In substantial part, with methods of research. Most sociologists, especially those with an interactionist perspective, embrace the participant/observer methodology as the one that yields the most valid results in studying a particular subgroup.
For instance, the editors of our textbook, the Adlers, spent a significant amount of time early in their respective careers studying drug dealers and drug smugglers.
Of course, drug dealing and drug smuggling are types of deviance that are proscribed in criminal law (in other words, one can potentially get arrested, prosecuted and imprisoned for engaging in these acts).
The Adlers freely admit that they broke the law while being participant/observers of this particular subculture, arguing that "it would have been impossible for a non-user to have gained access to this [drug-dealing] group to gather the data presented [in this article]."
The Adlers then went on to say that they worried about whether they were exploiting their friendships merely for the sake of their research. Reflecting on the above, what do you think about the Adlers' methodology?
Were they, in fact, exploiting their friendships for the sake of publishing an article?
If so, was the end result worth the exploitation? While the Adlers admit that the research data they discovered was not subject to legal privilege similar to that of a treating psychiatrist, they did not dwell on the fact that, additionally, they, personally, were not immune to arrest for possessing and using marijuana and cocaine. Was the research goal worth the risk?
Would the Adlers have risked a possible criminal record if they were not, before undertaking the study, already possessors and users of two of the smuggled drugs?
Let me be perfectly clear: in asking these questions, I am not taking a side for or against the Adlers'.