To contribute something new in the field, change something, or expand base knowledge of the field. To tell others what we have found and convince them that the results are convincing.
The abstract comprises a concise summary of the paper, typically 100-300 words. Each sentence in the abstract should relate to an individual section–ie, the first few sentences should summarize the introduction, followed by a summary of the methods, results, etc. The abstract also needs to explicitly define the question that the paper is addressing.
The introduction should provide background and context about the field. The writing in this section is often similar to what you would find in a textbook, in which the start is often broad in scope, and gradually becomes more focused and less narrow in scope. In order to maintain your focus on the topic, your writing should address why this particular question is of interest to the field. It is also important to discuss previous studies that have already been done that relate to your topic. Papers are evaluated based on how well they build upon preexisting science. Citing existing research also demonstrates that you have an understanding of your field.
This section can vary depending on the area of research. For example, in the physical and life sciences, some typical things included are the materials, reagents, and assays used. In computer science, some common things included are the general framework and measures of performance.
Deciding which methods to omit and which to include generally involves taking into account which journal you intend to publish your paper in. Some journals ask that only the most crucial methods be included in paper and ask that the non-essential methods be omitted or put into the supplementary section. Theoretically, any scientists should be able to replicate your experiments from the paper alone. However, in practice, few papers actually describe the complete procedures for all of their papers. Often, the procedures are summarized or omitted because they are common to many other research groups.
When providing results, give a summary of the data, not the raw data itself, and package it in a way that is easy for the reader to understand what your major results were. In your results, do not interpret the data — no conjectures on the mechanism or reasoning for your results. This goes in the discussion section!
Use figures to supplement results section. Important factors here include the p-value, power and/or clinical significance. When using proofs, use lemmas and results to show the logic in which you prove the initial argument. Also, make sure to provide good evidence: legitimate citations and statistics. Lastly, your text should be independent of graphs and figures.
Results: Summarizing Data in Text, Figures, and Tables
In the results section, you should also include text, figures, and tables to summarize your findings. Remember not all results warrant a figure or table. If the data can be summarized in a couple sentence, then a chart is not necessary. All tables and figures require a title (i.e. Table 1 or Figure 1) and a caption that summarizes what the table or figure conveys. A good figure or table caption will guide readers through the main findings without them reading through the main text in the paper.
When you use text to summarize data using words, remember to include numbers in the text and quantify the differences, describe patterns and report observations. Text should be independent of any tables or figures and contain all relevant data. One good rule of thumb to follow is to write your text as if there are no tables or figures to supplement your results section.
For tables, give lists of numbers or text in columns and rows. Tables are useful for illustrating differences between groups. However, do not use tables if you want to illustrate a trend. A better representation of trends is to use graphs instead.
Figures are all visual representation of data. This includes graphs, maps, images and diagrams. Graphs are the most common figure, and as aforementioned, it is best to illustrate trends. There are many common graphs and they each have a specific purpose. Bar graphs are useful when you want to compare the value of a single variable across groups. For example, you can use a bar graph to compare average weight of pigeons in New York, Tokyo, and London. Histograms are useful when you want to show how your measured individuals distribute among the measured variable. You can also use scatterplots to show individual measurements across two variables. Lastly, line graphs are helpful when you want to show how Y (dependent variable) changes as a function of X (independent variable). The two variables must be from the same source.
Before delving into a discussion, briefly summarize what your experiment is. In your discussion/conclusion section, draw significance from the results shown in the ‘Results’ section. Answer questions like: Why do you think you got the results you did? What mechanism do you think caused this?
Also, include a summary of your most relevant data and compare and contrast with existing knowledge. Then use this data to draw meaningful conclusions that support your hypothesis. Discuss the implications of your results and also suggest possible directions for your field of research in the future
What kind of strategies would be used in a “good” scientific argument?
According to John Lokvan, the principal investigator in Paul Fine’s laboratory, he suggests that in an ‘ideal’ scientific argument: logical pieces fit together perfectly; there is no room for interpretation. “Good” scientific argument then one that recognizes where there are methodological shortcomings and discusses how these may affect conclusions or how they may motivate future work.
According to Isabelle Guyon, a good written argument isn’t always about how well written the argument is. It is also be about how timely the argument is presented to the public. In other words, timeliness refers to whether the audience is ready or not ready to accept the idea and whether or not they will resonate with the work presented.
According to Gary Hradek, you would want to weave your arguments in science into a story. You want to conduct experiments that will give data that can be tied together in a way that lets you tell the story. The goal is to gather multiple pieces of evidence that all contribute to substantiating your story or argument beyond a reasonable doubt. He also says that you should write scientific arguments adversarially, that is, writing it in a way that states your view is right, and alternative models are wrong. This is because scientific arguments are derived from experimental data. You do so by writing how your argument is correct, rather than arguing that the other model is incorrect. Though the majority of the scientific argument should be based on data, the way you write the argument is also important. You need to convince your peers that the finding in your data is worthy of being considered, even if your data is very convincing. You want to choose the argument that you know reviewers (people who review your scientific journal article before publication) and other scientists will find the most relevant or appropriate.
According to Leah, it is important to keep your arguments short, brief and clear. There is a page limit. For example in the methods section, consider using prepositional phrases to start your sentence. Even though it is grammatically incorrect, it saves space and the most important concern is that the audience needs to understand what is being said. Furthermore, to strengthen your argument, you should include citations from previous papers. Quoting other papers impart legitimacy and professionalism to your paper.
What argumentative tactics are useful in academia?
John Lokvan says, have your supporting points well organized and you have present them clearly in a manner that is easy to follow.
What is a good structure?
John Lokvan explains that a good structure is a major part of effective scientific argument. It must be easy to follow if it is going to be effective. The structure also must force scientists to reason arguments, and the structure should help the logic flow of the argument.
According to Diane Nava, a PhD candidate in Vision Science at the Berkeley School of Optometry, she states that the structure is linear in science. This differs from other fields, such as Computer Science, where the structure is more modular—every part connects to other fields. Therefore, it is important to follow this linear structure in scientific arguments, and recognize that your argument is only focusing on one specific area in a larger issue.
How do you write about a controversial topic?
According to Gary Hradek, a retired neuroscience researcher at UCSF’s Otolaryngology department, the most important strategy is to ensure the controversial viewpoint does not stray too far from the data. The interpretation of the data cannot be too far removed from the data itself. For example, some scientists, argue that data that is statistically significant, but the effect size is small. This reduces the strength of the argument.
Is usage of good logic important in a scientific argument?
According to Hao Gong, it is especially important to place emphasis on logos. He says: be rational, talk about facts, focus on facts, don’t use anything emotional, and be straightforward. Additionally, Hao Gong encourages writing arguments that can be driven and backed up with good data.