Issues and limitations

Things to consider that will constrain your project

Now that you have a general picture of how you should approach your independent project, you need to consider some of the limitations to what you can do. The space of possibilities for your project can seem huge; so these constraints are actually a good thing. They draw boundaries around what you can do, and make this choice more manageable.

You might not get to do exactly what you want, and there are variety of reasons this might happen. Consider the issues and limitations outlined here before you set your heart on a research question.

Time

Time: You only have so much of it, and there's no way to get more.

Good research consumes time. You only have a single semester to complete an extended study project, and two semesters for a dissertation. This will be going on while you're doing lots of other things (other modules, applying for graduate school or jobs, etc). The likelihood that, in this time, you will solve the puzzle of human language acquisition or reconstruct the evolution of English syntax back to Proto Indo European, is effectively zero. This is why we deal in concrete research questions instead of broad areas of interest.

However, even some specific research questions will simply be too time consuming for you to tackle. While you might be able to curate a small corpus of dialogue from online videos, you are not going to gather a billion word corpus of spoken European Spanish within a year. Be realistic about what you have time to do, and trust your supervisors' intuitions on what is feasible.

While we can't bend the space time continuum to get extra time for an independent project, it is possible to throttle the time you do have. The longer you dither in deciding what to tackle for your project, the less time you have to actually tackle it. Get started on thinking about your project as early as you can; you won't regret it.

Resources

Resources: They're limited, and we don't have everything.

We use lots of stuff to do our research, from complex and customised experimental software to specialised equipment. We have access to lots of cool stuff, but we do not have the resources to do anything. We cannot, for example, conduct research using an fMRI machine any more than we can pop over to the Large Hadron Collider. We don't have access to either of these.

Don't assume the resources are there to tackle the problems you're interested in exactly how you want to, and if they're not, don't get discouraged. Find another way to work your research question and keep moving forward.

Despite this limitation, we do have lots of resources available, particularly via the LingLab. However, these still take time to learn to use (which may exceed the time you have), and require expert supervision.

Expertise

Expertise: We know a lot, but we don't know everything (no one does)

Even if we could access the Large Hadron Collider, we wouldn't know what to do with it. We can't support projects that require in-depth technical expertise that we don't have. So, while fMRI is a very exciting approach, even if we did have access to the resources/equipment to conduct an fMRI study, we wouldn't know how to, so this isn't a viable option.

This limitation might extend to things we do have expertise in, but don't have time to impart that expertise. For example, unless you have some previous experience in programming (or want to get it independently over the summer), attempting an agent-based model or other computational approach is not a good idea.

This limitation extends to things like working with certain vulnerable populations or attempting complex analyses. (As an aside, this would also apply to an fMRI study: gathering and analysing fMRI data involves extensive training). You will definitely learn how to do new things during the course of your project, but there are some things that are too complex to learn to do in a single semester or year.

Permission

Permission: Just because you have an idea to do something doesn't mean you should

As researchers, we have a responsibility to conduct our research in an ethical way. This means everything form being honest and forthright with research participants about the purpose of a study and where their data will go, to thinking critically about assumptions and biases we might bring to the table in a modelling project. The ethics of human subjects research is arguably more complex than other approaches, but ethical considerations will form a part of every LEAD project.

At some point, either in Stage 2 methods module or as part of your onboarding into LEAD as a research student, you will need to undergo some ethics training. This training will give you a good grasp of the ethical issues you need to consider in formulating your research question and choosing a methodological approach.

There is what's called "block" ethical approval extended to dissertation and extended studies projects that are classed as low risk (if you don't already know what this means, you will after the ethics training). However, within LEAD, you must go through a "mock" ethical approval process with your supervisor. The reason this is a "mock" process is because your application will not actually go to the ethics board for approval, but you must engage in this process to be sure your research design is ethically low risk. Many projects won't encounter any ethical barriers, but the point of this process is that you can't assume this - you need to take a critical approach.

If your research question involves participants from vulnerable populations (e.g., children, people with autism or ADHD, immigrants, to name only a few), you would require special permission and potentially training collect new data from these populations. Currently, we cannot support this. The reasons for this will be clear once you have completed the short ethics training, but if you need any more information about this, do not hesitate to ask your supervisor.