Last I spoke with Support about this, they had let me know that unfortunately there was no way for the field to back-calculate if it was added as an embedded data field in the survey flow. If you add it as a formula in the Data tab, then it could calculate, but that would be restricted to simple math operations like +, -, power, etc.
The one you found, is it from this link? Here they use a similar regular expression but in the custom validation section. That way you don't have to worry about scripting.https://help.lafayette.edu/custom-validation-in-qualtrics/
You can add more embedded data fields to a survey in progress, yes. The dataset will update to include those fields once you publish the revision, but it will only fill the field for data moving forward. You would need to go back and manually fill in the added embedded data fields for your past data before the revision.Alternatively, if it is a simple calculation you want to add, you can go to your Data & Analysis tab for the survey and choose Create New Field, in which Formula is an option. Here is a support site that describes the math operations you can do, rather simple calculations: https://www.qualtrics.com/support/survey-platform/data-and-analysis-module/data/add-new-fields/formula-variable-creation/Hope this helps! :)
As far as I know, it's not possible since it's not in the API. What you could try doing is collecting the 5000 responses to delete into a list and then loop through each one, deleting each one. May take a while, but it's an option.
Sometimes I have had to delete responses multiple times for it to show that they are actually gone.If you do need to reindex, you can go into your Survey Flow and create a new variable that does nothing (name it "Reindex" and set it equal to 0, or something like that). Once you do that, go back to the Data & Analysis tab and it should start reindexing.
Thanks, I will try this now!
In your survey editor, you can click the settings gear next to the question and select "Recode Values" to make sure the column headers are correct. Then in your dashboard settings, you can recode the question so that the column numbers recode to the header labels. See attached.As far as the best widget, you may have better luck creating some embedded data fields in the survey flow to help you out with calculations. For example, create a field called "NOVO_Utilized" which is set to a 1 or 0 depending on if they checked it or not. Then you can create 9 Numeric Value variables in your dashboard settings, which will let you use a bunch of different widgets. For example, use a Gauge chart to display the average of NOVO_Utilized, set it to percentage, and you will see a percentage of how often that response was checked.EDIT: I did the classic blunder of not attaching the image the first time. In this image, column 1's header is "sp1", column 2's is "sp2", column 3's is "sp3". Here they are j
I just contacted support about this a few weeks ago. We use iPads to administer our surveys, and an iOS update changed the size/spacing of Safari, so it started to use the accordion-style questions. Their response was that the question is displayed that way depending on the size of the screen in order to fit everything nicely, and said there is no way to change it.
One quick suggestion would be you can make a dictionary with the keys being the Survey ID, and the values being another dictionary of all the different options. You could then iterate through each dictionary entry:for surveyID, options in idDict.items():And then access the different keys/values in the "options" dictionary to set those parameters.
If your "NA" will come from piped text, you can just generate the piped text string in another question (e.g., it might be ${e://Field/variable} for an embedded data field called variable), and then put that string as the fifth column header. Unfortunately you cannot hide the column with display logic.There is a messier option that I am aware of. You can create two copies of this question, one without the fifth column, and one with the piped text for the fifth column. Then just use display logic to hide the question with the fifth column if the piped text will be "NA", and to show the one without the fifth column. The main issue here would be that your responses would be split between the two questions, and you would have to combine them via embedded data in survey flow to analyze all of the responses together. Doable, but messy.I am sure there could be another way to solve it, but I would leave that to somebody else here :)
When you say the embedded data is displayed in blocks 4-8, do you mean the table of contents entries for blocks 4-8 will be named depending on the data? Or just some part of the blocks themselves depends on the embedded data?Edit: I tested something that may work. Show your generic blocks (1-3), then set your embedded data immediately before the Table of Contents. You can use two branches to set the embedded data before entering the Table of Contents (because you can't put embedded data under the ToC). Under the Table of Contents, you can then just put blocks 4-8. If you then use Piped Text in the blocks to port in the embedded data, it should work because you have already set the embedded data before the ToC was started.
I did a quick test on a survey with three blocks, 1 question each. I answered all three questions, then pressed back to go back to the first question. Closed the tab, and looked under "Responses In Progress." Only question 1 had a response recorded, questions 2 and 3 read "This question was not displayed to the respondent."My surveys won't let me close them out (it takes ages), and I would have to update later when the survey response actually times out, but it would seem going back to the first page stops pages 2 and 3 from being recorded. The responses aren't deleted, since you can go forward and still see them during the survey, but it seems they are just not recorded when the survey is complete because the user went back to blocks before those blocks.
I noticed this, since we have two different paths the user can take based on a condition in our survey. Even though a user may take the second path, the blocks for the first path are shown first in the summary. I think the blocks are ordered in the survey editor window, and in the response summary, in the order they appear in survey flow. Have not found a way to change it, so I assume there is not one.
Already have an account? Login
Enter your E-mail address. We'll send you an e-mail with instructions to reset your password.
Sorry, we're still checking this file's contents to make sure it's safe to download. Please try again in a few minutes.
Sorry, our virus scanner detected that this file isn't safe to download.
