Including Weighted Averages within your Customer Survey


This video tutorial discusses how to implement weighted averages within your customer survey instrument. The tutorial outlines how to author customer survey questions with weighted averages as well as how the weighted averages are displayed within the survey reports.


In this tutorial we're going to be discussing how to implement weighted averages within your survey design. And also take a look at how that's reflected within the reports.

So here we are in a test account in the SurveyManager. I'm going to jump up here to the Surveys tab, above the main bar, and go to Create New Survey. In fact, I've already created a survey so I'm just going to edit this survey and show you how I created this one question.

So I'm going to go to this sidebar here, drop-own menu, click Go to edit. And you'll see the main screen for authoring your survey. And I've just chosen to use the multiple choice vertical. You can also use different multiple choice or choose All that Apply questions for this feature. And I entered a simple Please Tell Us How You Heard About Us question type and inserted several different options. Recommendation from a Friend, let's assume this is a hotel. Expedia., Previous Customer, Other, Please Explain. And you'll notice here I selected the checkbox adjacent the Open Ended option so that they can explain if they did select Other, Please Explain.

So let's take a look here. I've made this a required response. I'm assuming that this is a question that I don't want my respondents to be able to skip. So this is a totally optional feature. But for this training tutorial, we're focused on weighted averages. So here you'll see down below, Weighted Average for Responses as well as Weighted Average for Questions. So you can use either or both. And in this particular survey question we've implemented Weighted Average for Responses.

And when you check this checkbox, you'll see that we get an option here adjacent, the Each Response. And I can put in whatever weighted average I want. Let's say I wanted highly weight the Recommendation from Friends because I value that more, let's say. And for the third party hotel sites let's give it a ten. A ten, sorry, I forgot the zero there. Ten and ten. Previous Customer, highly valuable. Let's say 50. And you need to enter a value for all the responses in the question. You can't leave one of these blank. And if you do you'll be prompted. But I'll just put a zero in this since I don't know how to weight that or value that response.

And then when I save this question, it will go ahead and continue through the Five Step Wizard here above. You know, you can customize your skin, brand the survey, use any optional features, schedule it for distribution or distribute it immediately. Either as a public or private survey. And, by the way, I created this survey as an online survey but you can also distribute it via our mobile survey app, mobile web for mobile browsers. Even as a kiosk survey, et cetera, since our SurveyManager has and accommodates multiple modalities.

So, okay. We've saved this survey and let's assume the survey's been published and we've, we have some data to analyze and report. So we can jump up to our Reports module. And let's go to Create Report. And this particular survey was using weighted averages. So I'll load that in. And we have 15 responses here. All completed, no incompletes and let's go ahead and just select a Summary report.

And you'll notice we have a checkbox here that appears only when weighted averages have been used within the survey design. And it indicates Show Response Computation. So let's check that checkbox so our report will indeed display the weighted average calculations for us. And I only have one question in this survey for demonstration purposes. We'll leave it as long format, but you can always use tabular or graph format. And let's generate this report.

So here we are. We have our one question. We have 15 of 15 created today. All the responses are from today. And we have an overall satisfaction score of 24.67 percent and we also have an average score of 3.1. Not to confuse the weighted score to the average score. Average is simply, is not taking into account any weighted calculations for each response.

So here we have for question one, answer one, Recommendation from Friend. And you'll notice this is highlighted or made bold because it has the highest responses. And that's just the default feature in our system. But 26.67 percent of the people of the 15 total responses answered question, I'm sorry, response one to question one.

And the average, the weighted average for this is 200. Why? We put in 50, remember, as the weighted score. So 4 times 50 is 200. So this is a calculation column here of the weighted average. This is a percentage, obviously. And this is a frequency column.

And if you were to take the total of these weighted averages and divide it by the total number of completes for this question, you would get the 24.67 percent. Whereas the 3.1 percent is telling you that 3.1 is the average of the total frequency divided by the 15.

So you can always go in and Show Detail which, you know, we had an open ended comment response. So this shows you for Other, Please Explain, the first person said, "Travel Magazine," the second, "Online Ad." So that is the Show Detail for the sixth response to this question.

You can always export a report in any of the formats that are default. You can also export the raw data. You can schedule reports, schedule the raw data, et cetera. So if I just go ahead and export this to an Excel file and open that file up, it then will show me my total scores exactly the way I saw them online. Also question one, any open ended question gets its own tab so you can see your open ended comments. And of course the Raw Data tab.

© 2017 Opinionmeter. All Rights Reserved.     |     Terms of Use     |     Privacy Policy     |     Site Map

Designed By