"round paused so people can play their hearts since the site was down etc etc we go thru this every round at least once now"
do you ever think when you type???? the round lasts 10 days max, the worst that can happen if you veg hearts is you miss out on a 10 day run that you can start over. not only that - hearts can be played on the next day (5 + 5). If the player didnt play his hearts for full 24 hours, chances are he does not care about the current run. this is the dumbest reason to pause the round, it pisses off 95% of the players while you try to please the 5%. demod the idiots who keep pausing rounds for this stupid reason.
ooh so to prevent bias they like, add what they're searching for in the registration or sign-up or whatever, so people don't realize they're doing the poll so they're less likely to skew the results? neat!
deletedover 6 years
ok totally serious but is this like in high school where in science class i would do 5 trials and throw out one if it caused my data to be incorrect or not a small enough standard deviation
yeah i do that few weeks in a year when i get unbanned. dont see anything bad about it. anyways im talking to a wall here. you are all idiots sorry Ginga to break your bubble but you are dumber than Ally
Some researchers in the academic community are adjusting their results, during their analysis (post-experiment), to show their own desired findings. These findings are known as "statistically significant.
In order to find "statistically significant" results, one must have measured their p-value (probability their results are not due to chance) is less than or equal to .05 (the numerical value is actually another debate, but I will not get into that).
So, after the experiments are conducted, some researchers are finding things like p > .05. This is not what they want, so they start to exclude certain trials and participants based on reasons unrelated to the study (i.e participant went through the trial too fast, even though timing wasn't measured). Eventually, this lowers the p-value to p < .05. Researchers think that statistically significant findings will get them published. Sadly, this is true, as there is a bias toward those findings.
One way to combat this is through pre-registering what you are measuring BEFORE the experiment is conducted, and having other non-related researches to give a blind review of the study. Then, BEFORE the experiment, you list why you would exclude participants so you can't make a reason up later.
Example: Amy Cuddy and Power Posing (https://www.nytimes.com/2017/10/18/magazine/when-the-revolution-came-for-amy-cuddy.html)
From the article: "Simmons called those questionable research practices P-hacking, because researchers used them to lower a crucial measure of statistical significance known as the P-value."
Thank you for letting me educate you guys on research!!!!!! Have fun!!!!!!!!
okay so you're playing for 20 hours a day. you are neglecting actually having friends and family to play epicmafia for 20 hours a day. i'd like to let that sink in. i have no further questions, your honor.
i wont have time to argue when i get unbanned, because i will have to actually play games, and i want rules to be sorted before i get unbanned. does that not make sense to you??
I DON'T KNOW ENOUGH ABOUT THE SUBJECT BUT I'M VAGUELY INTERESTED, BIAS IN RESEARCH AND POLLS IS A VERY INTERESTING TOPIC, PARTICULARLY WHEN YOU LOOK AT POLITICAL POLLS
IF YOU ARE INTERESTED, ALLOW ME TO ELABORATE ON P-HACKING AND WHY RESEARCHERS ARE UNFAIRLY ADJUSTING THEIR RESULTS TO HAVE DESIRABLE, BUT MOSTLY UNTRUTHFUL, FINDINGS IN THEIR RESEARCH
In order to receive a (still imperfect) best-case sample size when conducting any form of survey and analysis, you must first produce a method of randomness; this eliminates researcher bias. You are reaching a point of bias when you handpick 100 people that you know, and, therefore, your survey is invalid and will not be accepted by the research community.
