Archive Hosted by the AFL-CIO

A Student, Two Teachers and the Limits of Test-Based Education

December 03, 2013 By Michael Diedrich, Education Fellow

Here’s a piece by a Massachusetts student, discussing her education's gradual erosion as test-based practices made their way through that state's school system. It’s a painful reminder of what education can look like when we’re not focused on the next bubble sheet.

The student also offers a perspective on a path forward. She writes, “The best, most interesting, and most intellectually challenging classes I took in high school were not the high level AP classes, which challenge mainly through the sheer amount of information, but the senior classes that teachers had been given the freedom to design themselves, with no MCAS or AP test writers peeping over their shoulders.” Great educators doing what’s best for students is incompatible with a test-based education.

So what is it about test-based schooling that quashes teachers’ ability to do what they do best? It’s not just the tests. They’re the starting point, but they remain simply one imperfect source of data. The real damage comes with the creation of a school climate that values the tests above and beyond any reasonable contribution they can make.

A ten-year New York City schools veteran recently called that climate out. Laurel Sturt started teaching at age 46, and was part of the Teaching Fellows alternative licensure program. When asked about her school climate in the No Child Left Behind era, she says, “I saw a lot of problems with all the testing, with all the slogans everywhere, as if you were in North Korea or something.... I resented the fact that we were test-prepping them all the time and we couldn’t give them a rich, authentic education.”

Contrast that with the Massachusetts student’s experience before the testing climate set in: “As young children, we spent time interviewing and talking to our fellow students about our varying countries and cultures, as well as interviewing and getting to know our custodians, principals, secretaries, and cafeteria workers. As older kids, we had many long class discussions about issues such as identity, racism, sexism, and what ‘America’ and ‘American’ mean and have meant throughout history.”

There’s a stark difference between that sort of curriculum and the one imposed on a Delaware teacher who recently protested the script handed to her by administration. She was explicitly told to teach the prepackaged curriculum word-for-word. It’s a story that will be familiar to many Minnesota teachers. Here’s an educator who, as she puts it, has “seen learning disabled, non-readers become college graduates; non-writers grow to be valedictorians; reluctant readers become bookworms.” Down to her bones, she believes in students’ innate ability to succeed. However, her freedom to facilitate that success has been hampered by the inflexible script she is forced to read to her students.

Again, to keep ourselves grounded in what’s possible, let’s revisit the Massachusetts student’s recollections of her middle school years. Of a teacher-designed Humanities class merging English and history, she writes, “My teachers and the curriculum they had designed challenged us to develop more complex thoughts and then helped us build up the capacity to express them, whether through talking, writing, or acting.” That sort of teaching is increasingly repressed by one-size-fits-all curricula. Those curricula are sold to districts based on their purported ability to increase test scores and their alignment with standards.

I distinguish between data-informed teaching and a test-based education. These are not synonyms, though they may look like it at first. Data-informed teaching takes data from a range of sources -- real-time checks for understanding, teacher-designed formative and summative assessments, standardized tests, qualitative data gleaned from relationships with students, and so on -- and constructs information from it. That information is then used to modify teaching to meet students’ needs.

This is what drove the Delaware teacher before the scripts were passed out. As she puts it, “[M]y job as a teacher was to discover each child’s pathway to learning and help them to embark on that path. My calling was to meet the needs of the child.” Call me crazy, but that sounds like a good teaching mindset to me.

Test-based education is different. This is education that starts with a single data source and reconstructs schools to focus obsessively on that one source. It’s unhealthy, and it narrows the scope of “education” in a way that’s disrespectful to students, families, and teachers.

Despite many good intentions, we have wound up with a school system that is increasingly test-based rather than data-informed. We are wasting students’ time and enthusiasm, teachers’ abilities, and our schools’ potential. The hope that increased attention to outcomes would translate into an instructional renaissance has gone unfulfilled in too many schools.

We don’t need to eliminate tests or data to improve our schools from their beleaguered state. We do need to show students we value them for more than their test scores and per pupil funding allocations. We need to unchain our teachers with the capacity to do more. If we don’t, we’ll continue to see the dreams of idealist elites twist into a lousy education for too many students.

Thanks for participating! Commenting on this conversation is now closed.


  • Stan says:

    December 9, 2013 at 5:56 pm

    Michael Diedrich writes…  “Great educators doing what’s best for students is incompatible with a test-based education.”

