Coursera - Writing in the Sciences - Week 5

Continue with Week 5.

Way more delay like a year and two months ago. Good I've keep a todo-list and revisit it from time to time. One good lesson when come to learning, the key thing here is not the distraction but the recovery from interference. One bite at a time. One step at a time. One paragraph at a time. You will not feel the task daunting. Otherwise you will procrastinate and won't be able to get started.

5.1: Tables and Figures
Did I wrote my thesis in the wrong sequence? Dr. Dr. Kristin recommended that you should write in such order:

(1) Tables and Figures. No shell table but actual collected data. These are the core idea or story of your paper.
(2) Results. Basically high-level summary of each table and figure.
(3) Methods. What you have done to achieve the results.
(4) Introduction. Background story of your research topic.
(5) Discussion. Probably the longest part of your thesis. If you've done (1) till (4), then this part should come naturally to you. You should already know what to say.
(6) Abstract. "Abstract means to pull out". Yes, that sounds a bit weird. Pull, in the context here means to extract summarized details from other sections.

In the past, my sequence is reverse from (6) to (1). No wonder I spent so much time rewriting the (4), (5), and (6). In other words, focus on the results first and write the other parts later. What about reading a manuscript, the sequence should be (6), (4), (5), (3), (2), and (1).

Some additional reading.
(1) Clinical Chemistry Guide to Scientific Writing.
(2) Essentials of Writing Biomedical Research Papers. Second Edition by Mimi Zeiger

Why tables and figures are important? Because these are the foundation of your manuscript. The two items should be self-explanatory without looking around the elaborated section like method and discussion.

Between tables and figures, which one should you choose? If you need to visually show trends, patterns, or distribution, pick figures. Otherwise, use table when you need to show precision (like number of decimal points), many values, or multiple variables. However, a table or figure may not be need, sometimes a single sentence is sufficient enough.

It's crucial to understand the anatomy of a table which includes title, legend, data, and footer. Likewise for the anatomy of a figure which contains title, legend, picture (primary evidence), diagram, graph, and label. Best to follow the journal guidelines on layout and styling. One trivial rule in styling, remove grid lines from a table (doesn't look professional), just three horizontal lines. Also, add unit of measurements to your variables, for example, Age (years) or Age (months) for toddler. Lastly, don't add unnecessary columns which clutters the table.

Next, graph. There are may types of graphs.
(1) Line graph shows trends over unit of measurement (time, age, or others).
(2) Bar graphs compares group of data at a time point.
(3) Scatter plots shows relationship between two variables or linear correlation (does A causes B?). This graph shows all the data.

If you graph is too complex or cluttered, maybe you should use a table.

When should you use diagram or drawings? If you need to illustrate an experimental set up, workflow (causal diagram), or anatomy of a human or an animal.

5.2: Results
The main focus in this section is to summarize the data by showing relationships and trends through citing data from tables or figures. Do not repeat the raw data from tables or figures data by data. Just focus on the high level summary or take home messages. Pay attention on complementing (not repeating but may highlight) the data in the tables or figures. In summary, result section tells the reader of what you've discover with the supporting data from tables or figures.

Additionally, one key point is result section is what your data shows and discussion section is that your data means.

What verb tense should you use in this section?
(1) Use past tense for completed action like experiment result. Example is "We found that ......"
(2) Use present tense for what is still to be true like showing the table data. Example is "Table 1 shows ......"

You can mix both in the same paragraph. For example, "We found that ...... as Table 1 shows ......".

5.3: Practice writing results
Revise and edit those part where the author is just reading the table, repeating the result line by line in writing. Only pick and highlight important and interesting statistics.

5.4: Methods
Overview of what have been done and instructions for someone else to replicate the study like recipe. Use who, what, when, where, how, and why checklist to guide you to draft the content. These questions should tell in details of material, participant/subject, experimental protocol / study design, measurements of the research, and analyses.

To make your life easy, reference to other papers if the approach is a general well-known method. Also, use flow-diagram to simply your approach, like participant flow again.

What about verb tense?
(1) Use past tense to report method. For example, "We measured ......".
(2) Use present tense to described how data is presented in the paper. For example, "Data are summarized ......". Why? When you read the paper, the data are still being summarized to you.

Is okay to use passive voice or mix of passive and active voice in this section. A lot of emphasis is on the method or the variables.

Read the BMJ Christmas issue or archive for some humourous and light reading.

5.5: Introduction
Some general rules, typically 3 or 2 to 5 paragraphs, don't focus on general ideas but hypothesis or aim of the paper. Read the details rules recommended by Thomas M. Annesley in his paper, "It was a cold and rainy night" : Set the Scene with a Good Introduction. Following this top-down structure to plan the content.

(1) Background, known information.
(2) Knowledge gap, unknown information.
(3) Hypothesis, question, purpose statement.
(4) Approach, plan of attack, proposed solution.

Or a similar structure.

(1) What's known or background. (paragraph 1)
(2) What's unknown or limitations and gaps in previous studies. (paragraph 2)
(3) Your burning question, hypothesis, or aim. (paragraph 3)
Example of phrases are "We asked whether ......", "Our hypothesis was ......", "Our aim/s were ......".
(4) Your experimental approach. (paragraph 3)
(5) Why your approach is new, different, and important to fill in the gaps. (paragraph 3)

5.6: Introduction practice
Find any papers that interests you and mark these three sections from the introduction section.
(1) What's known or background.
(2) What's unknown or gaps or limitations.
(3) The aims of approach of this specific study.

5.7: Discussion
Start this section by following these four rules.
(1) Answer the question asked.
(2) Support your conclusion (your data, others' data).
(3) Defend your conclusion (anticipate criticisms)
(4) Give the "big-picture" take-home message.

In the discussion section, your writing should answer the question that why should anyone cares?

For further breakdown of the structure of this section as shown below.
(1) Key findings that answers the question(s) asked in the Introduction section.
(2) Key secondary findings.
(3) Context.
(4) Strengths and limitations.
(5) What's next.
(6) The "so what": implicate, speculate, and recommend.
(7) Strong conclusion.

And what verb tense to use? Past tense for any discussion of the result of completed experiment (we found that ......) and present tense for the suggestion of the data (the result suggests ......).

5.8: Abstract
Abstract is the combination of "ab" (out) and "trahere" (pull). This means to "pull out" key point from each sections. Length wise, the paragraph should be around 300 to 500 words. The structure as follows:

(1) Background or context.
(2) Question / aim / hypothesis.
(3) Experiment(s) details on materials and methods.
(4) Key results.
(5) Conclusion or the answer to the question. The take home message.
(6) Implications, speculation, or recommendation. Why should reader cares?

Learning Objectives
(1) Understand how to write the sections of an original scientific manuscript.
The key take here is to understand the anatomy of the manuscript (more on this in future post), tables, and figures. Understanding each part and the visual styling will lead to professional looking and readable paper.

(2) Critique poorly formatted tables and figures.
Use the standard good practices.

(3) Practice writing strong Results and Introduction sections.
You will need to read good papers on these two sections.

(4) Summarize the elements of a Discussion section.
Following the steps in the structure.

No comments:

Post a Comment