I don't think the approach was the main reason, in fact as I said elsewhere analyzing things in El Salvador is a bit difficult because of their economy. In other words if we consider the adoption a failure (which I say that is also too soon to say) the problem was the terrible economy of El Salvador.

Another thing to consider is that during most of the time they adopted bitcoin as legal tender, bitcoin price hasn't been doing well. We already know that weird volatility or weird drops is always been a big discouraging factor for people to look towards bitcoin while more consistent market or a rising price has always done the opposite.
I am sure if we had a way to measure adoption of bitcoin elsewhere we would see pretty similar results to the adoption in El Salvador.

Isn't this chart showing number of Chivo downloads and isn't this considered normal?
For starters you download the wallet once and only once. You don't need to download it every day.
Also you don't have to use Chivo, in fact you shouldn't use it because it is centralized and closed source. Instead people should and may be using something else like Electrum.
I'm not saying they are but just pointing out that like most surveys on cryptocurrencies this one also has many flaws.