Most research stu­dents don’t seem to attend sem­i­nars. When asked, they usu­ally say the sem­i­nars are not on their topic, or they don’t under­stand them, or they find them bor­ing, or some other sim­i­lar rea­son. I think this is because stu­dents don’t under­stand the pur­pose of research sem­i­nars, and have not learned how to lis­ten to them.

Admit­tedly, many research sem­i­nars are badly pre­sented, and sem­i­nar speak­ers also fre­quently mis­un­der­stand the pur­pose of the sem­i­nar, which makes the prob­lem worse. In a pos­si­bly vain attempt to improve the sit­u­a­tion, here are some thoughts on attend­ing research seminars.

First, some advice to speak­ers: under­stand that sem­i­nars are intended to pro­vide brief and infor­mal tasters of a wide range of research. Peo­ple will get the detailed and for­mal ver­sion of research in the pub­lished papers. But often an infor­mal expla­na­tion with­out the details is more acces­si­ble. Also, a speaker can pro­vide the back­ground overview that is often miss­ing in the pub­lished papers. Speak­ers need to realise that there is no need to present detailed proofs, tables, and tech­ni­cal­i­ties — a sem­i­nar is a very poor medium for pro­vid­ing details about sta­tis­ti­cal and econo­met­ric research. You have given a suc­cess­ful sem­i­nar if every­one in the audi­ence has some idea about what you’ve done, most peo­ple in the audi­ence have been able to fol­low all of it, and you have inspired at least some peo­ple to read the paper.

Assum­ing speak­ers are doing that, here are some tips for stu­dents attend­ing research sem­i­nars. Note that much of this advice is adapted from Ravi Vakil, a Stan­ford mathematician.

• Go to research sem­i­nars from the time you begin your research degree. Don’t just go to sem­i­nars that you think are directly related to what you do (or more pre­cisely, what you cur­rently think you cur­rently do). Learn­ing to get infor­ma­tion out of research sem­i­nars is an acquired skill.
• Don’t worry if you lose the thread of a talk. Just try to get back on track again. Let the ideas flow past and try to learn some­thing from what is being said, even if it is just the words and terms that are being used.
• Ask ques­tions. Try to make a per­sonal rule to ask at least one ques­tion at each sem­i­nar you attend. Sim­ply try­ing to for­mu­late a mean­ing­ful ques­tion will help you focus on what is being said and aid your under­stand­ing. Fur­ther­more, you might learn some­thing from the answer. Often the best parts of a sem­i­nar are in the dis­cus­sion. But if there aren’t many ques­tions, the oppor­tu­nity is lost.
• At the end of the talk, you should try to answer these ques­tions: What research prob­lem was the speaker address­ing?  Why should we care about them? Is this prob­lem related to any other prob­lems I know about? It can help to write down these ques­tions at the start of the talk, and jot down answers to them dur­ing the talk.
• List any ter­mi­nol­ogy the speaker uses that you don’t know the mean­ing of. Then either ask the speaker to explain, or ask your super­vi­sor afterwards.
• See if you can get one les­son from the talk, no mat­ter how sim­ple. It might be about data analy­sis, or a math­e­mat­i­cal tech­nique used, or how to think about a cer­tain type of prob­lem, or what­ever. If you learn one les­son from each talk, your knowl­edge of sta­tis­tics and econo­met­rics will steadily grow. If you are unable to learn even one thing from a talk, think about what the speaker could have done dif­fer­ently so that you could have learned some­thing. Then you will learn about giv­ing good talks by think­ing about what makes bad talks bad.
• Ideas for your the­sis or next paper may well come out of an idea you have while sit­ting in a sem­i­nar. All of my papers on func­tional data analy­sis came about after I heard Jim Ram­say give a talk at a con­fer­ence. I didn’t even know func­tional data analy­sis was rel­e­vant to my work on mor­tal­ity fore­cast­ing until I heard his talk.
• Go to sem­i­nar din­ners when at all pos­si­ble, even though it might be scary, and no one else is going. Ask the sem­i­nar orga­nizer to add you to the list of peo­ple he/she cir­cu­lates about the din­ner arrangements.
• Go to work­shops and con­fer­ences, so you have a rea­son­able idea of what is hap­pen­ing in other parts of sta­tis­tics and econo­met­rics. It is amaz­ing what can become rel­e­vant to your research. You won’t believe it until it hap­pens to you. And it won’t hap­pen to you unless you go.

Tags: seminars