Hacker News Reader

Comment by petercooper | Hacker News Reader

I got put off science at high school when I'd get marked down for having "wrong" but truthful results to physical experiments, so I ended up going around and averaging out other people's results (not their conclusions or write-ups, just the raw numerical measurements) which yielded better marks. I wonder if an element of that is at play in corners of the broader scientific community.

View on HN

High school physics experiments are somewhat of a different case, as "wrong but truthful" results are probably evidence of an incorrectly performed experiment, performing which correctly is part of what one is assessed on. That, or bad luck, but presumably the experiments are sufficiently simple that performed correctly they all but guarantee a "correct" result.

i guess it depends how complex the experiments are at your particular high school, but in my experience we did not have terribly precise instruments or techniques, so the labs would have a high degree of random error. since the lab usually had to fit into 75 minutes, the procedure usually didn't call for many trials. labs where we would perform 5 or more (up to twenty in one lab) trials for each step in college might have 3 or even just one trial in high school.

if you have poor quality equipment and very few trials, the results are going to be all over the place. if you average the whole class's values, you might actually get close.

The time I learned how important it is to forge numbers is when we were measuring gravity in high school by rolling a ball off the end of a table and using a stopwatch to time how long it took to fall to the ground by starting it when the ball fell off the end and stopping it when it hit the ground.

To free fall one meter on Earth is about .45 seconds.

If you are systematically off and get .35 seconds, you get a gravity of 16m/s^2. If you are systematically off and get .55 seconds, you get a gravity of 6.5m/s^2. It isn't that hard to be systematically off in your button presses by that amount even in ten trials. The noise is huge compared to the signal, and averaging several attempts can only smooth out random noise, not systematic miscalculations of when the ball leaves the table or lands.

I got the second result. I got a bad grade. I learned my lesson. Remarkably every subsequent experiment I ever ran was within ten percent of the real value. My turnaround was truly amazing.

Manually triggered stop watches are a terrible way to measure gravity. Human error is going to dominate the result.

Reply on news.ycombinator.com