Degree In Sight
In an ideal world, dissertations have a logical progression: You research your topic, select a dissertation committee chair, submit a proposal, collect your data, write the manuscript and successfully defend it. Congratulations, Doctor.
But unexpected difficulties can and do arise. Your committee might create seemingly insurmountable hurdles, your adviser can leave, the ethics board can quash your study, or your study can go awry.
Don't fret. Students who've been through the experience say most every problem has a solution, and finding that solution is almost always better than starting over.
It's a truism that your research is only as good as your research subjects, so what do you do when you can't seem to find very many? That's the question psychology graduate student Rebecca Osterhout at Binghamton University in Vestal, N.Y., asked herself, and it's one she eventually answered.
Her research into long-term relationships among minorities relies largely on self-reports, so she put several months and thousands of dollars into mailing out about 3,000 questionnaires to Latino and black households. After a few more months of waiting, she realized she had a problem—only 10 households had responded.
"For whatever reason, people just weren't sending the questionnaires back to me," she says. "That was pretty terrible. I wasn't prepared for that."
All told, the delay probably cost her a year's worth of time, she figures. But rather than scrapping the project, Osterhout thought outside the box. She posted ads with Google, and opened a MySpace page explaining her project. Because MySpace allows you to search for people in various categories, including race and relationship status, Osterhout was able to contact hundreds of people who matched her research criterion.
Now, she's close to meeting her sample size goals and her research is back on track, she says.
Other times, what you plan on studying and what you wind up studying can be vastly different, depending on what passes muster with the institutional review board. If the board makes you change your methodology, how do you still maintain the integrity of your work?
Don't panic, advises Bryan Castelda, PhD, a clinical psychologist who now works at Science Applications International Corp. in La Jolla, Calif. His own experiences bear out his advice: As a grad student, he proposed a study where he'd ask people who had once been physically or sexually abused to read aloud generic vignettes about abuse and then observe their bodily responses such as skin conductance and heart rate. His dissertation committee approved his proposal and told him to start working on it. But five months into his project, a letter arrived from the institutional review board detailing their concerns about having abuse victims read aloud vignettes about abuse.
"It didn't matter how many great articles I had in front of me attesting to the validity and ethical soundness of the method," he says, "they weren't going to budge."
Castelda had no choice but to choose a new research design, so he measured participants' responses to loud noises instead. He was disappointed but says he learned an important lesson: Nonpsychologists or nonscientists may not be persuaded by psychological research. While Castelda and his committee knew that his methods were sound and ethical, members of the IRB didn't have the same background and interpreted the work through their own experience.
A little reconnaissance work ahead of time might have been helpful, he says. Castelda recommends "talking to students and faculty in psychology and other departments to learn if there are particular issues that you should be attuned to when proposing [a study] to the IRB." If he had done that, he says, he wouldn't have lost time pursuing a doomed paradigm.
Looking back, Castelda realizes that he gained a great deal of respect for the IRB's mission, despite his frustrations. "I witnessed firsthand just how much time and energy these dedicated individuals invest into ensuring the protection of human subjects," he says.
Adapting your research methods is one thing. Having to throw out your topic and start anew is something else. While uncommon, it can happen: Psychology student Anna Latysheva is on her third research topic.
Latysheva initially aimed to study eating disorders but couldn't find enough participants to test a new tool for measuring annihilation anxiety. Frustrated, she asked her committee for help, and they told her they didn't think her methods would get at the heart of her question anyway.
"It was upsetting because I spent four months on it," Latysheva says. She wishes she had found out about her committee's concerns earlier.
She then went on the hunt for a new topic. As many professors in her department were post-traumatic stress disorder experts, she drew up another proposal to study a new kind of projective test for people with PTSD. Again, her committee thought it wasn't a viable measure. Finally she suggested examining perfectionism among people with PTSD. Her committee approved her proposal and she hopes to graduate this year.
Having to change her topic has dampened her enthusiasm, she says, but she's glad that she's finally making progress in the program. If she could do it over again, she would have sought professors in a different department who were more interested in what she was working on, she says. By putting more time into finding the right people to work with, she realizes she could have avoided many of her problems.
But in some extreme cases, when an irresolvable problem arises, it might be worth it to cut your losses and find a new committee or chair, says Linda Chrosniak, PhD, a cognitive neuroscience professor at George Mason University. If that happens, Chrosniak stresses that you should remain as amiable as possible.
"Students really can burn their bridges," she says. "And later on, they're going to need those relationships and references."
Be diplomatic, she suggests. Explain to your committee that it would make for a better working environment if you were with someone whose objectives were more in line with your own. You probably won't be able to avoid stepping on a few toes, but step as lightly as possible, she says. Then, find someone who can support your research and is also dedicated to getting you through the program.
All said and done
Ultimately, grad students should remember that the goal is to finish the program, so every effort should be put into preventing and fixing mistakes rather than scrapping a project and starting over.
As many professors point out, no research project is ever perfect. Resources, both financial and human, argue strongly for getting the job done as best as possible rather than starting over.
If the problems are related to the research itself, be flexible, Chrosniak advises. Only if students are in the very early stages of their research should they consider switching topics. The proposal is a good tipping point, she says. Before you've written the proposal, it's probably acceptable to alter your topic completely. But at any point after that, it's better just to press on, she says.
"If you're given lemons, just make that lemonade," Chrosniak says. "It might not be the best lemonade, but at least it's something. Decent lemonade is okay. If it'll get you through the program, do it."
By Michael Price