I am a college professor and a mother. My Ph.D. is in Lifespan Cognitive Development, and I have taught developmental psychology and educational psychology for 25 years at Syracuse University, Ithaca College and Kenyon College. My children are 5 and 9; they are entering kindergarten and fourth grade in the Mount Vernon City Schools district. Despite state mandates, my children have had excellent educational experiences with their teachers in the local public school. I submit the following as partial testimony to the Ohio Board of Education in regards to some problems I have observed with the Third Grade Reading Guarantee (TGRG) and its implementation.
The TGRG testing has conflated criterion- and norm-referenced testing. The differentiation between the two types of tests is important. Criterion tests assess whether a learner has achieved specific skills or concepts, whereas the goal of norm-referenced tests is to rank learners by using a normal distribution (“bell curve”). The Third Grade Reading Guarantee test should be a criterion test, assessing whether a learning can demonstrate a percentage of a specific body of reading skills, and a passing score should be predetermined and known before the learner ever takes the test. However, the TGRG “cut scores” vary each year and are sometimes determined after the test has been administered. A norm-referenced test, too, is established a priori but by using a national sample to create a normal distribution, and a test-taker’s rank is determined based on where they score compared to that national sample.
The goal of the TGRG is not – or should not – be to rank learners. However, the TGRG seems to be an unconventional conglomeration of criterion-referenced and norm-referenced tests. In some years (e.g., 2016), the cut scores were determined after the third graders have taken the test. From 2016-2017, the cut scores went UP (perhaps the scores went up in other years as well) on at least one of the acceptable alternate tests. From a scientific tests-and-measurement perspective, how are these even possible? Why does the amount required to be known (the cut score determining whether one has demonstrated enough reading ability to “pass” the TGRG test) depend on the scores that the third graders themselves score as a body? The passing score or percentage should be known in advance (e.g., “We will require that students demonstrate 70% correct to be considered eligible to advance to fourth grade.”)
Further, why are the cut scores increasing each year? How can any subsequent group of children be expected to score higher than the previous year’s children? They should either meet an a priori cut-off, or they don’t meet the cut-off. The cut scores shouldn’t be changing or increasing from year to year. Is every district in Ohio populated by an expanding Lake Woebegone, where each subsequent cohort is increasingly above average? From a tests-and-measurement perspective, the creation of the cut scores themselves – and how they are used to determine passage or failure – seems to make no logical sense.
The Third Grade Reading Guarantee has created a nasty mess of academic push-down to lower grades. Kindergarteners are required to take a reading readiness assessment in the first weeks of school, just as classroom norms, expectations and relationships are being formed. We want entering five-year-olds to view their teacher and other school personnel as adults who they can trust and to turn to if they have a question or need help. And instead, we manipulate this developing relationship by giving them a test where they don’t know the answers, and the teacher is not allowed to help them. They learn that they cannot look to their trusted adult to provide the knowledge that they’re ostensibly in school to learn. They learn that their relationship with their teacher is not one of unconditional support, but a limited one instead.
The kindergarten reading readiness test is not in any way a valid assessment of reading readiness or reading ability. There is a 6-page passage that is read aloud to the child. Each page has 3-4 sentences on it, some of which may be “new” information to the child. There are then 6 multiple choice test questions, which require the child to keep MULTIPLE pieces of information in their working memory for way longer than is possible for a five-year-old. AND the child has to discriminate whether they learned the information from the text that was just read to them, or whether it is a fact they already might have known. They then have four more passages with six questions each to complete in a similar manner. Some of the multiple choice answers are completely confusing (e.g., asking the child to discriminate between “making pizza” and “cooking pizza”; what IS the difference between those actions?). A quick look and it’s easy to see that some of them are completely invalid.
This is not a reading readiness or comprehension test. It is a LISTENING test. It is a WORKING MEMORY task. It is a SHORT TERM MEMORY test. It is a DISCRIMINATION task. It is a SOURCE MEMORY task (“Did I learn this from the text? Or did I already know it? Or did I already know it AND it was mentioned in the text?”). It is a PRIOR KNOWLEDGE task. It is an “experience with testing” task. And because of curricular changes brought about by TGRG, kindergarteners now do this sort of stuff ALL DAY. But the sole reason our district decided to move to an all-day kindergarten in 2013 was to give students extra learning time in reading and test taking.
Kindergarteners and other young elementary students spend the BULK of their educational time doing stuff that is DEVELOPMENTALLY, EDUCATIONALLY, and MOTIVATIONALLY inappropriate. At that age – and until at least age seven, PLAY is the most important educational activity for developing minds and bodies (for at those ages, the same parts of the brain are recruited for cognitive and physical tasks – the very acts of reading and math are physical at those ages).
