Many students and professionals alike who are new in the field of scientific investigation have difficulty writing a scientific paper. Just what are the things required in a scientific paper? This article expounds on the topic by discussing, in simple terms, the elements of a scientific paper. It is written in an easy to comprehend way to provide the reader with the impetus to start off a scientific investigation.
Writing a scientific paper can be a challenge to both graduate and undergraduate students as well as those individuals who are compelled to write such as the faculty members of an academic organization. Many universities require students, especially those in the graduate level, to undertake research as one of their major activities to equip them with knowledge and skills in discovering new things that will help society in general. Faculty members in an organization are coerced, motivated and sometimes threatened just so they will be able to write scientific articles of professional significance as part, not only of their professional advancement, but also to help enhance knowledge imparted to their students.
The findings reported in a scientific paper add to the pool of knowledge in society. Ultimately, the information provided by scientific papers provides a solid foundation for a nation's advancement in various fields of endeavor, particularly the generation of new technology applicable to human lives that can uplift the living conditions of people. In fact, many developed countries invest a lot in scientific investigations to keep ahead of the pack. The United States, for example, allocates an average annual budget of USD 4.9 billion  just for cancer research alone.
How to Write a Scientific Paper
1. Start off with a phenomenon that strikes curiosity.
A scientific paper is written using the scientific method of looking at phenomena observed in nature. A phenomenon may be a fact, event, or anything that strikes the curiosity of the observer. A phenomenon is not easily understood unless broken down into bits of information that can be understood and measured in some way by the observer. A phenomenon is usually investigated by posing a question or a set of questions to clarify the issue at hand. This leads to the next step of the scientific method which is the statement of the problem.
2. Write the statement of the problem.
The statement of the problem is just a question that pins down the specific items that arise from observations made. For example, a faculty member might have observed that every Monday many of his students appear to be tired and perform poorly in class. He asks some of them why and found out that many have been hooked to playing internet games the previous day. He may ask the question, "Is there a relationship between the number of hours playing internet games and student's academic performance?" This question clarifies the issue and lends itself to further investigation and analysis.
3. Review related literature and studies and find gaps in knowledge.
Many students find the review of related literature cumbersome and time consuming. This also becomes a problem especially in third world countries who have no access to scientific journals because many of them are poor and could not afford to subscribe to such journals. This may also be the reason why many third world countries could not easily advance their scientific knowledge. Recently however, there are online journals that make scientific findings available to everyone. While some of these journals these may not be as rigorous as the paid ones, in many cases these just serve the purpose of a budding scientist. These open access online journals are compiled together in the Directory of Open Access Journals which is a helpful source of scientific information for free.
Why is there a need to review related literature and studies? This is to prevent you from "reinventing the wheel." In other words, this will keep you from doing what others have already done in the subject area of interest. Why invent the wheel when this has already been invented? This is done mainly to avoid duplication of work. This will also save you some money.
After you have read related literature and studies and have come up with your own idea about the subject at hand, you can draw up you own conceptual framework or a road map to chart the direction of your research. This is discussed in detail in the author's previous article titled "What is the Difference between the Theoretical and the Conceptual Framework."
3. Write the hypothesis
At this stage, you can now come up with your own hypothesis regarding the issue you are trying to investigate. For example, in the given phenomena discussed above, the hypothesis should be something like:
Null Hypothesis: There is no significant relationship between student's academic performance and the number of hours playing internet games on Sundays.
The null hypothesis is usually written in a "no relationship" form as the benchmark for investigation. This question also indicates the statistical tool to be used to analyze objectively the results of the investigation. To perform statistical analysis, the so-called dependent and independent variables of the study should be measurable. In this example, the variables are 1) student's academic performance, and 2) number of hours playing the internet on Sundays. Student's academic performance may be measured in terms of their student's scores in a quiz given in a subject during the first day of the week.
4. Test the Hypothesis
Testing the hypothesis requires designing a way to investigate the specific variables identified in the study. This begins with the idea of random sampling where samples are obtained in such a way that each of these samples has an equal chance of being selected. In the above case, a random sample may be obtained by listing down the names of all first year students in a university (as these are usually the ones who fall prey to internet games) on a piece of paper, rolled and placed inside a fish bowl. You may want to have only 200 students as your sample as it is impractical to study say a total of 15,000 students in the university. Two hundred samples approximate a normal distribution, a prerequisite in computing parametric statistics, where you can safely make conclusions.
Simple questions like the following may be asked in the questionnaire:
1) Do you play internet games on a Sunday during the semester?
2) How many hours do you spend playing internet games?
To correlate the number of hours to the student performance in the quiz, the researcher may request the student's score from the faculty handling the first subject on Monday.
5. Explain the results and discuss.
After gathering the data for 200 students, the researcher may now analyze the data using say Pearson's Correlation Coefficient. This is easy to compute using common software applications like MS Excel. A detailed discussion on how to do this is explained in detail here. If you get a correlation of something between 0.4 to 0.6, this will be a very good relationship where you can say that indeed there is a correlation between the number of hours of playing internet during Sundays and scores in a subject. You may then relate your findings to what you have read in the literature and studies you have reviewed.
6. Write your conclusion.
The conclusion is just a statement based on your hypothesis. If in the above example you discovered that there is a good relationship between the number of hours spent by the student in playing internet games and their scores in a quiz, then your conclusion will be this:
Conclusion: There is a significant relationship between student's academic performance and the number of hours playing internet games on Sundays.
7. Publish or disseminate your findings.
Writing the scientific paper does not end in your conclusion. To add to the pool of knowledge, you have to disseminate your findings in a conference, research forum, or publish it in a relevant journal for others interested in the topic to gain knowledge and make the necessary remedy to an identified problem in student perfomance.
 National Cancer Institute
© 26 February 2011 Patrick A. Regoniel How to Write a Scientific Paper