Which would make for a weak claim?
• a blog cited as the source
• a large sample size
• clear variables and controls
• multiple trials