These conclusions are frustrating to read because they haven't handled the numbers properly at all and have drawn poor conclusions. Take a scenario where Java/

These conclusions are frustrating to read because they haven't handled the numbers properly at all and have drawn poor conclusions. Take a scenario where Java/C# are popular and have 20 people applying for each position but Go isn't and only has 5 people applying for each position. You'd then have a far far better "pass rate" for Go developers. But it doesn't actually say anything about Java or Go in terms of developer proficiency.