At this point, all the hardest work of a paper is behind a writer. The research. The writing. The thorough and formal citation. All the stylistic questions have been answered, all the technical concerns addressed. The paper as an academic product is virtually complete. Yet what remains to be done can mean the difference between the paper being admired, and the paper being scorned.
Before taking this next step, however, a writer should shelve the paper for an hour, or a day, or a week, depending upon how much time has been allowed. Writers who are wise stewards of their resources, including time, give themselves this walk-away period to rest their brains and eyes. When they return, they will come nearer reading the paper as a stranger—or a professor—might.
This next step is called proofreading. The printing term refers to when an early version of printed matter is printed and closely examined to prove to a printer's satisfaction that no errors exist. If errors are found, they are corrected and another “proof” printed. Similarly, an academic writer should examine a paper closely and correct it as necesary before printing out a final copy.
What errors are sought? Misspellings, such as the intentional one in the previous paragraph! (Did you spot it? “Necessary” needs a second “s.”) Repeated words (“the the”). Errant capitalization. Faulty punctuation. Irregular spacing. Mislabeled lists or misnumbered pages. These are all mechanical mistakes discoverable by anyone willing to patiently, methodically review the text.
The next level of scrutiny required for clean copy centers on flawed or weak grammar. The more grammatical knowledge a writer has, the greater the chance of discovering an incorrectly used word, a poorly composed phrase, or an awkwardly structured sentence. In respect to the latter, here is a tip: If a sentence cannot be read smoothly, without stumbling, it should be rewritten.
The last proofing step is conceptual in scope. Does it hang together? Is there unity in its writing? Does it progress from introduction to conclusion, with the latter affirming the former? These questions should have been answered earlier, but the final proof of a paper's conceptual integrity is in the finished product. Is the paper coherent, structurally sound and error-free? If not, re-do it.