July 02, 2018, 13:10  #1 
Arrogant Bastard
Gold Contributor
FALaholic #: 96 Join Date: Jul 2000
Location: Surprise, AZ
Posts: 25,140

Reference  FAL  IMBEL Locking shoulder Size Distribution
moved to Ref & Data
So I was sorting locking shoulders and thought, it would be cool to calculate the average size and distribution of inservice IMBEL rifles. It's been almost 30 years since I took graduate statistics and math has never been my strong point. I have around 600 that I hadn't included in kits and with my summer help back in school  they weren't going to sort themselves. I set aside all that were "new" and counted only those I pulled from receiver sections (yeah, that was a chore!) I discarded all damaged ones. I discarded or set aside any that had been modified This is my distribution entered into Excel size Qty (fx size x qty) 0.268 12 3.216 0.267 17 4.539 0.266 20 5.320 0.265 34 9.010 0.264 74 19.536 0.263 59 15.517 0.262 91 23.842 0.261 61 15.921 0.260 69 17.940 0.259 45 11.655 0.258 45 11.610 0.257 29 7.453 0.256 38 9.728 0.255 21 5.355 0.254 15 3.810 0.253 4 1.012 Total 634 fx sum of size x qty = 165.464 fx mean (total size / total qty) 0.2610 median mode 262 (that's easy) range .015 Conclusions  .261" is the average (mean) size; .262" is the most frequently occurring size (mode) .264" is the second most frequently occurring size .260" is the third most frequently occurring size .263" is an anomaly making a nonstandard distribution How do I calculate the median using excel without entering 634 separate values? can I also calculate standard deviation without entering 600 data points?
__________________
T. Mark Graham, Master Gunsmith Arizona Response Systems, LLC 
July 02, 2018, 14:41  #2 
Registered
Bronze Contributor
FALaholic #: 34604 Join Date: Feb 2008
Location: SA Tx
Posts: 14,127

Your median is the average of values 316 and 317. Since .261 occurs from a low of 267 counts and a high 327. .261 is the median. On a spread sheet you could just add one box to the previous to find the values where the count lies between 316 and 317
__________________
Giving people the benefit of the means it’s ok to screw you over. Expect the worst, and you'll never be disappointed. Before trying to beat the odds, make sure you can survive the odds beating you. 
July 02, 2018, 19:28  #3 
Registered
Bronze Contributor
FALaholic #: 34604 Join Date: Feb 2008
Location: SA Tx
Posts: 14,127

I don’t understand your method if you want a decent approximation iof standard deviation. I would take the data as a percentage of the population and then run the STD on the resultants you got away from the normalization of the units. 1 is always the number statistics are an estimation. That data is fairly normal. 1 standard deviation is 62% 2 is 97.5 and 3 is 99.7. Divide by the average and you get the variance of which plus or minus is what you see 62% of the time
__________________
Giving people the benefit of the means it’s ok to screw you over. Expect the worst, and you'll never be disappointed. Before trying to beat the odds, make sure you can survive the odds beating you. Last edited by meltblown; July 02, 2018 at 19:41. 
July 15, 2018, 19:42  #4 
Registered
FALaholic #: 28033 Join Date: Apr 2007
Location: USA
Posts: 139

My experience with DSA is on the low end (.249  .253)
How do Coonan Receivers compares? Last edited by Chester Nimitz; July 15, 2018 at 19:50. 
Thread Tools  
Display Modes  

