Importing possible answers from CSV: replace vs append

1

Hi Team,

When we update a Dropdown list with Updated CSV file, the stored data in data table for that question in deleted.

Here in The CSV file, we may have added 2-3 options to existing list of options. And in order to maintain uniformity of dropdown options against multiple projects, we uploaded CSV and replaced options instead of manually adding it.

Expectations is that only the dropdown list should show the newly update list of options.

What was not expected that the Previous Saved Data in the Main Data table for the question from previous entries gets deleted. This has lead to thousand rows of saved data for the Dropdown question from Saved entries is deleted.

(Request to not close question after Admin replies, as There may be subsequent doubts to it)

2

We have already responded to your question →

we uploaded CSV and replaced options instead of manually adding it.

This is the key point: if a list of possible answers is REPLACED, all previous references to that list are removed since they no longer exist.

Behind the scenes, each possible answer is associated with a unique identifier (answer_ref). For example, your original list might look like this:

[
    {
        "answer": "yes",
        "answer_ref": "58d92c2b50435"
    },
    {
        "answer": "no",
        "answer_ref": "58d92c2b50436"
    },
    {
        "answer": "n/a",
        "answer_ref": "58d92c2b50437"
    }
]

When you replace the list, it might change to:

[
    {
        "answer": "yes",
        "answer_ref": "58d92c2b50439"
    },
    {
        "answer": "no",
        "answer_ref": "58d92c2b50438"
    },
    {
        "answer": "n/a",
        "answer_ref": "58d92c2b50432"
    },
    {
        "answer": "another option",
        "answer_ref": "5ad92c2b50432"
    }
]

Even though the possible answers may look the same to the user, the unique identifiers (answer_ref) are different. This means there is no longer any reference linking the previous “yes” response to the new “yes” response. The system considers those to be two completely different responses when looking at the anwers_ref identifiers (58d92c2b50435 vs 58d92c2b50439).
The answer_ref 58d92c2b50435 cannot be found anymore, so all those responses are gone.

This principle applies to any software in general. For example, if you replaced a folder with another folder of the same name on your PC, you wouldn’t expect to find the previous content in the new folder, despite the folder having the same name.

To correctly add possible answers to an existing list, and therefore preserve existing responses, Epicollect5 provides the option to APPEND possible answers to an existing list when importing the possible answers from a csv file.

append-csv
append-csv1136×848 87.3 KB

User Guide reference →

As a rule of thumb, we highly recommend taking a backup of your data before making changes to an existing project. Backing up your data ensures that you have a safety net in case anything goes wrong during the update process. This is particularly important when dealing with complex forms and datasets, as even small changes can sometimes lead to unexpected issues or data loss.

We hope this clears your doubts and we apologize if Epicollect5 does not meet your needs.