The summary.
STATE YOUR QUESTION IN ONE SENTENCE
If it cannot be, it is probably several questions.
Establish what has already been written before starting, or you may repeat work that exists.
THE MOST EFFICIENT ROUTE TO SOURCES
Reference lists of relevant works. One good paper leads to many.
Record every reference completely, immediately. Reconstructing them later is slow and error-prone.
NEVER CITE A SOURCE YOU HAVE NOT READ
Citing from someone else's citation propagates their errors as yours.
MARK QUOTATION, PARAPHRASE AND YOUR OWN THOUGHT DISTINCTLY IN YOUR NOTES
Confusing them later produces accidental plagiarism, which is judged the same as the deliberate kind.
And record the source and page with every note — one without them cannot be cited.
THE METHOD FOLLOWS THE QUESTION
Not preference, and not convenience. Plan the analysis before collecting the data.
STATE LIMITATIONS
Every study has them, and acknowledging them strengthens work rather than weakening it.
REPORT WHAT YOU DID NOT EXPECT AND WHAT DOES NOT SUPPORT YOU
Selective reporting is misconduct, and it is detectable.
"The evidence does not settle this" is a legitimate finding.
BACK UP EVERYTHING, IN MORE THAN ONE PLACE
Losing months of research is a well-known and entirely avoidable disaster.
