“Teach me,” she said.

[Cecily]

Hello and welcome to my thread where you lovely people get volunteered to fill in the gaps of my education. How this works is I ask a question, probably math / statistic oriented and what you do is helpfully explain it to me, ideally, or try really hard to troll me. Haha you... you devils you.

Anyone else feel free to contribute academic-related questions. Please avoid topics on gender studies, feminism, or politics as they already have their own threads.

[America]

pose me a question i'm spinning up an "accomplished academic" trolling angle

[Cecily]

So to start this off, I’ve taken a statistics class years ago at a 4y college and all I learned was how to input numbers into a program. I’d really like to relearn the useful information to help me understand research journals, but for now I have a question about calculating percentile rank.

I have data for 5th%, 95th%, mean, and total number.

Can I use this information to plug in a number to determine its percentile rank?

Ex:
My weight = 196 lbs
Mean = 153.2 lbs
5th percentile = 105.5 lbs
95th percentile = 225.6 lbs
N = 8764

How do I figure out my rank?

Using data from this table:
http://andreaportman.tripod.com/averages.html

[America]

that website is ADORABLE

[loramin]

Originally Posted by Cecily [You must be logged in to view images. Log in or Register.]

So to start this off, I’ve taken a statistics class years ago at a 4y college and all I learned was how to input numbers into a program. I’d really like to relearn the useful information to help me understand research journals, but for now I have a question about calculating percentile rank.

I have data for 5th%, 95th%, mean, and total number.

Can I use this information to plug in a number to determine its percentile rank?

Ex:
My weight = 196 lbs
Mean = 153.2 lbs
5th percentile = 105.5 lbs
95th percentile = 225.6 lbs
N = 8764

How do I figure out my rank?

Using data from this table:
http://andreaportman.tripod.com/averages.html

I think you can't, because you don't have enough info. But stats isn't my forte so hopefully someone else has a better answer.

The way I see it ... the mean is 153.2 lbs right? That could be because there was one insanely fat fuck on the heavy end, or it could be a lot of "on the heavier side but not that heavy" people on that end. Same deal on the other end, you could have like three bulimic people who weigh as much as a concentration camp survivor, or you could just have a lot fairly fit people. Either case could give you that same mean of 153.2 but your placement will be different depending.

The only way to truly know where you place in a dataset is to have that whole dataset. That being said, with the percentiles I would imagine you could guess pretty close to the correct answer by using math ... but again stats aren't my strong suit so someone else would have to provide that math.

[Cecily]

Literally one of the websites 19yo me read for research back when I played EQ.

[America]

Originally Posted by Cecily [You must be logged in to view images. Log in or Register.]

Literally one of the websites 19yo me read for research back when I played EQ.

your old [You must be logged in to view images. Log in or Register.]

[Lune]

Originally Posted by loramin [You must be logged in to view images. Log in or Register.]

I think you can't, because you don't have enough info. But stats isn't my forte so hopefully someone else has a better answer.

The way I see it ... the mean is 153.2 lbs right? That could be because there was one insanely fat fuck on the heavy end, or it could be a lot of "on the heavier side but not that heavy" people on that end. Same deal on the other end, you could have like three bulimic people who weigh as much as a concentration camp survivor, or you could just have a lot fairly fit people. Either case could give you that same mean of 153.2 but your placement will be different depending.

The only way to truly know where you place in a dataset is to have that whole dataset. That being said, with the percentiles I would imagine you could guess pretty close to the correct answer by using math ... but again stats aren't my strong suit so someone else would have to provide that math.

This is correct, the distribution could be bimodal, steep, shallow, you don't know, and your rank/percentile depends on that. Think you need access to the data set, and with N=thousands, a computer/spreadsheet to calculate this, because it involves ranking the data.

Even if you had just the standard deviation you could assume a normal curve and estimate your percentile based on how many standard deviations you are from the mean, but you don't have that.

[Cecily]

Naw. I’m not in a new age demographic bracket till April. Nice thing is I blend in with college kids still. It scares me how roughly 30+ hits a lot of people. My mom looks really good for being close to 60, so I think I’m going to be very fortunate in that regard.

Originally Posted by loramin [You must be logged in to view images. Log in or Register.]

I think you can't, because you don't have enough info. But stats isn't my forte so hopefully someone else has a better answer.

The way I see it ... the mean is 153.2 lbs right? That could be because there was one insanely fat fuck on the heavy end, or it could be a lot of "on the heavier side but not that heavy" people on that end. Same deal on the other end, you could have like three bulimic people who weigh as much as a concentration camp survivor, or you could just have a lot fairly fit people. Either case could give you that same mean of 153.2 but your placement will be different depending.

The only way to truly know where you place in a dataset is to have that whole dataset. That being said, with the percentiles I would imagine you could guess pretty close to the correct answer by using math ... but again stats aren't my strong suit so someone else would have to provide that math.

Thank you, Loramin! I was wondering if lacking the whole range would be a problem. I’m considering outliers when I check my results vs the ones on the table. Like 95 for waist is around 46” and I looked at that in the mirror with measuring tape and was sad people have to deal with that

[Cecily]

Originally Posted by Lune [You must be logged in to view images. Log in or Register.]

This is correct, the distribution could be bimodal, steep, shallow, you don't know, and your rank/percentile depends on that. Think you need access to the data set, and with N=thousands, a computer/spreadsheet to calculate this, because it involves ranking the data.

Even if you had just the standard deviation you could assume a normal curve and estimate your percentile based on how many standard deviations you are from the mean, but you don't have that.

Ty Lune!

She did cite the research she took the data from so I might be to dig a little deeper and get that information. Seems like a fun project to get the hang of stats again.