Writing Is Not a Report, It’s Thinking

Let me be honest: most academic papers are boring. They hide good ideas behind bad writing. The reader sees a wall of text, a thousand citations, and twenty tables of numbers—and gives up. This is not necessary. Writing is not just reporting your research. Writing is a level of critical thinking. These are my personal ways of article writing, and I am sharing them for your reference. Let me walk you through how I think about each part, from the very first sentence to the last reference.

I always start my introduction with a hook that grabs the reader’s attention, but the hook is not just a clever phrase—it is a clear statement of where the academic conversation currently stands. The reader needs to know immediately what the current debate is, what the gap in that debate is, what my novelty brings to the table, and what my contribution will be to the field. I do not hide these things or save them for a dramatic reveal. I state them clearly and early because my job is to help the reader understand why this paper matters, not to make them work to figure it out. Then, at the end of my introduction, I add a short outline of the paper. This outline is like a promise: here is where we are going, here is the path, and here is why I am taking you in this order. A good introduction is a map that shows the destination and the route. A bad introduction is a fog where the reader gets lost before the journey even begins.

Now let me talk about the literature review, because this is where many writers go wrong. I never make a list of papers. Saying “Smith said X in 2015, Jones said Y in 2018, and Lee said Z in 2020” is not a review. That is a phone book. It tells the reader nothing except that I can copy titles and dates. Instead, I tell a story. I take the theories and weave them into one big picture, showing how they connect, where they dance together, and where they bump into conflict. I show how the theories agree on some points and disagree on others. I challenge some of those points openly—not to be rude to past scholars, but because challenging assumptions is how knowledge moves forward. I position these theories carefully within the larger intellectual landscape, asking questions like: where do these theories succeed, where do they fail, and what have they accidentally left out? Where is the empty space on the canvas? Then, and only then, do I highlight what I add to fill that space. My contribution is not a small footnote or a tiny tweak. It is the missing piece of the puzzle that makes the whole picture make sense. This is how I show a broad mind: not by citing everything, but by seeing the shape of the entire conversation and stepping in exactly where something is missing.

Next, let me explain my approach to data and methods, because this is where some people simply give up and copy what someone else did. I always build a clear framework that operationalizes my big picture and directly addresses my research question. The framework is the bridge between my abstract theories and my concrete numbers. I never copy and paste a method from another paper. Doing that tells the reader one thing clearly: I do not actually understand what I am doing. Instead, I take the time to explain my choices from first principles. Why this data and not that data? Why these variables but not others? What would I lose if I used a different dataset, and why am I willing to accept that loss? Every choice I make is an argument, not an accident. I also need to understand the principle of the method—not just how to run it in software, but how it works under the hood, what assumptions it makes, and how those assumptions fit my specific research question. I show how the method works in plain terms. Then I modify and customize the method to fit my unique problem, because standard methods are rarely a perfect fit for non-standard questions. Good researchers often change standard methods to fit their situation, adding a lag here, dropping a control there, adjusting a threshold somewhere else. Some modifications in methods often happen, and that is not a sign of weakness—it is a sign of thinking. This is not cheating. This is thinking. I write my methods section as if I am generating new findings, not just following an old recipe from a textbook. I want the reader to see that I made deliberate choices, and that those choices shaped what I found.

Then we arrive at results and discussion, and this is where my way is perhaps the most different from what you usually see in journals. I am highly selective about what I report. I do not report every test I ran, every robustness check, every alternative specification. It is not about a statistical report where I dump every number I produced. It is about the story, and my statistics exist only to support that story. If a test does not help tell the story, it does not belong in the main text. I put descriptive statistics—means, standard deviations, correlations—in the appendix because I am not too interested in showing them in the main text. The reader does not need to know the average age of my sample to understand my main argument. Instead, I use visualization whenever I can. A good graph tells a story faster than a table full of numbers, and it stays in the reader’s memory longer. Visualization is the best way to tell the story because humans are visual creatures. If I must use a table, I redesign it completely. The way I format and redesign my tables helps reframe the story that the numbers are trying to tell. I bold the key coefficients. I group things logically instead of alphabetically. I add interpretive notes right there in the table. I help the reader see what matters without having to squint or cross-reference. Then, in my discussion, I do not just compare my effect size to other studies—”Our coefficient is larger than Smith’s” is not a discussion; it is a measurement. I ask the deeper question: what is the driver behind this result? What mechanism is actually operating here? What did I learn from this finding that I did not know before? That is the heart of my discussion: extracting meaning, not just reporting differences.

My conclusion is not a copy of my abstract, and this mistake drives me crazy when I see it. The abstract is a preview. The conclusion is a reflection. I always include limitations—what my study could not do, where my data fell short, and what assumptions I had to make that might be wrong. Limitations are not confessions of failure. They are honest statements about the boundaries of my knowledge. And I always point to future research. I tell the reader exactly where they should go next if they want to build on my work. This shows honesty and opens the door for others to continue the conversation. A good conclusion says three things: here is what I learned, here is what I could not learn, and here is where you should go next if you want to push this further. That is generous writing. That is how a field grows.

Finally, let me talk about references, because this is where many papers reveal their quality instantly. I always check the quality of my references. I rarely cite anything other than a peer-reviewed article or a book from a reputable press. Working papers, blog posts, unpublished manuscripts—these do not belong in my reference list unless there is an extraordinary reason. And I never put a hundred citations behind one claim. That is lazy writing. That tells the reader that I have not done the work of figuring out which papers actually matter. Selective referencing reflects the quality of my work. It shows how well I know the topic—not how many PDFs I have in my folder, but which three papers actually shaped the conversation. I cite the most relevant work, the most influential work, and the most recent work. That is the triple standard: relevance, influence, and recency. I usually cite no more than three references per claim. More than that is just noise. Instead of grouping many citations together in one pair of parentheses, I break my claim into small pieces. For example, I do not say “Evidence in the US and UK” with two references thrown together as if the two countries are the same. I say “Evidence in the US” with one reference, and then in the same sentence or the next sentence, I say “evidence in the UK” with another separate reference. This is because the two countries are clearly separate contexts, and lumping them together erases that distinction. Breaking claims into small pieces shows precision, respect for the reader, and genuine understanding of the literature.

As for the outline of the paper itself, I have a practical recommendation. If you are a fresher, I recommend sticking to the traditional layout: introduction, literature review, data and method, results and discussion, and conclusion. It is safe, it is clear, and it gives the reader a familiar structure to hang onto. But you do not have to follow it forever. You can tailor the layout to fit your specific research idea. If your two research questions are deeply interconnected—for example, the outcome of the first question is the direct motivation for the second question, so that you cannot understand question two without already knowing the answer to question one—then you should make your own layout. The rule is simple: help the reader follow easily. The layout is not sacred. Clarity is sacred. Everything else is secondary. Writing is not a chore I do after the research is finished. Writing is the final test of my thinking. If I cannot explain my work clearly, I do not truly understand it. These are my ways of writing. I hope they help you write papers that are not just published, but read, understood, and remembered.