Degree In Sight
"Many students reach this stage of their careers having been focused for several years on the 'trees,'" says Yale University cognitive psychology professor Brian Scholl, PhD. "This section of the dissertation provides an opportunity to revisit the 'forest.'"
Fellow students, your adviser and your dissertation committee members can help provide that outside perspective, adds Yale clinical psychology professor Susan Nolen-Hoeksema, PhD, who teaches a course on writing in psychology.
And while the discussion should put your research into context and tell a story, say experts, it should not overstate your conclusions. How do you find the balance? Follow these do's and don'ts.
DO: Provide context and explain why people should care.DON'T: Simply rehash your results.
Your discussion should begin with a cogent, one-paragraph summary of the study's key findings, but then go beyond that to put the findings into context, says Stephen Hinshaw, PhD, chair of the psychology department at the University of California, Berkeley.
"The point of a discussion, in my view, is to transcend 'just the facts,' and engage in productive speculation," he says.
That means going back to the literature and grappling with what your findings mean, including how they fit in with previous work. If your results differ from others' findings, you should try to explain why, says Nolen-Hoeksema. Then, launch into "bigger picture" issues. For example, a clinical study might discuss how psychologists might apply the findings in a clinical setting or a social psychology project might talk about political implications.
By exploring those kinds of implications, students address what Scholl considers the most important-and often overlooked-purpose of the discussion: to directly explain why others should care about your findings.
"You can't and shouldn't rely on others to intuitively appreciate the beauty and importance of your work," he says.
Sounds simple, right? In fact, choosing what to include can be overwhelming, warns sixth-year Yale University social psychology graduate student Aaron Sackett.
"It is easy to get caught up in the desire to be extremely comprehensive and to bring up every potential issue, flaw, future direction and tangentially related concept," says Sackett. "However, this will make your dissertation seem like it has raised more questions than it answers."
Limit your discussion to a handful of the most important points, as Sackett did on the advice of his adviser.
"No reader wants to wade through ten pages of suppositional reasoning," says Roddy Roediger, PhD, chair of psychology at Washington University.
DO: Emphasize the positive. DON'T: Exaggerate.
One of the biggest errors students make in their discussion is exaggeration, say experts. Speculation is fine as long as you acknowledge that you're speculating and you don't stray too far from your data, say experts. That includes avoiding language that implies causality when your study can only make relational conclusions.
"If your study was not a true experiment, replace verbs that imply causation with words and phrases such as 'correlated with,' 'was associated with' and 'related to,'" write John Cone, PhD, and Sharon Foster, PhD, in a forthcoming revision of "Dissertations and Theses from Start to Finish" (APA, 2006).
Steven David, PhD, who successfully defended his dissertation in clinical geropsychology at the University of Southern California last May, found this point to be particularly difficult. When he defended his master's thesis, his committee told him his conclusions went too far out on a limb. He used more restraint with his dissertation and his committee thought he wasn't positive enough.
"The moral here is to try to find a balance where you set a tone that indeed celebrates interesting findings without too many leaps, while at the same time reporting limitations without being unnecessarily negative," says David.
Indeed, every discussion should include a "humility" section that addresses the study's limitations, write Cone and Foster. But avoid beginning the discussion with a long list of study limitations, says Nolen-Hoeksema.
"This makes me think 'Then why should I care or believe anything you found,' and want to stop reading right there," she says. "Limitations should be noted, but after you've discussed your positive results."
DO: Look toward the future. DON'T: End with it.
Along with noting your work's limitations, it's helpful to also suggest follow-up studies. But don't dwell on the future at the expense of the present,says Scholl.
"I think that too many discussions make the mistake of ending with 'the future,'" he says. "Too often I am left excited not by what was in the dissertation, but by what was not in the dissertation."
Roediger agrees: "Conclude the general discussion with a strong paragraph stating the main point or points again, in somewhat different terms-if possible-than used before."
Remember, adds Scholl, you want readers to remember you and your work. The discussion section is the place to leave your mark. So instead of simply summarizing your data and suggesting a few obvious follow-up studies, think about presenting your data in a novel way, showing how the work might resolve an existing controversy in the literature or explaining how it connects to an entirely different literature.
By the time readers get to your discussion, they're tired, adds Sackett. Give them something clear, concise and interesting to read, and they're sure to appreciate it.
Beth Azar is a writer in Portland, Ore.
10 most common dissertation discussion mistakes
Starting with limitations instead of implications.
Going overboard on limitations, leading readers to wonder why they should read on.
Failing to acknowledge limitations or dismissing them out of hand.
Making strong claims about weak results.
Failing to differentiate between strong and weak results as you make conclusions about them.
Lapsing into causal language when your data were correlational.
Repeating the introduction.
Restating the results without interpretation or links to other research.
Presenting new results; such data belong in the results section.
Offering no concluding statements or ending with the limitations.
Source: Susan Nolen-Hoeksema, PhD, Yale University