I'm sitting in a meeting for my kids' school... they did a parent survey. They only got 160 responses out of 350 families... To me that seems like a REALLY HEALTHY sample... and I'm no statistician - but I think 45% is pretty damn skippy good.... and might even be called "statistically significant" sample. The High School science teacher in the room flat out disagreed with me and told me I wasn't being mindful of the families that didn't respond and that I was trying to silence them. He then likened our survey to the US census and tried to make it sound like we're trying to oppress some families.... Maybe my basic statistical sense is off... but a 45% response rate is PRETTY DAMN GOOD... and science man can f$%& off back to the hippie hole he crawled out of.
I'm doing this thing from the Society of Manufacturing Engineers called Tooling U. Before you start the chapter you have to take a pretest. The pretest can run to 20 questions. Tooling U loves to point out how much better everybody does after they do their study program because the end-of-chapter tests are usually a couple points better than the pretests. They don't like to point out that we often hit SKIP on eighteen fucking bullshit questions because goddamnitletmegetonwithmyliiiiiiiiife. Was their survey an email? Did it come from SurveyMonkey? If so, 20% is fan-fucking-tastic. 46%? That's positively North Korean. If the science teacher wants a better response rate, he should go door-to-door to grok why we only conduct the census every ten years.
Fuck him. He doesn't live in the real world. Out here in industries where we use EXTENSIVE surveying to measure everything from the efficacy of advertising, to determining entire political platforms that will dictate how the lives of 360 million Americans are going to proceed... We know that a 4% response rate to ANY targeted survey is a HUGE win. Four percent. Wanna know how many responses a politician has to get to be "statistically significant" and to induce the fucker to get off his ass and create some new legislation. ONE HALF PERCENT. There are 2.1 million people in my county. If I can get 1000 of them to sign a petition, my legislator will JUMP exactly as high as I ask him to. That's 0.0004%. Mr Teacher is a clueless hack. You know why people didn't respond? They DIDN'T CARE. You know whose responses you actually WANT, and are VALUABLE? THE ONES WHO CARE ENOUGH TO RESPOND. He should be fired. He can't be trusted around children.
Do kids still have school bags or just digital school bags?they did a parent survey
Did the kids have to bring home a form telling the parents to go on line and do a parent survey? If kids have to bring anything home to give to the parents, 55% of the notices are still at the bottom of their school bags.
Got back to Canada and tried a different power cord. Wow... it worked. so I didn't have to spend a day trying to convince DELL to send me a guy. Heading to NYC tomorrow. When will I be distractable enough to write something on the CURRENT pubski. I have SOOOO much to chat about in the pub. xxoo
Doesn't it depend on how confident you want to be that your sample response is representing the population? I. E. Alpha = 0.95 vs. Alpha =0.98?
What's the subject of the survey? Is it something like changing like which children will be eaten to feed the others or minor like whether they should play baseball or soccer this spring? For mundane subjects, I think that's a healthy sample.
"The teachers at the school care about me" "I feel like I belong at school" "My schoolwork is meaningful to me" "I feel safe at school" etc. According to one online calculator... (yes, I have wasted WAY too much time on self-vindication tonight) with a population of 350, 161 responses gives me a 7.5% margin of error with a 99% confidence level. but what do I know...