Forum Thread

Using Excel to for survey score analysis

Ram|bunc|tious 24,564 1,944 October 23, 2015 at 01:33 PM
For a youth sports team, I sent out a survey form (ratings 1-5) and received a decent number of responses for feedback. The survey results are broken up by age group so what I want to do is compare results across age groups. For a single result, ie "On a rating of 1-5, score ..." what is the most useful measuring method? (ie its not the average or median score, it may be combination of tossing out the outliers)

Best I could find online was: http://www.measuringu.com/blog/in...ponses.php

21 Comments

1 2

Sign up for a Slickdeals account to remove this ad.

#2
It sounds like you're asking for statistics help rather than excel help.

I know enough about statistics to know that I don't know enough about statistics. What I'd probably start doing is plotting a histogram of the responses so that I could look at the distribution. I think that will give you the quickest idea of what's going on. This is with the obvious caveat that these are the people who chose to respond to your survey.
Reply Helpful Comment? 0 0
#3
I guess I would ask though, what is an outlier in this case? Your responses are bounded between 1 and 5.
Reply Helpful Comment? 0 0
Joined Feb 2004
L10: Grand Master
6,953 Posts
944 Reputation
#4
Quote from Ram|bunc|tious View Post :
For a youth sports team, I sent out a survey form (ratings 1-5) and received a decent number of responses for feedback. The survey results are broken up by age group so what I want to do is compare results across age groups. For a single result, ie "On a rating of 1-5, score ..." what is the most useful measuring method? (ie its not the average or median score, it may be combination of tossing out the outliers)

Best I could find online was: http://www.measuringu.com/blog/in...ponses.php
Sounds like any of those mentioned in your link would work...depends on what the respondents were scoring and what your goal is with the survey.
Reply Helpful Comment? 0 0
Marshall: Have the rest of you guys figured out by now that mmathis is the smartest guy on SlickDeals?
#5
Reply Helpful Comment? 0 0
Joined Jul 2009
L10: Grand Master
9,300 Posts
2,205 Reputation
#6
Is this for your personal amusement or are you presenting the information? If so, who is the audience? How much time do you have? What is the purpose, or do you have an agenda?
Reply Helpful Comment? 0 0
Joined May 2004
You like that!
24,564 Posts
1,944 Reputation
Original Poster
#7
Quote from quotidian View Post :
I guess I would ask though, what is an outlier in this case? Your responses are bounded between 1 and 5.
Weird, guy on our board is manager at Macys and he says they way they do it for customer service is subtract top 1 and bottom 3 ... but that just leaves the 2nd highest score. Checking on clarification, that makes little sense.

Quote from quotidian View Post :
It sounds like you're asking for statistics help rather than excel help.

I know enough about statistics to know that I don't know enough about statistics. What I'd probably start doing is plotting a histogram of the responses so that I could look at the distribution. I think that will give you the quickest idea of what's going on. This is with the obvious caveat that these are the people who chose to respond to your survey.
Ugh, scars from statistical analysis class in college have faded but the night tremors remain.

$20? You new to SD? Wink

I think I need to load the Excel Analysis Toolkit Doh
Reply Helpful Comment? 0 0
Last edited by Ram|bunc|tious October 24, 2015 at 08:07 AM
Joined May 2004
You like that!
24,564 Posts
1,944 Reputation
Original Poster
#8
Quote from Foreveryours View Post :
Is this for your personal amusement or are you presenting the information? If so, who is the audience? How much time do you have? What is the purpose, or do you have an agenda?
Purpose is to identify areas of strength and those for improvement across our program. The survey was designed such that parents would provide feedback for coaches on a age group basis. I want to share those results with our coaches and use it as a learning tool.

May be tough based on limited feedback. We have 110 kids playing across 4 age groups and a total of 45 surveys. The smallest age group only returned 4 surveys, but that's mine so I'm not as concerned Wink
Reply Helpful Comment? 0 0

Sign up for a Slickdeals account to remove this ad.

#9
Quote from Ram|bunc|tious View Post :
$20? You new to SD? Wink

I think I need to load the Excel Analysis Toolkit Doh
I'm not advocating any paid tools suggested at the link. An open source statistics package could be an option.

Did you give out paper surveys or email our a survey using survey monkey, google forms, or other similar tool?
Reply Helpful Comment? 0 0
Joined May 2004
You like that!
24,564 Posts
1,944 Reputation
Original Poster
#10
Quote from jkee View Post :
I'm not advocating any paid tools suggested at the link. An open source statistics package could be an option.

