My daughter came home from school with a research paper assignment that felt dangerously open-ended. The teacher had listed a bunch of broad themes—consumer behavior, community health, media influence—and told the students to pick something, come up with a question, and gather original data to answer it. No big deal, right?
My daughter, a high school junior, stared at the rubric like it was written in code. Then she announced: “I’m going to survey people at McDonald’s.”
I made a face before I could stop myself. “Why McDonald’s?”
“Because everyone goes there,” she said. “And it’s easy. You can ask people about their eating habits.” She already had her mileage. The question was something like, “Do you think McDonald’s is a treat or a regular part of your week?” She wanted to know if people were making conscious choices or just going through routines.
I figured it was mostly a convenient excuse to buy french fries, but I said okay. The assignment required a survey of at least thirty people, and she had a plan. She printed out a neat little questionnaire on her school laptop, with ten questions and a Likert scale. She was very proud of it.
The first time we went to McDonald’s was a Thursday afternoon, right after school. The place was a sleepy wasteland between the lunch and dinner rushes. A few retirees sat with coffee, one mom was wiping down a high chair, and two teenagers were sharing earbuds. My daughter stood by the condiment station with her clipboard like she was about to sell insurance.
“Maybe wait until someone’s done eating,” I suggested. “Or catch people at the register before they sit down.” She nodded, took a deep breath, and walked over to the mom with the wipes. The woman listened, looked at the paper, and said, “I’m not really a McDonald’s person, but my kid likes the playplace. Is that helpful?” My daughter said yes, but her question had no place to record that kind of answer. She wrote it down anyway in the margins.
The next ten minutes were rough. She approached a man reading his phone, who barely glanced up. She asked two teenagers, who laughed and said “No thanks.” She caught someone on the way out, and the person said, “I don’t do surveys.” I watched her shoulders drop a little each time. She started mumbling, “This isn’t going to work.”
Then an older woman in a floral blouse asked if my daughter was doing a school project. My daughter explained, and the woman spent a solid four minutes answering every question, then added a story about how she used to bring her kids there after soccer practice, and now she only comes for the egg McMuffins on Saturdays. My daughter said “thank you” about five times. When she came back to the table, she looked at the answer sheet and said, “That didn’t fit the scale at all.”
That’s when we actually started talking about how research works in the real world. Not in the textbook sense. I sat down with her over a burger and we went through her questions one by one. She had asked, “How often do you eat at McDonald’s?” with boxes like “rarely, sometimes, often, very often.” But the woman’s answer was really about a specific memory. The man on his phone didn’t answer because he was at McDonald’s alone, passing time. The teenagers laughed because the question felt too personal. The mom said she wasn’t a McDonald’s person, but then she ate a chicken sandwich with one hand while pushing a stroller with the other.
“Your data isn’t clean,” I said. “But that’s not a bad thing.”
She looked at me like I had just said the floor was the ceiling.
We tried again the next Saturday morning. Different McDonald’s, busier location, more families. She had rewritten her survey. Instead of a Likert scale, she started with, “Can I ask you a few quick questions about why you came here today?” She added a spot for “other” and an open-ended question at the bottom. She also learned to say, “You can just say ‘next’ if you want to skip one.” That made people more willing to talk. One dad said he was there because his daughter had a swim meet and this was the only place between the pool and home. A couple said they were on a road trip and McDonald’s was “the same everywhere.” A kid in a little basketball jersey answered every question with “chicken nuggets,” and my daughter wrote that down, too.
By the end, she had thirty-two responses. The data was a mess. Some people circled two numbers. Some skipped the question about income. One person wrote “salary is none of your business” in ink. My daughter spent the next two nights trying to make sense of it. She had expected to find a simple answer—maybe “most people eat McDonald’s for convenience.” Instead, she found that convenience meant a hundred different things. For the swim mom, it was efficiency. For the retirees, it was cheap coffee and a place to sit. For the road trip couple, it was predictability. For the kid, it was pure joy.
I helped her organize the survey results, but I didn’t do the analysis. She had to decide what to do with the outlier responses. She had to explain in her paper why some people refused to answer. She had to acknowledge that her sample was biased—mostly middle-class families, not many teenagers who walk to McDonalds after school, no one who works the late shift. Her paper ended up arguing something more nuanced than she originally planned: people don’t just eat at McDonald’s because it’s fast. They eat there because it serves a specific role in their day, and that role changes depending on who you are.
When she got the paper back, she got a B-. The teacher wrote, “Good effort, but your survey sample is too small to support a strong conclusion.” My daughter was a little deflated. Then she shrugged and said, “At least I know why.”
I think about that McDonald’s survey a lot now. Not because the paper was groundbreaking, but because it taught her something that no textbook could. Research is not about following a perfect procedure. It’s about dealing with people who don’t follow the rules, who laugh at you, who give you too much information, who write angry notes in the margins. You can practice your script, you can refine your questions, you can pick a good spot and a good time—and still, the best answer might be from a five-year-old who just says “nuggets.”
As parents, we want to jump in and fix things. I wanted to wave my hand and say, “Here’s how you write a proper survey.” But what she needed was the mess. She needed to feel the awkwardness of approaching a stranger, the sting of rejection, the strange gift of a random person telling you her Saturday routine. She needed to learn how to adapt. The B- didn’t reflect that. But the experience did.
We still go to McDonald’s now and then. My daughter sometimes points out people and guesses which survey box they’d check. She’s half-joking, but I also notice she watches how people order—how some say “the usual” and some study the menu like it’s a foreign language. She’s not doing a research project anymore. She’s just paying attention.
I don’t know what that paper would have looked like if I had just told her how to do it. Maybe she’d have gotten an A. But she wouldn’t have learned why you leave space in your survey for the answer you didn’t think of. She wouldn’t have learned that some people don’t want to be studied, and that’s fine. She wouldn’t have learned that data is just a collection of stories, and some are messier than others.
Her next assignment is a history research paper. She hasn’t mentioned any surveys. But she did ask if she could interview my grandmother about growing up in a different city. I told her that’d probably be easier than the McDonald’s crowd. She said, “Yeah, but I might have to use follow-up questions.” That sounds about right.
Please indicate: Thinking In Educating » My daughter came home from school with a research paper assignment that felt dangerously open-ended