It was an econometric approach. Our data showed a regression. I don't have the data in front of me. However, it concerned about a hundred people who decided to replace their doors or windows or both.
Our study isn't the only one to show that the returns were lower than predicted. I can actually tell you about another study that we carried out. We reviewed 39 papers in the literature that also examined investments in new doors and windows. The result was the same in multiple cases. These investments didn't lead to any reduction in energy consumption.
As I said earlier in my response to Mr. Bonin, we didn't carry out any randomized controlled trials beforehand. So we couldn't really know whether receiving money to replace their doors or windows influenced people's decision to carry out these renovations. They might have carried out the renovations even if there hadn't been any air leakages. We can't determine this without having established an evaluation system beforehand or offered the investments to households at random.