PhD/PsyD Assigning dissertation projects

[futureapppsy2]

---

Members do not see this ad.

I've known of a few labs where advisors basically assign their students portions of a large ongoing project (say, a certain set of analyses from a large dataset or a particular variation on an ABA study) as the student's dissertation study and have them lead that portion of the data collection/cleaning, analyses, write-up, etc.

On one hand, this seems efficient and probably more reflective of how large research groups actually run. On the other hand, it does kind of deprive students of the process of planning a dissertation project as the PI. Thoughts?

 

[WisNeuro]

I've known of a few labs where advisors basically assign their students portions of a large ongoing project (say, a certain set of analyses from a large dataset or a particular variation on an ABA study) as the student's dissertation study and have them lead that portion of the data collection/cleaning, analyses, write-up, etc.

On one hand, this seems efficient and probably more reflective of how large research groups actually run. On the other hand, it does kind of deprive students of the process of planning a dissertation program as the PI. Thoughts?

I've seen it happen, and I think it's doing a great disservice to students. I'd rather see a student design and run a project, from start to finish, with some flaws, rather than get a more polished study hand picked for them. You learn so much more when you get your hands dirty. And, with proper supervision, you learn what mistakes you made so that you hopefully avoid those same mistakes in the future.

 

[FadedC]

When I get my MA (seperately from my PhD), I did a Masters thesis where I was given a piece of a large, well funded study to analyze and write up. It gave me the chance to engage in some pretty detailed analysis and coding of a wide subject pool that I'd never have been able to do with my own project, but there was no study design or data collection. My doctoral dissertation was more traditional, I designed my own study, collected the data, etc.

I was pretty happy to have had both experiences, but I think I probably learned alot more from my dissertation and designing my own study. Still, at the time, I would have loved it if someone had just given me a piece of their own project to work on again, just becuase it would have been much easier!

 

[Therapist4Chnge]

There are definitely a lot of pros and cons to "assigned" projects. From purely a learning perspective I think it is more instructive to do it from scratch. There is something to be said about a student having to think through scope and design and practicality of the project. I'd argue that a student would be more invested in their own research question, though it may also increase the risk of a project being too unwieldy (and thus taking much longer to complete, if at all). Proper mentorship should curb this risk for the most part, so it is probably more of a hypothetical risk than an actual risk. The other side of the argument are probably some students that would prefer to, "just be done with it" after too many hours of struggle and frustration because their recruitment, analysis, and/or other parts took far longer than anticipated.

 

[singasongofjoy]

Definitely arguments either way. On the one hand, a good dissertation is a done dissertation, and doing a small part of a larger study may give you opportunities re: sample size and different types of analyses than if you designed a project yourself. On the other hand, I agree you learn more about the full process of designing a study if you, ya know, actually have to do that. A lot of it may depend on the type of mentorship you have. If your advisor is a pretty hands-off one who won't give you a lot of guidance on designing your study (which is sad but happens) I'd say just do the first option and get it done, then seek out a research-heavy postdoc if you still want more research experience. If you've got an advisor who will help you through the process to help you avoid and solve unexpected pitfalls as you occur so that you're not mired down in a project that will take too long to finish, I'd go the latter route. It could be that you could find some sort of middle ground and tweak a small component of the bigger study (add an additional measure or two? Piggyback on recruiting efforts?) to avoid some of the most common frustrations of designing your own from the ground up. For what it's worth, I did the first for my thesis to get it out of the way, and designed my own study for dissertation. I'm glad I had both experiences, though dissertation did make me way more anxious as I didn't get as many participants as I'd hoped and the whole thing took longer than I expected. Still managed to finish by end of internship but internship would have been even more fun without that stress.

 

[Justanothergrad]

I've seen it happen, and I think it's doing a great disservice to students. I'd rather see a student design and run a project, from start to finish, with some flaws, rather than get a more polished study hand picked for them. You learn so much more when you get your hands dirty. And, with proper supervision, you learn what mistakes you made so that you hopefully avoid those same mistakes in the future.

This. I see folks who dictate a design as frequently serving the advisors interest (publications and work distribution for their area) although their are admitted benefits to the student. I think the risks of project scale and common pitfalls that people cite as a common reasons/benefits for dictating an already designed project are bigger issues when good/close mentorship is not being provided. To me, the purpose of a dissertation is to develop a competent way of thinking empirically. It is a hurdle, but one that has implications for our thinking as a field.

 

[PsyDr]

I've known of a few labs where advisors basically assign their students portions of a large ongoing project (say, a certain set of analyses from a large dataset or a particular variation on an ABA study) as the student's dissertation study and have them lead that portion of the data collection/cleaning, analyses, write-up, etc.

On one hand, this seems efficient and probably more reflective of how large research groups actually run. On the other hand, it does kind of deprive students of the process of planning a dissertation program as the PI. Thoughts?

I've seen it happen, and one person in my lab was "given" a small part of my work to publish for her dissertation.

It's typical of the greed of the established profession. "I got mine, and want more.". If they wanted to teach they wouldn't do this. But more publications=more money/prestige. Same concept as older people refusing to retire due to their greed which has led to their financial mistakes. Or why non cms funded post docs pay bs. Or why established clinics ask for insane cuts.

 

[EmotRegulation]

As a mentor, my "rule" is that my students have to do an experimental design (laboratory based, ideally) for at least one of their projects, either masters or dissertation. The other one can be more survey/correlational (typically using advanced statistical techniques, sometimes measure development) or, in theory, a project using pre-existing data. As I get more and more into experience sampling research, the amount of data I'm going to have will be massive and I suspect this may become data for masters theses for future students, because I just won't get to writing up all of the interesting stuff we're going to get.

I actually think designing a project from the ground up AND secondary data analysis are both worthwhile research endeavors. Right now, my students have done their own studies for their milestone projects but pull from existing data for publications.

 

[deleted343839]

I've known of a few labs where advisors basically assign their students portions of a large ongoing project (say, a certain set of analyses from a large dataset or a particular variation on an ABA study) as the student's dissertation study and have them lead that portion of the data collection/cleaning, analyses, write-up, etc.

A variation on this theme is for the student to tack an original study onto the mentor's parent project (an additional experiment, recruitment of a specific subgroup, etc.). Research supplements are a good way to fund this kind of work. I've also known a few people with lucky timing who managed to incorporate their ideas into their mentor's large grant proposal and then work one of the project aims as their own. This can be good experience for future involvement in team/collaborative science. It does require more mentoring on the front end, though.

 

[smalltownpsych]

I've known of a few labs where advisors basically assign their students portions of a large ongoing project (say, a certain set of analyses from a large dataset or a particular variation on an ABA study) as the student's dissertation study and have them lead that portion of the data collection/cleaning, analyses, write-up, etc.

On one hand, this seems efficient and probably more reflective of how large research groups actually run. On the other hand, it does kind of deprive students of the process of planning a dissertation project as the PI. Thoughts?

I tend to favor the original research design from the ground up approach. Although the students in my program who worked on small parts of a larger project tended to get done a lot quicker. The only reason I didn't join that lab was I would have had to drive a couple of hours each way to get there and with practicums it would have been impossible so I chose to design my own study. Learned a lot and am grateful for that experience, but it cost me a year probably and had to take out a student loan to fund it despite teaching several classes during that time.