excellent instruction when it is not a common daily practice in a school.
The bottom line is this: If potential corruptibility is an absolute disqualifier of any given accountability measure, then we will be left with no accountability measures. How many more accounts of cheating on tests, teaching to tests, and gaming test scores (for example, by focusing on “bubble students”) will it take before we admit that so-called “objective measures” are just as potentially corruptible as measures like inspections that rely on human observation and judgment? Indeed, relying on multiple kinds of evidence collected through a variety of methods, from testing to surveying to observing, is one way to discourage cheating and gaming of any kind.
*Like American states, each country in the United Kingdom of Great Britain and Northern Ireland has its own education system. As a result, England, Scotland, Wales, and Northern Ireland implement separate school inspection systems that share some similarities but also differ in significant respects.