Is it recommended to use "we" in research papers? If not, should I always use passive voice?
We is used in papers with multiple authors. Even in papers having only one author/researcher, we is used to draw the reader into the discussion at hand. Moreover, there are several ways to avoid using the passive voice in the absence of we. On the one hand, there are many instances where the passive voice cannot be avoided, while, on the other, we can also be overused to the point of irritation. Variety is indeed the spice of a well written scientific paper, but the bottom line is to convey the information as succinctly as possible.
It's definitely OK to use "we" in research papers. I edit them professionally and see it used frequently.
However, many papers with multiple authors use such constructions as "the investigators," or "the researchers." In practice, there really aren't that many occasions when the authors of a scientific paper need to refer to themselves as agents. It happens, sure. But not that often.
Rather, the Introduction, Methods, Results, and Conclusion sections should speak for themselves. Any reference to the authors should be minimal as except in rare cases they are not germane to the findings.
APA (The American Psychology Association) has the following to say about the use of "we" (p. 69-70).
To avoid ambiguity, use a personal pronoun rather than the third person when describing steps taken in your experiment.
Correct: "We reviewed the literature."
Incorrect: "The authors reviewed the literature."
For clarity, restrict your use of "we" to refer only to yourself and your coauthors (use "I" if you are the sole author of the paper). Broader uses of "we" may leave your readers wondering to whom you are referring; instead, substitute an appropriate noun or clarity your usage:
Correct: "Researchers usually classify birdsong on the basis of frequency and temporal structure of the elements.
Incorrect: "We usually classify birdsong on the basis of frequency and temporal structure of the elements"
Some alternatives to "we" to consider are "people", "humans", "researchers", "psychologists", "nurses", and so on. "We" is an appropriate and useful referent:
Correct: "As behaviorists, we tend to dispute...
Incorrect: "We tend to dispute..."