During my junior year of high school, my Principal and mentor, Dr. Keith Guy, introduced me to my first Independent Study (IS) course.
Then, during my AI Master’s degree, my professor, Dr. Erhan Guven, advised me during two semesters of IS to focus on AI research.
These three classes have been the three most valuable courses I’ve taken in over 40 years of continuous learning. However, I had to put a lot of effort into these courses to receive the benefits of IS.
During these three IS courses, I learned several important lessons about what you should do and what you should avoid doing in an IS.
So, to help you prepare for the next step in your educational journey, here are the most important things you need to know to succeed in an IS.
The first question you need to ask yourself is, “Why do I want independent study?” The goal of an IS is to allow you to explore a subject or topic of interest in-depth beyond what is covered in a standard curriculum. IS students are self-directed and work at their own pace under the guidance of an advisor.
If you’re pursuing a Ph.D., taking an IS course is a smart choice. Ph.D. programs are highly competitive, and having a few published papers can greatly increase your chances of being accepted. If you’re considering IS because you think it will be less work, think again; IS is more work than a regular class.
Also, consider “why” the university and your professor are willing to allow you to pursue an IS. There is a higher cost to the university and your professor to facilitate an IS. So, in return, they are expecting you to include both the university and your professor’s names in your research – to earn them more prestige.
The next step is to select an advisor for your IS. Choosing the right advisor is the 2nd most important decision in an IS. Your advisor will often be a professor who you’ve had in a previous course — someone you’ve “clicked” with. However, it can also be a professor you haven’t worked with yet.
You want to find someone with similar research interests, a compatible work ethic, and experience advising students. They also need to have the time necessary to take on an advisee and be allowed by the university to advise IS students. Beyond this, you want an advisor you trust and work well with.
After you’ve selected an advisor, you need to create a research proposal for your IS. You need to explain the purpose of your research, provide background on the subject, and discuss your plan to research the topic. My first IS research proposal was three pages long, half of which was a list of references.
Then, you fill out an application form for IS. You may also need to email the head of the department to advocate for your IS. However, your advisor may also advocate for IS on your behalf. If approved, the head of the department will sign off on your application so the registrar can add IS to your class list.
Once approved for IS, you need to develop a plan for the semester. Your plan explains what you will do, how you will do it, and when you will do it. It’s not super detailed, just a week-by-week high-level list of bullet points. This plan will change as you progress through your IS, so don’t worry about the details.
I recommend that you structure your IS into three equal parts. In the first part, you will read prior literature to fully understand the topic. In the second part, you will conduct an experiment to test a hypothesis. In the final part, you will author a paper and prepare a manuscript for submission.
You should plan to meet with your advisor once per week. During your weekly meetings, you should come prepared to answer to the following three questions: What did you accomplish last week? What do you plan to do next week? What questions, help, or guidance do you need from your advisor?
Add this meeting to your calendar and set an alarm to avoid missing it. Come prepared. Show up a few minutes early so your advisor doesn’t have to wait for you. If you cannot attend a meeting, inform your advisor well in advance. Then, send them an email recapping the three questions listed above.
The first 1/3rd of your IS should focus on reviewing the prior literature on your topic of focus. You want to know this subject matter inside and out. You need to know what progress has already been made on this topic to avoid duplicating existing research. You also need to be able to support your research with this info.
Do not skip this step. Nothing is worse than developing a research project, running an experiment, and writing up your results only to find out that someone else already beat you to it. Also, while reviewing prior literature, record everything you’ve read using a reference manager like Mendeley or a BibTex file.
After reviewing the prior literature, it’s time to develop your hypothesis. During your literature review, you will likely have encountered many interesting questions. Some of these questions will be answered in other literature. However, unanswered questions become the foundation of your research hypothesis.
Your hypothesis should answer a question that is novel, interesting, and incrementally advances your field of study. It should be specific, testable, measurable, realistic, and can be completed within about five weeks or so. I recommend you create the simplest possible experiment to test your hypothesis.
With your hypothesis in hand, you are ready to conduct an experiment. This typically involves designing an experiment, running the experiment, collecting data, and analyzing the results. This process is roughly another 1/3rd of your semester. So, manage your time wisely and know when to wrap things up or pivot.
It is easy to make errors while conducting scientific experiments. So, you need to double-check all your work and avoid various biases. Unfortunately, doing good science is really hard work. However, being the first person in the world to discover something new and interesting is also genuinely fulfilling.
After you’ve analyzed your results, you’re ready to author a research paper. The paper is the primary output for your IS. It is how you communicate your findings to a wider audience. It is also how your professor will grade your performance. This step and the step below are the final 1/3rd of your IS.
I highly recommend you learn how to use LaTeX and then write your paper in LaTex using Overleaf with BibTex for your references. LaTex has a relatively steep learning curve. However, once you’ve learned it, it will make your life much easier – especially when you get to the next step, submitting for publication.
The final step in an IS is to submit a manuscript for publication. This may happen during your IS, or you may submit it after your IS to a Call for Papers (CFP) for a conference. You must format your paper individually for each journal or conference. This process is often difficult, tedious, and frustrating.
Articles in journals are the most prestigious but also the most difficult and time-consuming. Articles in conference proceedings are the next most prestigious and medium difficulty. However, you can also write a paper only for your advisor to read, but there’s minimal prestige in an unpublished paper.
Remember, your professor and the university have likely granted you IS in return for the prestige a published paper provides them. So, if you are pursuing a published paper, you should include the university’s name and your professor as a co-author. Your name, however, should always go first.
But wait, there’s more! After you’ve submitted your manuscript for publication, you will have a few post-IS follow-up tasks. You need to respond to feedback provided during the peer review process. If accepted, you must also make any required changes and de-anonymize your manuscript for publication.
You may also want to submit your paper to a pre-print site like arXiv to claim primacy for novel research. You should also add the paper to Google Scholar, Semantic Scholar, and your website. In addition, if you submit your manuscript to a conference, you will be expected to deliver a presentation there.
What is most important in your independent study is that you get what you need out of it. If your goal is to learn a new topic, then read. If your goal is to learn how to do research, then conduct an experiment. If your goal is to publish your first research paper, then publish. Everything else is secondary.
I hope this article has helped prepare you to succeed in an Independent Study course. If you’d like to learn more, check out my articles on Choosing a Degree in AI and Advice for College Students.