gettingbye
New Member
- Joined
- Apr 24, 2012
- Messages
- 1
I've been searching for a straightforward formula for my situation, but nothing quite fits yet.
I have a tab ("raw data") with a list of names (1,000+ rows) and data points:
Jamie Park 1,230
Rob Smith 745
Jamie North 340
Jamie K Park 1,100
Rob Smith 2,502
Jamie Drake 950
Glenn Scott 853
Jamie K. Park 0
Jamie North 650
Stacy Kern 2304
Mary Ross 921
Jamie Park 253
Glenn Scott 640
Jamie Kox Park 1,420
I am using another tab ("analysis") to tally up my results.
Objectives:
Extract the average for each person listed on my Analysis tab.
Include only non-zero values
(if possible) Exclude datapoints are are erroneous. And erroneous would be something outside of a defined standard diviation.
Example:
So, let's take "Jamie Park" for example. She appears 5 times. I will need the formula to look at the search term on my Analysis tab, and then search the list on the Raw Data tab. For argument sake, let's say she is the only "Park".
Once the formula can identify the rows containg her name, I need it to look at the column containing the data (1230, 1100, 0, 253, and 1420). The rows reporting zero can be ignored. It would also be amazing if there was a way to exclude the '253' value, since it will not accurately reflect her "average" data. (yes, the data I've been given IS a nightmare)
Current Process:
Looking at my Analysis sheet, I have the following:
A1 = "Park"
B1 = [formula looking at A1, checking list on Raw Data tab and returning filtered average]
I'm using the string "*"& A1 &"*" to help my formula find all the cells in that range that contains "park".
I've used sumproduct in other situations to check a condition and have it sum the remaining results (which could be easily averaged with a count function). However, I'm finding it difficult to have the condition say, 'return True only if the name is "Park" and the data in that row is non-zero'. Plus, it would save me hours if there was a way to "AND" the non-zero check with some kind of standard diviation trick.
I saw some formulas online using a isnumber/search function to find the rows with the applicable values. Then there's if/countifs and the index/match method.
I freely admit that I have very little background in statistics. So, trying to accomplish the last part of this problem seems impossible to me! I'd like the formula to look at Jamie's numbers and know that the 253 value is too high/low from the rest of the numbers.
In another example, person x could have (45, 60, -15, 459, 10, -27, 36, 0, 12). I still don't want it to count the zero, and the crazy number here is the 459. That means, I can't just provide a static range to exclude. Sucks, I know.
My head is truly spinning at this point and would love some help! Thanks in advance.
I have a tab ("raw data") with a list of names (1,000+ rows) and data points:
Jamie Park 1,230
Rob Smith 745
Jamie North 340
Jamie K Park 1,100
Rob Smith 2,502
Jamie Drake 950
Glenn Scott 853
Jamie K. Park 0
Jamie North 650
Stacy Kern 2304
Mary Ross 921
Jamie Park 253
Glenn Scott 640
Jamie Kox Park 1,420
I am using another tab ("analysis") to tally up my results.
Objectives:
Extract the average for each person listed on my Analysis tab.
Include only non-zero values
(if possible) Exclude datapoints are are erroneous. And erroneous would be something outside of a defined standard diviation.
Example:
So, let's take "Jamie Park" for example. She appears 5 times. I will need the formula to look at the search term on my Analysis tab, and then search the list on the Raw Data tab. For argument sake, let's say she is the only "Park".
Once the formula can identify the rows containg her name, I need it to look at the column containing the data (1230, 1100, 0, 253, and 1420). The rows reporting zero can be ignored. It would also be amazing if there was a way to exclude the '253' value, since it will not accurately reflect her "average" data. (yes, the data I've been given IS a nightmare)
Current Process:
Looking at my Analysis sheet, I have the following:
A1 = "Park"
B1 = [formula looking at A1, checking list on Raw Data tab and returning filtered average]
I'm using the string "*"& A1 &"*" to help my formula find all the cells in that range that contains "park".
I've used sumproduct in other situations to check a condition and have it sum the remaining results (which could be easily averaged with a count function). However, I'm finding it difficult to have the condition say, 'return True only if the name is "Park" and the data in that row is non-zero'. Plus, it would save me hours if there was a way to "AND" the non-zero check with some kind of standard diviation trick.
I saw some formulas online using a isnumber/search function to find the rows with the applicable values. Then there's if/countifs and the index/match method.
I freely admit that I have very little background in statistics. So, trying to accomplish the last part of this problem seems impossible to me! I'd like the formula to look at Jamie's numbers and know that the 253 value is too high/low from the rest of the numbers.
In another example, person x could have (45, 60, -15, 459, 10, -27, 36, 0, 12). I still don't want it to count the zero, and the crazy number here is the 459. That means, I can't just provide a static range to exclude. Sucks, I know.
My head is truly spinning at this point and would love some help! Thanks in advance.