Folks can data dredge and present their results as significant. Lay people trust the conclusions. They had the data, but they didn't understand that the data presents a narrow view. They trust it blindly.

Worse are the people that treat conclusions drawn from polling and surveys as scientifically rigorous.

Happens all the time with Vox, NYT, Quartz, you-name-it articles. Policy is enacted from information like this.

Not always. Data can quite often completely mislead us. See Simpsons paradox for an example.