Learnings

Lessons from random books/courses ^___^

Insecurity

A valuable piece of advice from one of my indirect mentors:

insecurity

It came to mind this morning, when I finally decided to face my long-standing problem at work: uncharted GGY AXIS territory. I was so intimidated at first by using processes that I’ve never used before but have to figure out on my own, but I finally made huge progress today. I realized that sometimes my main time-waster is that I’m scared to do something just because I don’t know how it works yet. Same goes for studying for a major actuarial exam. You just have to take the plunge, dammit! Should’ve done so sooner.

How to write a research paper

2013-0708-phdcomic-my-thesis-is-written.jpg

Dr. Guido David hosted his annual research paper writing seminar yesterday. I was supposed to attend last year but I was uninformed of the schedule for some reason. Fortunately I was able to create my thesis paper simply by observing the format of several thesis papers of graduates and with some guidance from him. Being a professor who produces scientific papers for a living, he insists that years of experience merits him enough credibility to give a lecture on it, haha. Anyway, for the benefit of students who weren’t able to attend the seminar, and for those who want to have a general idea of the research paper writing process, here are the main points of his talk (made 1000% more fun with some writing humor and my go-to gallery of academic woes, PhD Comics).

RESEARCH PAPER FORMATS

  • Thesis/dissertation
  • Research proposal
  • Journal publication
  • Conference proceedings

Thesis/Dissertation

  • In full detail: in-depth literature review, definitions & description of all formulas, even basic ones used in the study
  • Around 40-100 pages, self-contained (built with necessary background, so that even lay-academics can grasp it)

Research Proposal

13002366825_7e1433385b.jpg

  • Proposal for research funding
  • Includes only the most important parts (problem statement, objectives, clearly-outlined methodology, significance, etc)
  • The investigator must prove that he/she knows how to approach the problem and has a clear plan (i.e., the research grant will not be wasted). Preliminary results are helpful.
  • Contains brief related literature review (what has already been done about the problem/using the method) -> put everything very related or similar to your work
  • Required for masters and dissertation

(Scientific) Research Paper

  • New results. Publication-worthy. Content/findings must be strong enough
  • Concisely written. Well-known results/technologies/models should not be explained. Even derivations may skip details.
  • Usually 5-10 pages, sometimes more (for full research paper)

Conference Proceedings

  • Same parts as research paper
  • Not as strong/in-depth. Meant for laypersons in the field

TYPICAL RESEARCH PAPER LAYOUT

  • Title
  • Abstract
  • Introduction
  • Methodology
  • Results & Discussion
  • Conclusion

Title

  • Concise, describes research well enough, distinguishes it from other studies/papers on the same topic
  • Title is specific enough (which model is used, the exact problem being solved)

Abstract

  • The searchable part/quick description of the paper, 200-300 word limit
  • Includes:
    • Brief introduction of concepts (1-2 sentences)
    • Short overview of methodology
    • Summary of major results
  • If topic is obscure/not well known, terms must be defined

Introduction

  • Defines the problem, significance of the study, scope & delimitation
  • Background of the study: brief review of relevant literature to highlight the distinguishing features of the present study. usually just 1-2 paragraphs on relevant lit.
  • Necessary terminology is defined
  • Significance: from a scientific standpoint (you’re contributing new knowledge), not necessarily in terms of societal contributiondone.png
  • Scope: limitations of the paper, restrictions of the models, assumptions
  • Theoretical framework: methods/theorems that are fairly well known (but not well known enough to be excluded)

Methodology

  • Defines steps/approach to solving the problem
  • Most applicable to empirical/experimental studies
  • May use equations/theorems that lead to your main problem
 381743_10151184159063429_372225777_n

Results and Discussion

  • Include most important results
  • Graphs, plots, figures, tables are the preferred way of presenting multiple data
  • Highlight findings that are new or surprising. Discuss in sufficient detail.
  • If thesis: you can put as much non-redundant results as possible
  • Captions: under figure, above table

