Year over year data filtering, CX dashboard

We have an annual survey that we send out every year, the survey changes slightly year to year and what has historically been send is a new survey every year rather rather than using the same survey with the minor edits. In 2023 there was as admin error where contacts were sent incorrect personal links… The issue was identified and remedied. This issue was narrowed down to a specific subset of respondents that we were able to identify based off embedded data, it did not affect all participants, is there’s a way other than deleting the “admin error respondents” to exclude those responses from analysis/dashboard? 

My initial thought was to export, remove the known error respondents, delete those responses, then import only those correct ones back into the survey. My concern is this would, at least I would image as I’ve never done it before, update the recoded date to the date I did the import. *The data I would be adjusting was from 2023, and I would like to keep the record date as 2023 as I’ve got multiple surveys mapped into the dashboard and then matrix groups grouped by years, if I was to import now it would group the true 2023 data with the 2024 data, right? I went into the survey data & analysis tab and  I can’t edit the recorded/stop/start dates.