Say you have a survey, where you ask more than one question. For example, for each person you ask,
1. What did you like about your stay?
2. What did you not like about your stay?
Then you would get multiple columns of open ended questions with one row per person.
The Explorer helps you find the common Topics in text data. As it depends on why you are asking multiple questions in the first place, and if you would like multiple perspectives on different things. If you would like to find the common topics for Answers #1. Then you could upload them and create a project called Answers #1. And similarly for Answers #2. We can suggest simply making two projects.
If you are thinking of combining the two text answers into one text then there is a challenge with combining text which means you might not know which Topics refer to answering which question. For example, with the two above questions, if you got the response Topic: Bathroom frequently, it is difficult to understand if that Topic was answering the positive or the negative question. So it is not insightful to extract Topics when you have combined the text to multiple questions. However, instead of actually combining both texts into one cell, you could do some Excel Transformations to upload both questions in the same project.
The benefits of doing a transformation instead of a combination:
- Easy Comparison between answers using the Filter Function
- Common Topics between both answers. That is, what is most top of mind overall
- Easy Excel Exports. That is, you just have to Export Once
The negatives are:
- You might be making Topics that only make sense for one Set of Questions and do not make sense for another. Thus, you could be missing Topics that are not that prevalent.
- This is assuming you have very similar answers across the different questions. You might need to do separate projects anyways.
Recall that The Explorer only analyzes one column. What we can do is take both columns and stack each of them on top of each other. But, to make sure a new column is added that says the Question Tag. This allows you to upload one file to The Explorer with one column. The Explorer takes credits based on how many rows are in the Excel file so adding more columns is okay. Having an extra column lets you to filter through each of the answers in addition, to be able to see individual answers. The only draw back is to remember that it is all stacked so when a Topic is showing a Count and a Percentage, those are respective of the entire set of all answers stacked together. Which you would then have to either Filter in The Explorer or Use Excel to cut the data.
So what you would want to do for this example is
- Insert a new Column beside B and name it Question Tag
- Fill the column to the last row that "What Did you Like" occurs with the value "What Did you Like?"
- Select cells from the column "What Did you not like?" and then all the way down to the last cell of the Column "What Did you not Like?".
- Cut the cells. The hotkey is Control X or Command X. Note if there are other columns that represent data about one row over all the questions then it is good to save the ordering of that and try and Cut out that data as well. So for example in below there is a Grade column which applies to the entire row. So when you cut out the column "What Did you not like", re-Copy the column Grade and paste that as well. That way the value for grade is in the same row as the original "Like" and "Dislike" answer.
- Go to the bottom of column b which is the last cell of "What Did you like?" and paste the cells. The hotkey is Control V or Command V
- Go back to the new created column called Question Tag and then fill the cells with the value "What Did you not like"
- Repeat for all the columns of open ended questions
- Now you should have something that looks like below.
1. This is what you have.
What we can suggest is to stack the answers and add another Tag. Note that the Grade and ID column were preserved across the Question Tag Rows so they repeat in the same column.
