Thank you for this video. I learned a lot, better than in attending the class I am in. Please don't put down your videos and continue making them like the Lean, Six Sigma, and more of statistics with use of excel as I am using your videos to learn. I am sharing your videos to my classmates too. Really appreciate this!
Thank you for your kind words. I believe my content is of decent quality. I recognize that the quality of the videos themselves (sound, editing, etc), aren't the best on RUclips. I originally created these videos for my students to use to augment the classroom lectures. I wasn't really focusing on the quality of the video so much as the content. Some of these videos I created at 2-3 a.m. 😁 Again, thank you. Your comments humble me and encourage me to create more videos!
Great question! Please forgive my delay in responding. If I understand your question correctly, you can replace the "100" in the denominator of formula with a "3" and replace the "P" in the subscript and the formula with a "T." So, the formula for the LOCATION of the tertile would be L(subscript)T = (n+1)(T/3). Take that location value and apply it the same way I show in the video for the 33rd percentile (which I just realized has said, "33 percentile" instead of "33rd percentile"): a) calculate the location value, b) if the location value is a decimal, calculate the difference between the value of the data point located at the whole number position below and above the location value, c) multiply the difference by the decimal value, and d) add the product to the value at the data point located at the whole number position below the calculated position value...yeah, you're better off watching the video again. If you are looking to just create equal groupings of low, medium, and high within which you want to separate your data, a) subtract the lowest value from the highest value, b) divide that difference by 3, c) add the difference/3 value to the lowest value; the range from the lowest value to the lowest value + (difference/3) is your "LOW" category, d) add the difference/3 value to the highest value of the LOW category; the range from the highest value of the LOW category to the LOW category highest value + (difference/3) is your "MEDIUM" category, and e) add the difference value to the highest value of the MEDIUM category; the range from the highest value of the MEDIUM category to the highest value in your data set is your "HIGH" category. NOTE: you will likely have to adjust these instructions based on the decimal results of the calculations. For example, you really should not have the highest value of the LOW category as the same value as the lowest value in your MEDIUM category; it should be a value greater than that of the highest value in the LOW category. Caveat: as I state in the video regarding the manual method of calculating percentiles, the manual method of calculating tertiles that I just stated is arguably not the most correct method. Another method is to use the percentile function to create proxies for the first, second, and third tertiles. In Excel using the formula below, you could create proxies =IF(DATA POINT
this was uploaded 7 years ago and i never knew this day would come where i'll be needing it. tysm!
You're very welcome, LK!
Thank you for this video. I learned a lot, better than in attending the class I am in. Please don't put down your videos and continue making them like the Lean, Six Sigma, and more of statistics with use of excel as I am using your videos to learn. I am sharing your videos to my classmates too. Really appreciate this!
Thank you for your kind words. I believe my content is of decent quality. I recognize that the quality of the videos themselves (sound, editing, etc), aren't the best on RUclips. I originally created these videos for my students to use to augment the classroom lectures. I wasn't really focusing on the quality of the video so much as the content. Some of these videos I created at 2-3 a.m. 😁
Again, thank you. Your comments humble me and encourage me to create more videos!
Can you please tell how to divide a given data into tertiles? Like into 3 categories - Low,Medium and High
Great question! Please forgive my delay in responding. If I understand your question correctly, you can replace the "100" in the denominator of formula with a "3" and replace the "P" in the subscript and the formula with a "T." So, the formula for the LOCATION of the tertile would be L(subscript)T = (n+1)(T/3). Take that location value and apply it the same way I show in the video for the 33rd percentile (which I just realized has said, "33 percentile" instead of "33rd percentile"): a) calculate the location value, b) if the location value is a decimal, calculate the difference between the value of the data point located at the whole number position below and above the location value, c) multiply the difference by the decimal value, and d) add the product to the value at the data point located at the whole number position below the calculated position value...yeah, you're better off watching the video again.
If you are looking to just create equal groupings of low, medium, and high within which you want to separate your data, a) subtract the lowest value from the highest value, b) divide that difference by 3, c) add the difference/3 value to the lowest value; the range from the lowest value to the lowest value + (difference/3) is your "LOW" category, d) add the difference/3 value to the highest value of the LOW category; the range from the highest value of the LOW category to the LOW category highest value + (difference/3) is your "MEDIUM" category, and e) add the difference value to the highest value of the MEDIUM category; the range from the highest value of the MEDIUM category to the highest value in your data set is your "HIGH" category.
NOTE: you will likely have to adjust these instructions based on the decimal results of the calculations. For example, you really should not have the highest value of the LOW category as the same value as the lowest value in your MEDIUM category; it should be a value greater than that of the highest value in the LOW category.
Caveat: as I state in the video regarding the manual method of calculating percentiles, the manual method of calculating tertiles that I just stated is arguably not the most correct method. Another method is to use the percentile function to create proxies for the first, second, and third tertiles. In Excel using the formula below, you could create proxies
=IF(DATA POINT
Thank u :)
You're welcome. I hope it was helpful!