Writing a research publication as a student is more achievable than most think, and more valuable than most realise. Universities at the Oxbridge and Ivy League level are looking for students who think independently, engage with real ideas, and produce original work. A published paper is concrete, verifiable evidence of all three.
Why publishing matters more than you think
Most Singapore students assume research publishing is something you do in university, not before it. That assumption is costly.
Top admissions offices receive tens of thousands of applications from students with near-perfect predicted grades. A 45 IB score is table stakes at Oxford or Harvard, not a differentiator. What separates applicants is depth: evidence that you’ve gone beyond the syllabus, engaged with a field on its own terms, and produced something original.
A published paper does this more credibly than any personal statement paragraph. It’s not something you claim. It’s something you can link to.
Start with a research question, not a topic
The most common mistake student researchers make is choosing a topic when they should be choosing a question.
“Climate change and developing economies” is a topic. “To what extent did the 2015 Paris Agreement alter foreign direct investment flows into Southeast Asian renewable energy sectors?” is a research question. The distinction matters because journals evaluate papers on the quality and originality of the central question. A good research question is specific, contestable, and answerable with the evidence available to you.
Here’s what a well-formed question looks like across three disciplines:
STEM: “Does classroom CO2 concentration affect cognitive performance in JC students? A controlled study across five Singapore schools.” This is narrow, testable, and grounded in a local context that has not been widely studied.
Humanities: “Colonial language policy and linguistic identity in post-independence Singapore: a discourse analysis of parliamentary debates, 1965 to 1980.” This is interpretive, uses a named methodology (discourse analysis), and is anchored in a specific archive.
Law: “The adequacy of Singapore’s Personal Data Protection Act 2012 in regulating AI-generated personal data: a comparative analysis with the EU General Data Protection Regulation.” This frames a clear legal gap, invites a comparative method, and is directly relevant to Singapore’s regulatory environment.
How to find your question: Start with a topic you’re curious about. Read two or three recent papers in that area. Look for a gap, something the existing literature hasn’t addressed, a finding that seems contradicted by other evidence, or a context (Singapore, Southeast Asia, post-COVID) that hasn’t been studied. Your question lives in that gap.
The anatomy of an academic paper
The structure of an academic paper is not arbitrary. Each section does a specific job.
Abstract
A 150 to 250 word summary of the entire paper: question, method, key findings, and conclusion. Write it last. Be precise. “This paper finds a statistically significant correlation (p less than 0.05) between classroom CO2 levels and short-term recall scores” is stronger than “This paper explores the relationship between air quality and learning.” Editors decide whether to send your paper for review based largely on the abstract.
Introduction and literature review
The introduction establishes why your question matters. The literature review maps what’s already known, where it contradicts itself, and where your paper sits in that conversation. It is not a list of papers you have read. It is a curated argument.
For the STEM paper on CO2 and cognition, the literature review would cover prior studies on indoor air quality and cognitive function, establish what sample sizes and methods those studies used, and note that no equivalent study exists for Singaporean school environments. That gap is your justification.
For the law paper on PDPA and AI, the literature review would cover the existing scholarship on data protection frameworks, identify the specific provisions that current literature treats as inadequate for AI contexts, and show that a Singapore-EU comparison has not yet been done systematically.
For STEM papers, the literature review is typically concise and focused on methodological precedents. For humanities and law papers, it is more extensive because you are building the interpretive or analytical framework through which you will read your own material.
Cite primary sources wherever possible. For Singapore-related research, MOE annual reports, data.gov.sg datasets, MAS working papers, parliamentary Hansard records, and A-STAR publications are authoritative and frequently underused by student researchers.
Methodology
This is where STEM and humanities papers diverge most sharply.
In a STEM paper, the methodology section describes your experimental design, data collection process, instruments used, and analytical approach. It must be replicable: a reader should be able to follow your methodology and reproduce your results. For the CO2 study, you would specify the CO2 monitoring equipment used, the cognitive tests administered (for example, a validated recall task such as the Rey Auditory Verbal Learning Test), the number of classrooms and students sampled, and the controls applied (time of day, ventilation conditions, student baseline performance). A study involving 40 participants has very different validity claims than one involving 400. State this honestly.
In a humanities paper, methodology refers to your analytical framework. For the parliamentary debate paper, you would name discourse analysis as your method, explain whose theoretical framework you are applying (Fairclough’s critical discourse analysis, for instance), and describe how you selected and coded your corpus. Are you analysing all debates mentioning language policy between 1965 and 1980, or a stratified sample? Justify that choice.
In a law paper, methodology typically involves doctrinal analysis, comparative analysis, or both. For the PDPA-GDPR paper, you would specify that you are applying a functional comparative method, explain which provisions of each instrument you are comparing and why, and note the limits of that comparison (different enforcement architectures, different political contexts). Law papers also need to be clear about the jurisdiction and the date at which the law is stated: Singapore’s PDPA was amended in 2021 and the amendments matter for any AI-related analysis.