Conclusion/Summary

  • Repeat most important parts of the study. Include significant findings, pioneering techniques or methods
  • Highlight limitations & future work (extensions)

Bibliography

  • Avoid Wikipedia & other online references not based on publications (no accountability or authorship, could’ve been edited by anyone)
  • References should be in the prescribed format (APA, MLA, etc.)
  • Refer to books and articles from research journals. Unpublished research articles/personal communication may be included too if based on a relatively new topic

Extended Abstract

  • Condensed thesis as guide for the examiner, with preliminary results (deadline usually on defense week)
  • Intro usually not in past tense. Result is never in future tense.

THESIS WRITING TIPS

 

Before anything else, here’s a useful infographic I found about plagiarism. Pretty scary. D:

did-i-plagiarize-the-types-and-severity-of-plagiarism-violations_5418b1efd7f4a_w1500

 Anti-Plagiarism

  • Never copy-paste. Quoted phrases that are cited are acceptable but usually not used in scientific papers
  • Paraphrasing other papers without citation is considered plagiarismselfplagiarism.jpg
  • Do not plagiarize even your own work (self-plagiarism)
  • Safest way to go is to always cite sources
  • Tip: read something, put it down, write from memory and own understanding (your writing will formulate itself!)

Writing Flow

  • Don’t start with introduction. It will give you writer’s block.
    writing
  • Start w/ methodology (try to connect the equations and theorems used ) & results (numerical results, plots, derivations, etc) first.
  • Write in a very factual and formal way.
  • A boring way of writing is okay. It’s not a crime, lol
  • Keep tense consistent.
    • Scientific papers: usually all in past.
    • Thesis: intro & method in present, results & conclusion in past (his general approach)
    • Intro & methodology could be in future
  • After methodology & results, write the intro. Since there are already results, you should have a better idea on which objectives & significance you want to emphasize in the study.
  • Define relevant terms that aren’t general knowledge in the field
  • Review of related literature should be short. Stick to the most relevant references.
  • Write the conclusion. highlight important results. by describing the limitations, you already propose how the study can be extended.
  • Write the abstract last. Compress entire paper in one paragraph, should be readable from layman’s perspective
  • Doesn’t necessarily present results, just mention that the paper was successful was successful in obtaining results.
  • No citations & related lit. No equations.

English

  • Spell-check it, have it proofread
  • Note: not explaining well/writing well could be grounds for rejection of paper
  • Write formally. No shortcuts (&, etc) and abbreviations/acronyms should be defined
  • Numbers 1 to 10 must be written in words, unless numerical value is being highlighted
  • Concise but clear and factual. remove unnecessary lines, words, or redundancies.
  • Avoid vague words that serve no purpose (e.g.: The process was quite fast. The thing that was most relevant…)

FOR JOURNAL SUBMISSIONS

 

Choosing a journal for publication

  • If high quality work, try high quality journals (known in the field)
  • Index of journals: Scopus, Thomson Reuters, etc
  • Local: Matimyas, Phil. Journal of Sciences(?), Phil. Computing Journal(?), etc
  • Multidisciplinary work: choose where the work will be more appreciated (e.g.: for math finance work, submit to applied math journals)
  • Try the journals where the papers you referred to were published

Condensing Thesis to Journal format

  • Include very main result. Choose representative values for multiple similar data
  • Ask adviser to help you remove inessential stuff (most likely what I’m going to do, haha)
  • Paraphrase your own writing!

Citations

  • 20-40 is a good number for research papers. (thesis papers usually around 10)
  • Shows that the author did their homework and are very familiar with the topic
    how_to_write_a_research_paper_by_kaidafaye-d4tj79q

    How NOT to write a research paper

     

I’ll work on transforming my own thesis into a journal submission soon. I hope I’m ready! XD

Learning how to learn

metalearning.jpg

One of the online courses I took recently was about learning how to learn, hosted by Dr. Barbara Oakley. Theoretically, if a person knows how they learn best, they will be able to maximize their learning. I was right to finish this course first before the tougher ones! My only regret is that I hadn’t come across it before high school, I would’ve done a lot better with my studies way into college. But I’m glad that I’ll be able to apply what I learned into my preparations for my SOA exams in the future.