In our district, keyboarding instruction will begin in kindergarten, starting with the Fall, 2017 incoming kindy class. This is in addition to the sometimes hours of “seat-work” that so many kindergarten classrooms spend much of their time doing. The kindergarten classroom in our building has been required to remove the construction, loose parts, creativity-arts area to make room for a bank of 5 computers with keyboards so incoming kindergarteners – my daughter among them – can participate in keyboarding instruction. Keyboarding instruction for five- and six-year-olds is completely developmentally and educationally inappropriate. Phonics involves matching letter-to-sound and involves visual coordination of the eyes with sound. This is a MOTOR activity as well as a COGNITIVE activity; it involves integration of at least 3-4 systems (ocular-motor, cognitive-attention, and listening and/or producing sound).
Kindergarteners need to be learning to associate a visual letter with an auditory sound (either A with “A”, the letter name, or A with “aaaa,” the phonological sound the letter makes). They should NOT be searching for and learning a geographical location for each letter and involving the arms and fingers; this is not a helpful association to make when learning letters and their associated sounds should be prioritized for beginning reading. It is not only overly complicated and likely motorically very difficult, it is also teaching them associations that could confuse and delay the eye-letter-sound associations that need to be learned to a point of automaticity.
Further, we want to develop the efficiency of their eyeballs left-to-right across text, NOT their eyeballs from screen-to-keyboard-to-screen. The development of their motoric base is not yet ready for keyboard as a mediated vehicle between letters-and-sounds. Learning activities that are appropriate to this age, stage and pre-reading involve large gross motor activity with both sides of the body, integrating left-and-right.
Kindergarteners’ hands are not yet large enough for a keyboard and for many, the fine motor control and eye-hand coordination necessary to master keyboard typing – or even be minimally efficient – has not yet developed. Most children of this age should be working on (pencil-crayon-marker) grip strength, providing appropriate pressure to produce writing (or, more developmentally appropriate, coloring and drawing), and the eye-hand coordination for an activity that is a direct manipulation (eye to hand-crayon-tip) and which immediately produces an outcome, NOT a keyboard stroke-to-screen that is mediated by the keyboard. Developing fine motor coordination, like drawing and coloring to improve eye-hand coordination, so their eyes get used to looking at and producing print and drawing is appropriate.
They also need the language skills of listening to and producing words; listening to stories, learning and performing choral chants and singing are more valuable to a strong linguistic base than is attempting to compose writing on a screen through a keyboard. They are being asked to learn keyboarding at this age for the sole purpose of preparing them to take the TGRG test.
We see children’s random movements as “attention deficit” when it is more likely to be immaturity – or typical age appropriateness – in the physical/motoric system. Their proprioception and vestibular systems are still developing. Rather than keyboarding and seatwork, the curriculum needs perceptual-motor activities and opportunities to challenge and practice postural control: gross motor physical activity, not training the children to sit for longer and longer time periods. This type of training will only result in more children being identified and labeled as needing help.
Every moment that children spend in developmentally inappropriate activities are moments that are (1) lost to DAP and (2) moments that will contribute further to their frustration, dislike of the learning process, a-motivation for school, and alienation from school and the learning process. We know that one of the most developmentally and educationally important activity that children – and other social mammals – can participate in is play. The frontal lobes develop during play. Opioids are released during play, and when animals are deprived of play as children, they develop into adults who exhibit emotional dysregulation, including increased anxiety, depression and social inhibition and aggression.
It is not a stretch to hypothesize that deprived of play, humans will seek positive stimulation elsewhere (e.g., “self-medicate” with other activities or substances). Humans aged 3-6 need up to three hours of play a day in order to not exhibit problems associated with play deprivation. This is in addition to the recommended 2-3 hours/day of physical gross motor activity. By age 6-7, there is enough cortical inhibition to sit still and pay attention for short periods, but the bulk of a kindergarten day should be spent in play.
By third grade, children are 8-9. Eight- and nine-year-olds do not yet uniformly have the working memory necessary to answer a multiple choice test question with three alternates, let alone a test with questions on one digital page referring to text that they read several digital pages back. These are not reading tests; they’re tests of working memory capacity, which at this age is still very small. Let me explain how this is a problem. Children have to answer questions – on different screens – about text that is several screens back. There are questions about editing sentences – sentences that they cannot see unless they click back several screens to see the text again…and then click several screens forward to the question and its answers….hopefully the text they had clicked back to is still in their working memory. If it is not, the child will have to click back and forth several times. Why is the sentence in question not printed ON THE QUESTION PAGE itself?
