Why Standardized Testing Doesn’t Measure What Parents Think It Does Anymore
I was idling in the elementary school pickup line last May, scrolling through my phone and waiting for the first day of state standardized testing to let out, when my 10-year-old son Luka rounded the corner of the building. His backpack was slipping off one shoulder, the blue crayon he’d stuffed in the side pocket was poking out, and he was already halfway through the granola bar I’d packed that morning. When I asked how testing went, he shrugged and said “Fine. I finished the reading section 22 minutes early, so I drew a T-Rex fighting a robot on the back of my answer sheet.” I laughed and asked if any of the other kids did that too, and he said half the class did.
A month later, the score report showed up in my email. Luka, who reads three chapter books a week for fun, beats me at Scrabble every Sunday, and can explain the plot of every Star Wars movie in 10 minutes of detail, was marked “approaching standard” in reading. I stared at the screen for a minute, then emailed his teacher, Ms. Henderson, to ask if we could chat for a few minutes after school.
She pulled out a redacted copy of Luka’s answer sheet when I got there, and pointed to the first wrong answer. The reading passage was a personal narrative about growing up on a corn farm in central Iowa. The question asked what the author meant when he wrote “we hunkered down to get the harvest in before the rain.” Luka’s answer: the family took shelter inside from the coming storm. That’s what hunker down means to him—every time we get a bad thunderstorm warning here, I tell him we’re hunkering down inside until it passes. The test makers assumed every kid would know “hunker down” in this context meant working fast to get the crops in, but Luka never set foot on a corn farm. He’d never experienced a harvest. That wrong answer wasn’t because he couldn’t read, or couldn’t infer meaning. It was because the question tested his background knowledge of a life he’d never lived, not his reading skill.
We went through the rest of his wrong answers. Three out of the five had the same pattern: the question relied on context or experience Luka didn’t have, not the skill the test was supposed to measure. I’ve volunteered to proctor testing at the school for the last three years, so I already knew what else these tests measure that no one talks about. I’ve seen a 9-year-old who gets panic attacks have to leave the room halfway through because her hands were shaking too hard to hold a pencil. I’ve seen kids who get free school breakfast, but testing ran late that day, so they were too hungry by the third hour to focus on the last section. I’ve seen kids who are great at critical thinking get tripped up by questions that have three technically right answers but only one the test maker considers correct.
I don’t think standardized testing is useless entirely. Districts need some way to spot gaps in programming, like if a whole group of kids in a school are struggling with a specific skill that the curriculum isn’t covering well. But it’s not measuring what most of us think it is: it’s not measuring how smart your kid is, how much they’ve learned over the school year, or how ready they are for the next grade. It measures how well they can sit still for three hours straight, how well they handle high pressure, how much shared background experience they have with the people who write the test questions, and whether they happened to feel well that day.
That’s not to say there’s nothing parents can do to make sense of test scores, or to take the pressure off our kids. After Luka’s score came out, I started doing a few small things that have helped. First, when I get the score report now, I don’t fixate on the overall rating. I pull up the breakdown of individual questions, then sit down with Luka for 10 minutes after dinner and go through the wrong ones together. Half the time, it’s not that he didn’t know the skill—it’s that the question was confusing, or relied on context he doesn’t have. That tells me way more than a single number ever could.
Second, if the score is way lower than what I expect from what I see at home, I email my kid’s teacher and ask for 15 minutes of their time to look through class work from the semester. Ms. Henderson showed me Luka’s in-class reading responses that semester, where he analyzed a story we read for the book fair and wrote a two-page response asking deeper questions about the main character. That’s the work of a kid who’s exceeding reading standards, not approaching it. Teachers see what your kid can do every day, and that’s almost always more accurate than a single test taken on one Tuesday in May.
Third, I changed what I say to Luka before testing week. I used to say “make sure you try your best, this is important” which just made him nervous. Now I say “This test just helps the school figure out what programs they need to improve. It doesn’t tell anyone how smart you are, and it doesn’t change anything for your next grade. Just do your best, and if you finish early, you can draw on the back if you want.” It takes the pressure off immediately, and it’s honest.
This year, Luka’s in 5th grade, and testing week is next month. We’ve got his favorite gummy bears packed for snack, we’re going to keep bedtime the same as usual, and we’re not talking about the test every night leading up to it. I know I’ll get a score report in a few months, and I’ll look at it, but I’ll also keep looking at the kid who stays up late reading graphic novels, who explains chess strategies to his little sister, who asks me questions about how the world works that I can’t even answer. There’s still a lot that doesn’t add up about how we measure learning right now, but that much is clear: the real measure of what a kid knows isn’t going to show up on a bubble sheet.
Please indicate: Thinking In Educating » Why Standardized Testing Doesn’t Measure What Parents Think It Does Anymore