Data analysis
This section is where most student papers are weakest, and where strong ones pull ahead.
For STEM papers, data analysis means applying the appropriate statistical tests to your results and interpreting what they show. For the CO2 study, you would likely run a paired t-test or a linear regression analysis, depending on whether you’re comparing two conditions or measuring a continuous relationship. State your significance threshold in advance (typically p less than 0.05) and report exact p-values, confidence intervals, and effect sizes. Effect size matters as much as statistical significance: a result can be statistically significant but practically trivial if the effect size is small. Use R, Python (with scipy or statsmodels), or SPSS. Know which test is appropriate for your data type. Parametric tests assume normally distributed data; if your data isn’t normally distributed, use a non-parametric equivalent such as the Mann-Whitney U test. Graphs and tables should be clearly labelled, with units, and referenced in the text.
For humanities papers, data analysis means applying your analytical framework systematically to your sources. For the parliamentary debate paper, you would code the debate excerpts for specific discourse features: lexical choices around “dialect”, “mother tongue”, and “national identity”; presuppositions embedded in ministerial statements; and shifts in framing across the 15-year period. Your analysis section should walk through specific examples from the Hansard, quoting directly and explaining what the discourse features reveal about underlying ideology. Don’t just assert a pattern. Show it, with evidence, and explain the mechanism.
For law papers, analysis means applying your chosen comparative framework to the provisions you’ve identified. For the PDPA-GDPR paper, you might structure analysis around three or four key dimensions: the definition of personal data, consent requirements, rights of the data subject, and enforcement mechanisms. For each dimension, you state what Singapore law provides, what EU law provides, how they differ, and what that difference means for AI-generated data specifically. The strongest law papers also engage with case law and regulatory guidance, not just the text of the statute. The Personal Data Protection Commission publishes enforcement decisions; the European Data Protection Board publishes opinions. These are your primary sources.
In all three disciplines, the analysis section should not simply restate your data or your sources. It should interpret them in light of your research question and the existing literature, acknowledge where the evidence is ambiguous, and build toward your conclusion.
Results and discussion
For STEM papers, present your results first, then discuss what they mean. Don’t conflate the two. “The mean recall score in high-CO2 conditions was 14.3 compared to 18.7 in low-CO2 conditions, a statistically significant difference (t(38) = 3.42, p = 0.001, d = 0.87)” is a result. The discussion then asks: what does this mean, why might this be, how does it relate to existing findings on CO2 and cognition, and what are the limitations?
For humanities and law papers, results and discussion are typically integrated. You are simultaneously presenting your analytical findings and interpreting them.
The discussion is where the paper’s quality becomes visible. Strong papers grapple with complexity, acknowledge counterevidence, qualify claims, and are honest about what the data can and cannot show.
Conclusion and references
The conclusion should be brief and specific. Don’t restate the entire paper. Synthesise your key finding and state what it contributes. A useful test: if someone read only your abstract and conclusion, would they understand the paper’s core argument? They should.
References must be formatted consistently. Most student journals accept APA or Chicago. Law papers typically use OSCOLA. Use a reference manager: Zotero is free and handles formatting automatically. Inconsistent references are an easy reason for an editor to decline a submission.
Where to publish as a student
There are several well-regarded journals specifically for secondary and pre-university students. These are peer-reviewed and taken seriously by admissions offices.
Journal of Emerging Investigators (JEI) is peer-reviewed, focused on STEM, and designed specifically for secondary students. Papers are reviewed by graduate-level scientists. Find it at jemerginginvestigators.org.
Curieux Academic Journal accepts submissions across STEM and humanities and is peer-reviewed by undergraduates and faculty.
Columbia Junior Science Journal is STEM-focused and run by Columbia University undergraduates.
Young Researchers Journal is interdisciplinary and accepts humanities and social science work.
For Singapore students with strong empirical research, SSEF (Singapore Science and Engineering Fair) is worth targeting separately. It feeds into Regeneron ISEF, the world’s largest international pre-college science competition. Placing at SSEF is itself a meaningful admissions signal.
Rejection is normal. Most professional academics have rejection rates of 50 to 80 percent across their careers. Submit to multiple journals, incorporate reviewer feedback, and resubmit.
How we help you develop and publish original research
We don’t just help students prepare for exams; we help them build the kind of intellectual profile that top universities find genuinely compelling. Our services include:
- Research mentoring: One-on-one guidance from research question development through to submission-ready draft, across STEM, humanities, and law.
- IB Extended Essay coaching: Expert support for students who want to use their EE as the foundation for a broader research publication.
- University admissions mentoring: For students applying to Oxbridge, Ivy League, and top-ranked universities, we help frame research experience into a coherent admissions narrative.
- SSEF preparation: Structured support for students entering Singapore Science and Engineering Fair, from research design to presentation.
Reach us at enquiries@qeducation.sg or book a consultation through qeducation.sg.
Quintessential is the leading specialist in Oxbridge and Ivy League admissions mentoring in Singapore. Call us today to find out how our personalised admissions counselling can support your application journey.
