Editing Talk:1877: Eclipse Science
Please sign your posts with ~~~~ |
Warning: You are not logged in. Your IP address will be publicly visible if you make any edits. If you log in or create an account, your edits will be attributed to your username, along with other benefits.
The edit can be undone.
Please check the comparison below to verify that this is what you want to do, and then save the changes below to finish undoing the edit.
Latest revision | Your text | ||
Line 29: | Line 29: | ||
*There is a tendency among non-scientists to underestimate the certainty of scientific or statistical results. Some examples are "you can't teach evolution as settled science because it's only a ''theory''" and more recently "this poll only sampled a thousand adults, which is 0.03% of Americans, so it's meaningless because we have no idea what the other 99.97% think". So scientists are expected to keep verifying and re-verifying fairly basic results. [[User:D5xtgr|D5xtgr]] ([[User talk:D5xtgr|talk]]) 16:29, 21 August 2017 (UTC) | *There is a tendency among non-scientists to underestimate the certainty of scientific or statistical results. Some examples are "you can't teach evolution as settled science because it's only a ''theory''" and more recently "this poll only sampled a thousand adults, which is 0.03% of Americans, so it's meaningless because we have no idea what the other 99.97% think". So scientists are expected to keep verifying and re-verifying fairly basic results. [[User:D5xtgr|D5xtgr]] ([[User talk:D5xtgr|talk]]) 16:29, 21 August 2017 (UTC) | ||
:That last one is hilarious, for reasons I doubt you realize. Some of us here know how to get a proper population sample--and, well, with that poll? I don't even know where to ''start'' on the potential sources of error. 'Too small' should appear on the list several times, though, since a lot of the sources of error can be controlled for by simply adding to your sample. (Yes, yes, changing your sampling techniques will also do it, but it's just plain '''''easier''''' to add more people, so you start there.) This is actually part of why being able to replicate results matter--plus, the training in how to do it properly does give you a pretty good idea how to rig your poll to produce the desired results, and what to watch for when reading others' research. [[Special:Contributions/108.162.237.130|108.162.237.130]] 21:39, 24 August 2017 (UTC) | :That last one is hilarious, for reasons I doubt you realize. Some of us here know how to get a proper population sample--and, well, with that poll? I don't even know where to ''start'' on the potential sources of error. 'Too small' should appear on the list several times, though, since a lot of the sources of error can be controlled for by simply adding to your sample. (Yes, yes, changing your sampling techniques will also do it, but it's just plain '''''easier''''' to add more people, so you start there.) This is actually part of why being able to replicate results matter--plus, the training in how to do it properly does give you a pretty good idea how to rig your poll to produce the desired results, and what to watch for when reading others' research. [[Special:Contributions/108.162.237.130|108.162.237.130]] 21:39, 24 August 2017 (UTC) | ||
β |