Here are my main takeaways:
  • There are 2 modes of thinking. The focused mode is responsible for concentrated (as the name suggests) and sequential approach to thinking, while the diffuse mode is for holistic and creative thinking. When a person is focusing on something, the focused mode is activated. And during relaxation when one’s thoughts drift and the mind wander, the diffuse mode is at play. From my understanding of the Myers-Briggs type theory, people are predisposed to have one of these modes as their default approach to thinking and problem solving (Judging = focused, Perceiving = diffuse). To learn something effectively, the mind must switch back and forth between the two modes of thinking.
  • The working memory is limited, and it can only hold 3-4 items at a time. Chunking is tying up multiple ideas/information using a common use or meaning into a compact “chunk” that can be recalled more easily. A chunk is formed little by little, and can get bigger with practice (as the neural connections get stronger). To form a chunk, one must focus intensely on the information, understand the basic idea and gain context, then repeatedly practice using it. One can also compare chunks across different topics to find a common ground and use it to build analogies and metaphors.
  • To master a material, one must make it persist in memory. Testing and retrieval practice produces deeper learning. Re-reading from a material should be utilized during practice, not during the information absorption stage itself, as it might lead to an illusion of competence (thinking that one has a solid grasp of the material/solution just by looking at it).
  • Sleep and exercise is important. It is in sleep that the brain flushes out toxins, information are consolidated into the long term memory, and when ideas are rehearsed/played around with subconsciously to better prepare a person to respond to it in the conscious state. Exercise improves blood circulation of the body (and consequently, to the brain), promotes neurogenesis (birth of new neurons), and allows the diffuse mode to take over (which makes it a good brainstorming activity).
  • Einstellung (mindset in german) is the concept in which a perceived mastery over something becomes a roadblock to the formation of better ideas. One must avoid a confirmation bias and always be open to new ideas (under scrutiny, of course).
  • To learn better, it is recommended to practice interleaving (mixing up different ways of learning) to build flexibility and creativity. My mentor is a fan of this and makes it a point to apply it to the courses he teaches for the benefit of his students.
  • No matter how introverted you are (I’m talking to myself, haha), you’ll learn better with a study group. You’ll benefit from sharing academic resources with others, and more importantly, you have people to brainstorm with to overcome blindspots in your thinking. From experience, simply having a study buddy (hi Ron!) gave me the motivation to stick to a study plan and work on problem sets regularly.

 

Working smarter

tasks.jpg

 

On an ordinary day when I have to go through a personal to-do list, I tend to do the mindless work first in the belief that productivity with small things will snowball into productivity with bigger things. Taking a course on working smart presented me with the idea that this isn’t the optimal way to act. These mindless tasks are referred to as “unproductive work”. It seemed counterintuitive at first because how can work be unproductive? But it merely suggests a shift to a more efficient working system. Unproductive work is a category of tasks which aren’t essential or related to your main goals. I realized that I should tackle on the biggest and most important/urgent tasks first while these resources are at their maximum. Once they’re at their lowest and I would need to mentally slow down, that’s a better time to do the mindless tasks. By adopting a perspective that emphasizes priority, I have become more meticulous with how I manage my time, focus, willpower, and energy (especially mental energy. with the growing list of challenging items on my checklist, it’s the most precious attribute to me right now). The other main takeaway from this course is the importance of self-imposed discipline. This should be obvious: without it, it’s hard to see plans through to the end and the inner resources are wasted instead of invested into things that serve a greater purpose.

Let’s work more efficiently towards our goals this year! 🙂