But even if it were, eight- and nine-year-olds still do not have the working memory capacity for a multiple choice question to be a valid assessment. The child has to hold the stem – a question or sentence fragment – in working memory while reading all the alternates (the possible answers). The longer the question or question fragment and the longer each possible response is, the more the child has to hold in their working memory. The validity of the item goes down; it’s now a test of working memory, but not a valid measure of whatever ELA skill that it is purported to assess, because the item conflates working memory ability with that ELA skill being assessed.
I watched a third grader take a computerized ELA assessment in preparation for the TGRG test (not the TGRT itself, but a mid-year benchmark test that was a part of the ELA test-prep curriculum purchased by our district). There were multiple passages with many multiple choice items for each passage similar to what I’ve described above – multiple screens and flipping back and forth. Then the child got to number 53 – FIFTY-THREE – and it was an ESSAY. The very last question – out of 53 – was an ESSAY, the third essay of the test! Note that the very definition of a benchmark assessment is that it is short. Do these test makers know NOTHING about making up tests? Testing fatigue sets in after 15 items; this is true for college students and adults, not just elementary students. Testing fatigue will occur at fewer items than 15 if the items are complex (like clicking back and forth to edit sentences or reference text to answer questions, as described above).
When testing fatigue sets in, accuracy and speed decline, so that items after item 15 are more likely to be wrong that items 1-15. Tests longer than 15 items begin to lose their reliability – for any age. Furthermore, eight- and nine-year-olds do not have the sustained attentional capacity for a test of this length, even if they didn’t have to face testing fatigue. Sitting and maintaining posture is still a challenge at this age. If we ask 8-9-year-olds to read or count backward while standing still, they will have increased postural sway. Cognitive performance is sacrificed for balance. Even in third grade, reading is a PHYSICAL challenge and still relies on areas of the brain (cerebellum) that we think of as “motor” areas of the brain.
The essay this test taker faced at item #53 was, “Most people have faced an interesting problem that they had to solve. Think about an interesting problem that someone might solve. Now write a story about a character who solves that problem.” Consider this, if you can, from an 8-year-old’s perspective: They have to THINK up an “interesting” “problem” (What is “interesting”? What makes a problem “interesting”? Is that the key word here? What kind of “problem” – a math problem? A problem like saving the earth? Is “problem” the key word?)…figure out how to solve it…write a STORY about a CHARACTER that solves the problem. Are you kidding me? What child has the stamina to not only THINK of an INTERESTING problem, think up the SOLUTION to the problem (because even though you and I, as adults, know how to read between the lines and could write a story about SOLVING a problem without actually solving the problem, we know that most 3rd graders will interpret this as “write a story about a problem that has been solved, where solving it is the action of the story”.
I have no idea what the child wrote, as I was not privy to that. The child asked whether the story had to start “once upon a time” (when he read “story” that’s what he thought). I said no. He asked whether he could write about something that people already knew about. I said, “Like what? Give me an example. What are you thinking?” He said, “The three little pigs” (The Three LIttle Pigs was playing in the room next door). I said, “No. You have to think up something that everyone doesn’t already know about.” And then I left him alone. His only questions after that were how to spell things, which under a standardized testing administration, we would not be allowed to assist.
I will close with an anecdote. I teach a class of ~30 students in Educational Psychology every year at Kenyon College, a highly selective liberal arts college. A few years ago, when I first introduced the idea of the Third Grade Reading Guarantee, and that children who did not “pass” the one standardized assessment would not advance to fourth grade, my junior- and senior-level students were astonished. They were incredulous that advancing a grade level for eight-year-olds would be dependent on passing ONE test.
At that point, a student raised his hand and said, “But I didn’t learn to read until fourth grade.” Then three other students raised their hands and expressed the same sentiment – that they hadn’t learned to read until fourth grade. Three of the four did not have any known learning disabilities. Fully ten percent of the class did not learn to read until fourth grade; they were all typically developing and ultimately academically successful. There is a normal distribution around the average age at which children learn to read; some children will be on one or the other end of this distribution.
These four students were, by any definition of the concept, academically successful. They were juniors and seniors at a college that is known for its English writing program; some of them were about to graduate in a few weeks. The student who raised his hand first was a National Merit Scholar in high school (I just checked his LinkedIn page), and he is a published author and currently teaches English Language Arts at a middle school in New York.
~Andrea S. White, Ph.D., Kenyon College