Did you give out paper surveys or email our a survey using survey monkey, google forms, or other similar tool?
I set it up with Google Surveys, great little tool. Provides a nice summary for the overall program, just trying to divvy it up by age group (first question was "what was your age group")
Reply Helpful Comment? 0 0
Joined Jul 2009
L10: Grand Master
9,300 Posts
2,205 Reputation
#11
Quote from Ram|bunc|tious View Post :
Purpose is to identify areas of strength and those for improvement across our program. The survey was designed such that parents would provide feedback for coaches on a age group basis. I want to share those results with our coaches and use it as a learning tool.

May be tough based on limited feedback. We have 110 kids playing across 4 age groups and a total of 45 surveys. The smallest age group only returned 4 surveys, but that's mine so I'm not as concerned Wink
I think it would provide some meaning over long term showing trend.
Reply Helpful Comment? 0 0
#12
You've already measured. The most useful measuring method is likely to be one you did not select (this follows from, Murphy's Law). However, now it's time to interpret the measurements. Mathematically manipulating the numbers (averaging, taking standard deviations, etc...) is not really appropriate - there's no way you can demonstrate that the distance between a 1 and a 2 is the same as the distance between a 3 and a 4 across all respondents. Just create a series of histograms, look at them, and go with your gut.
Reply Helpful Comment? 1 0
Joined Feb 2004
L10: Grand Master
6,953 Posts
944 Reputation
#13
Quote from Ram|bunc|tious View Post :
I think I need to load the Excel Analysis Toolkit Doh
Quote from Ram|bunc|tious View Post :
Purpose is to identify areas of strength and those for improvement across our program. The survey was designed such that parents would provide feedback for coaches on a age group basis. I want to share those results with our coaches and use it as a learning tool.

May be tough based on limited feedback. We have 110 kids playing across 4 age groups and a total of 45 surveys. The smallest age group only returned 4 surveys, but that's mine so I'm not as concerned Wink
I'm a little confused now - do you need help figuring out what info to calculate, or do you need help figuring out how to get that info from Excel?

Not knowing how the survey was worded makes it a little tough to suggest ways of analyzing the resulting data. As mentioned before, histograms of the data would be a good starting point, but if you're looking for something like "x% of the people said you did a good or great job!" or "y% of the people thought you were terrible at teaching the kids how to bunt" we need more info.
Reply Helpful Comment? 0 0
Joined May 2004
You like that!
24,564 Posts
1,944 Reputation
Original Poster
#14
Quote from Foreveryours View Post :
I think it would provide some meaning over long term showing trend.
Totally agree, I hope we can use the survey questions as a basis every year (ie minimal deviation from wording/content)

Quote from BoscoBearbank View Post :
You've already measured. The most useful measuring method is likely to be one you did not select (this follows from, Murphy's Law). However, now it's time to interpret the measurements. Mathematically manipulating the numbers (averaging, taking standard deviations, etc...) is not really appropriate - there's no way you can demonstrate that the distance between a 1 and a 2 is the same as the distance between a 3 and a 4 across all respondents. Just create a series of histograms, look at them, and go with your gut.
Probably right, especially as the sample size is significantly decreased when I look instead of one big group as the 4 separate age groups.
Reply Helpful Comment? 0 0
Last edited by Ram|bunc|tious October 26, 2015 at 08:27 AM
Joined May 2004
You like that!
24,564 Posts
1,944 Reputation
Original Poster
#15
Quote from mmathis View Post :
I'm a little confused now - do you need help figuring out what info to calculate, or do you need help figuring out how to get that info from Excel?

Not knowing how the survey was worded makes it a little tough to suggest ways of analyzing the resulting data. As mentioned before, histograms of the data would be a good starting point, but if you're looking for something like "x% of the people said you did a good or great job!" or "y% of the people thought you were terrible at teaching the kids how to bunt" we need more info.
My fault, sorry to make this a bit misleading.

My preferred calculation tool is Excel, I need assistance trying to figure out how to use Excel to interpret the results (if its reasonably straight-forward). Using a Pivot Chart I can quickly desperate the age groups out (u7, u9, u11, u13) and create histograms. I suppose quick copy & paste into different sheet or Word page to compare against the same question for different age groups.
Reply Helpful Comment? 0 0
Page 1 of 2
1 2
Join the Conversation
Add a Comment
 
Copyright 1999 - 2016. Slickdeals, LLC. All Rights Reserved. Copyright / Infringement Policy  •  Privacy Policy  •  Terms of Service  •  Acceptable Use Policy (Rules)  •  Interest-Based Ads
Link Copied to Clipboard