Answer: A random selection of participants.
In order for professor Newberg's survey to be considered valid, he should have had a random selection of participants, instead. By doing so, all members would have an equal chance of being included in the study.