Getting an automated reminder doesn't necessarily mean it isn't anonymous [in the context of what matters i.e. visibility to management], if the people who review the results of the survey don't have a means to see who's been sent a reminder. Behind the scenes, it has to be attributed to individuals in some way to ensure legitimacy e.g. people can't spam multiple responses, or respond to surveys they shouldn't.
As for the demographics, the surveys we used hid the detailed responses from any teams smaller than 5 people so e.g. a manager with 4 direct reports wouldn't get to see their ratings and it would just roll up to the higher level. Admittedly, dependent on the overall team hierarchy, you could sometimes narrow things down a lot by applying various filters and a process of elimination. Definitely not perfect in terms of anonymity, but I know it wasn't possible to link the textual responses to individuals, you could only really deduce it for the quantitative metrics. In fact in some cases the fact it was anonymous caused problems, because you'd get a one line comment out of context making it difficult to know how to address it, so in meetings with the team you'd have to call out such things and request further info (again anonymously if people didn't want to put their name against it).
Overall, I don't have a problem with such surveys. They provide useful insights that can drive change, dependent on the culture of the organisation.
If I take me as an example, one of the things I picked up on was that my team felt they didn't have enough understanding of our future vision and strategy. So I was able to focus a bit more on that in team meetings etc.
More broadly, I know from sitting on the leadership team for my division that a LOT of time end effort was put into addressing survey feedback. Way more than I had realised before I moved into management.
I once did one of these where I was in a small team and my writing style made it obvious it was me who wrote it
I used to tailor my writing style for the freetext survey elements, flattening it out into almost bullet points (very little use of commas, brackets etc). To someone who knew me it could probably still be attributed if they looked hard enough, but there wasn't really a confrontational culture anyway so worst case scenario would be having an honest conversation about perceived issues.