suffer from two forms of bias. first, samples unrepresentative of target population as its drawn from a very specific area so findings can't be generalised to target population. secondly, researcher has complete control of selection of participants and may avoid people they don't like the look of (researcher bias)