    Building his argument on this statement illustrates the incomplete thoughts that have come to permeate the debate on how to improve education.  This “absolute” statement diminishes the abilities of “great educators” to use all of the tools in the tool-box, including test-based education, to accomplish the objective of education.  The problem is that while some subjects are ideally suited for test-based education; other subjects are completely incompatible.  However the converse is also true; some subjects are incompatible with the “freedom to design” based education especially if the teacher is unqualified or ill-prepared.

    Michael goes on in this article to practice “poor science”.  The anecdotal reflections of a Massachusetts student are the basis for the core value of teacher flexibility to create intellectually challenging learning opportunities.  The anecdotal comments from Laurel Sturt support the conjecture that the school climate was “test-prepping all the time and we couldn’t give them a rich, authentic education”.  And finally the anecdotal report from a Delaware teacher who saw “learning disabled, non-readers become college graduates; non-writers grow to be valedictorians; reluctant readers become bookworms”.  I am sure that all of these are true; certainly there are not any advantages to be gained by advocating for their ideals which might have biased their recollections. 

    Unfortunately we all know that the other side of the anecdote was also present.  There were students who were not intellectually challenged by the teacher in Massachusetts and who thought the teaching was trite.  There were students in New York that craved the thrill of the test and received powerful motivation at seeing their scores.  There were disabled students in Delaware that never saw college opportunities; valedictorians who in fact were non-writers in spite of their class rank; and readers for whom reading comics qualified as being a “bookworm”.  The world is far from perfect and anecdotes do not equate to outcomes.

    Nowhere in Michaels’ opine is there any “scientific method” support based on development of hypothesis; prospectively assignment of a randomized study cohort; or longitudinal assessments of a targeted outcome.  If the outcome was to develop students who in retrospect felt their teachers developed “intellectually challenging” learning opportunities then the student’s teacher in Massachusetts has succeeded.  Does this equate to the ability to be a productive contributing member of society for the subsequent five decades; time will tell but this intellectually challenging subject is not even broached. 

    Allowing for the potential of a normal distribution of traits in biologic populations (including students) will lead to the conclusion that one size or method will not fit all.  I suspect that there is a fairly large sub-population (say 1/3) that can pretty much teach themselves all they need to know if they have basic skills (reading, writing and mathematics) regardless of the nature of the teacher, or the tests.  Then there may be a second sizable sub-population for whom the individualized personal instruction methods are critical to learning.  Then there may be a sub-population for whom the rigor and a deadline of a test will become the motivation for acquiring knowledge.  Finally, there will be a sub-population (hopefully a very small proportion) for whom the specific subject being learned will never become understood, internalized, retained or utilized.  Ideally an academic leader will help channel and guide students along the way so that they can discover their abilities and ways to acquire the knowledge that may be required.

    Michael closes by appropriately advocating for data-informed processes to guide the teaching methods.  However, we also need a data-informed processes to align specific teachers to specific students in order to achieve specific outcomes.  and in some cases, data-informed process to help some teachers seek a new occupation or some tests to never used again.  In some relationships “teacher discovery of individual student learning pathways” will be appropriate.  In others, curriculum rigor defined by testable-outcomes will be appropriate.  What becomes unworkable for a large percentage of learning outcomes is either the freedom-only, teaching based on good intentions as advocated by the anecdotes, or the script-reading, teach to the test.

    In the end we need to be honest that not all teachers can be treated equally, not all students can be treated equally and not all outcomes can be valued equally.  Once we arrive at this conclusion than we can let a “marketplace” process begin the prioritization, allocation and distribution of education; and there will neither be complete freedom or equality. 


    • James says:

      January 24, 2014 at 10:25 am

      Well Stan, you lost me as soon as you mentioned “poor science” and the “scientific method.”  I happen to be of the opinion that societal matters, as well as human action and thinking, cannot possibly be quantitatively measured by the limitations of science, which is only effective in describing material things.  When describing human action, I believe that anecdotal evidence is perfectly appropriate.  You would try to lump every human being neatly into one sterile, homogenized, measurable formula or theory, and that is an impossible task.  Even if a pure isolated concept of “science” were able to accurately describe human action, science, as it is practiced by man and his limitations would still be limited by limited human senses.  I know materialists like Mark, et al, like to lump all humans together and describe their actions by scientific, mechanical explanations, chemical reactions, etc., but humans are greater than the sum of our parts, and thus cannot be so described.  Anecdotal evidence, on the other hand, can accurately describe the possibilities of human potential, a thing which cannot possibly be measured, and thus give people something to strive for - a chance to be better than they are, unfettered by clod-like scientitsts, who put far too much faith in their limited senses.

      • james says:

        January 24, 2014 at 11:08 am

        Please forgive my typos - never took typing.  Also, I meant to say “Marx,” not Mark.