3 Ways to Use Factor Average
This article is part of a series that focuses on report building for efficiency. The series, prepared by Daniel Landelius, Customer Support Specialist, highlights different functions in the system and how to best utilize them when setting up your project. The ultimate goal is for you, the project administrator, to save time and cut down manual work.
In our last article, we went over NPS calculation and mentioned the use of factor average. Now, we’ll dive a bit deeper into factor average, as this function can be used for more than to calculate NPS.
Before we get started, in case you aren’t familiar with factor average, it is a numerical value that can be assigned to every answer category of a single response question. We’ll look at 3 different cases for using it. Users have likely come across the first two cases quite often. While the last case is an example of how factor average can be used to complete calculations that initially seem impossible.
The three cases we’ll detail are how to factor average to:
- 1. Change the value order in mean value calculations
- 2. Show a numeric value in a respondent table
- 3. Take your benchmarking one step further
1. How to use factor average to change the value order in mean value calculations
This is the most obvious use of factor average. In the previous article, the values were changed to -100, 0, and 100 in order to change the mean value calculation to an NPS calculation. No limitations exist in terms of which factor average numbers you can switch your original numbers to.
Let’s say, for example, you simply want to switch your scale so 1 is the highest value and 5 is the lowest in the calculation.
Image 1: Answer block with original Factor Average numbers.
To do that, you simply manually change the numbers in the factor average box to reflect the new values.
Image 2: Answer block with Factor Average numbers set to a reversed scale.
2. How to use factor average to show a numeric value in a respondent table
This example, technically, isn’t a factor average function – rather, it’s a respondent table function. Occasionally, you have an answer block with a descriptive rating scale, but in the respondent table, you’d prefer to show just the numerical value.
Achieving this is simple. First, make sure that the factor average is showing the values you want to display. Then, under the Format section select “factor average” instead of “answer text” from the dropdown. Ta da!
Image 3: Format dropdown in Respondent table.
3. How to use factor average to take your benchmarking one step further
In this last example, we’ll highlight how to use factor average to your benefit when you want to do more complex benchmarking.
Comparing positive vs. negative responses:
Follow along by looking at the three tables on Image 4 below.
Table 1: Let’s assume that you do an annual study in which you ask your customers how positively or negatively they perceive an aspect of your company. This is done with a scale ranging from “very negative” to “very positive” as the answers appear in the Original values percentage table below.
Table 2: In order to get an average of the overall perception, we want to solely look at the negative and positive experiences, and leave the neutral ones out. To do this, we create a grouped answer called Negative (including all “very negative” and “somewhat negative” responses) and Positive (“very positive” and “somewhat positive” responses) from the answers.
Table 3: Then, we calculate a benchmark between both of these values. In our example, the benchmark between Positive and Negative gives us 0.54% fewer positive answers for the year 2015 than negative answers.
Image 4: Comparing positive vs. negative responses
This solution works if you want to compare results within just 2015. However, you won’t be able to benchmark the result between years since you are using the benchmark function to compare your values within a single question. To solve this, you can use factor average to do the calculation above without involving the benchmarking function. Doing so allows you to then add the benchmark function to compare the results between years.
Comparing results between years:
Image 5 below will help you follow along.
Table 1: Give your negative answers -100 in factor average, your neutral value 0, and your positive values 100. Set up your table to show the original percentages for both years.
Table 2: Set up your question as a mean value question instead of a percentage question. As you can see, the calculation will result in the same numbers as in Table 3 in Image 4 above.
Table 3: Calculate the mean benchmark between the year 2014 and the year 2015. You will see that 2015 had 3.06 fewer positive answer points than 2014 – implying a better performance in 2014 than in 2015.
Image 5: Comparing results between years
This series of articles tries to reveal hidden gems and shortcuts you can use within the system to make your reporting really stand out! If you have any questions, feedback, or ideas on topics you’d like for us to cover, please don’t hesitate to email Daniel directly at firstname.lastname@example.org.