You mean like opinion surveys, or small sample sets?
Both. And then some.
Unrepresentative samples can occur either due to small sample size or due to poor sampling methods.
Conscious or subconscious bias on the part of those gathering the stats can play a role.
Questions that are deliberately worded or structured or sequenced so as to be leading can produce misleading outcomes.
Unclear definitions of terms can be an issue. "Have you ever cheated on your partner", for example, could mean many things.
I'm referring to statistics in general, not to this specific case. Which I only shared in order to rile the broflakes.