I went and read the article but have not tracked down the original paper. It is at least 26%, because that was the percentage of studies that provided access t

I went and read the article but have not tracked down the original paper. It is at least 26%, because that was the percentage of studies that provided access to their data that proved faked or fatally flawed. It may be substantially higher.