A large manufacturing firm is interested in improving worker productivity. It randomly selects 150 workers from its factory and over a period of one month collects data on the number of ocertime houirs worked and the total output produced per worker within the sample. Is this an observational or experimental syudy? If experimental, why?