As I mentioned in a previous post, this last summer, I assisted in the research by Dr. Elmira Kalhor at the Center for Policy Research on Energy and the Environment (C-PREE) in examining the effect of anomalous weather on economic activity, as part of an internship funded by the High Meadows Environmental Institute (HMEI) at Princeton University. Under Dr. Kalhor’s guidance, it became a fulfilling experience as I had the opportunity to formulate a model of how business activity is affected by extreme weather and thus apply many of the economic tools and theories that I learned to a practical space.
Now that it has been about three weeks since I gave my final presentation of my work to HMEI, I wanted to take the opportunity to reflect on my time there. While the internship was very enjoyable, it also presented me with many challenges – technical ones related to the statistical analysis and the research itself and practical/logistical issues in terms of having a valuable and fulfilling internship. In this post, I hope to discuss some of the more practical issues and guidelines that you could use to help maximize the efficiency of your internship experience, especially if you are working on a research-intensive project.
My first and most important piece of advice is to make a broad plan for what you want to do even before your internship begins and let your adviser or supervisor know about your research interests. Of course, just because you think that your internship should go in a certain direction does not mean that your adviser will agree – and you should always listen to what they have to say – but if your adviser knows about what you find most interesting about their work or what you are hoping to get out of the internship, they may be able to tweak the experience to your benefit. In my case, this was very important.
Initially, Dr. Kalhor thought that my interests would be primarily in literature review, learning about her model, and learning about the writing process for a research paper. She was not sure whether I would have the necessary background in econometrical techniques to perform any significant analysis on my own. Of course, I did want to learn about all of these things, but I also hoped to contribute to the nitty-gritty of the model itself and try to apply the tools I had learned in the classroom in a concrete fashion. Once my adviser came to know about my objectives, she was able to adjust my roles and responsibilities. As a result, I ended up working on a sub-topic within her paper by trying to justify a vital assumption she made about how uncommon (but not extreme) weather affects different kinds of businesses. Thus, effectively communicating your ideas about your expectations from the internship is an essential part of working with your adviser to constructing a satisfying, valuable, and mutually beneficial internship experience.
Once you have received a plan or objective for your internship, it is important to take the time to predict potential logistical “bottlenecks” and get them out of the way as quickly as possible. For example, if you think that you may need a particular data set at some point in the future, it may make sense to request that data early, even if you are not sure that you will need it. Such planning is especially true if you expect to work with massive data sets or require special permissions. In my case, I did not do this, which led to several unplanned mini-stoppages in my work. Most notably, Dr. Kalhor’s research involves human subjects. Although the subjects are anonymized, for a particular data set that I ended up needing, there is a small risk that the data could end up identifying the subjects, and because of this, I needed to complete an IRB certification to ensure that I knew the recommended practices when it came to handling such sensitive data. What is the IRB? Well, if your study involves human subjects, Princeton’s Institutional Review Board will review your study to ensure the safety of those human subjects. As I did not have that certification, it required me to complete a course, submit an application, and get approval from Dr. Kalhor’s own adviser – a process that temporarily put my work on hold. Had I realized that this was a required step, I could have tried to complete this application at the beginning of the internship or even before it started, as I had already contacted Dr. Kalhor for other things. Getting the necessary IRB approval to handle the data set was relatively simple. (If you are specifically looking for tips to navigate the IRB form, here is a post from 2018 that can help you with just that.) Still, the process had multiple steps and took several days, during which I could not continue with my work.
Finally, I would highly recommend setting up a calendar to meet with your adviser regularly at standard times during the week. In my case, initially, Dr. Kalhor and I were meeting “as necessary,” that is, whenever a particular problem or a research issue cropped up, we would schedule a meeting. However, it quickly became clear to us that having a formal weekly meeting would be beneficial. Not only did these meetings create a systematic method by which Dr. Kalhor could check in on my work and make sure that I was staying on the right track, but it also provided me with a valuable and streamlined opportunity to discuss new ideas with her and learn more about the economic modeling process in general. The experience was very valuable to me: I learned a lot in terms of creating a model that is technically correct and useful in practice. In these weekly meetings, I received a significant amount of mentoring from Dr. Kalhor. Although they may not always have been immediately helpful in solving specific issues with my research, they certainly helped me grow as an econometrician.
I hope that these tips from my own internship experience are helpful to you as you look to make the most of your own internship experiences and craft a plan that helps maximize your efficiency and growth. In my next post, I hope to delve deeper into the more technical challenges that I faced regarding the statistical analysis of extremely large data sets and introduce you to some of the resources at Princeton that you could use if you have to handle big data in your work.
– Abhimanyu Banerjee, Social Sciences Correspondent
