If you collect data, very often it's just an assembly of numbers. If you have a hypothesis, then you can generate something from the numbers.
There was the case of Sam Milham, who was an epidemiologist from the state of Washington. He had the idea that there was a link between the incidence of leukemia in children and electrification. Electrification didn't occur in the United States at the same time; the north and east had it long before the south and west.
He went to different places and started collecting data on the incidence of leukemia, and lo and behold, he found that when electricity was introduced, within a few years there was a jump in leukemia. It correlated with the introduction of electricity. If you look at his data, you see there's a bump in there, and that was the origin of the linking of ELF with leukemia. He knew what he was looking for, and luckily the incidence of leukemia had somehow been collected.
You have to find something that will have the kinds of numbers you want; otherwise, you will just have file cabinets full of numbers.