Data Analysis in Research: A Step-by-Step Guide for Students

Research Data Analysis: A Step-by-Step Guide

This article focuses on data analysis in a research and the steps to be successful in this very important chapter of a research paper. A final year project or any other type of academic research is composed of different parts. Although they are distinct, these parts are complementary and aim to achieve a single objective: discovering and interpreting the facts of a given reality.

Data analysis is one of the last and most important parts of the structure of academic research. It comes right after data collection and before conclusion. Doubts or insecurities about data analysis are constant, especially among young researchers. It is for this reason; this article gives you tips on how to successfully analyse data for your research paper.

What is data analysis in academic research?

As pointed out earlier, data analysis is a very important phase in any kind of academic research. It is from this that the results and conclusion of the research are elaborated. The objective of data analysis is to understand the data collected so that it is possible to find answers to the investigated problem. And confirm or deny the formulated hypotheses. It is, at the same time, a process of description and interpretation. Thus, data analysis is characterized as a process of meaning formation. Several experts have divided research writing into three major parts namely exploratory, fieldwork, and material treatment.

  1. Exploratory: This is the point where the researcher seeks the theory relevant to the subject and what is the appropriate methodology to carry out the investigation.
  2. Fieldwork: This is the point at which, although referring to Field Research, talks about data collection in general; and finally
  3. Material treatment: This is the stage where the researcher systematizes and analyses the data previously collected based on the theory and methodology proposed at the beginning of the work.

As you can see, the processes performed earlier culminate in data analysis. Therefore, it is not possible to reverse the order or even ignore some of the parts of the cycle. This is because data analysis must be compatible with the type of collection method chosen; and the type of collection method must also be in accordance with the theoretical framework that underlies the work.

In short:

The purposes of data analysis for scientific research are:

  • Understand the collected data.
  • Confirm or not the hypotheses formulated at the beginning of the work; and finally,
  • Expand knowledge about the researched subject.

Treatment of material and data analysis in research paper

In addition to dividing the research cycle into three parts, it is also important to point out that the treatment of the material also has subdivisions.

1. Ordering

We can say that this is a familiarization stage, since it is the first contact the researcher has after data collection. But ordering is not only the first view of the dataset, but also the process of organizing them. After all, the organization is an essential point for academic research to be successful. Without it, the researcher cannot present the data in a clear way or interpret them reliably. Thus, in this first moment, the researcher identifies the type of data (qualitative or quantitative) and selects the most appropriate way to present it (in table, graphs, transcripts, etc.).

2. Classification

Once the data is sorted, it’s time to apply the analysis parameters consistent with the type of research being carried out. It is at this stage where the data is divided into objective categories. Thus, it is essential that the problem and research objectives are well defined and clear to the researcher. They will serve as a guide for ranking the results. After all, you need to look at the results and understand which ones will be considered in your analysis and which ones are beyond the scope of your research problem.

But in these cases, never delete data collected from your survey! Even if they aren’t interesting to your research objectives, that doesn’t mean they don’t matter. During data analysis, other problems are often detected (which may or may not be related to your investigation). In such cases, they can be cited at the conclusion of your research and the full data presented in the appendices of the final paper. Remember that the construction of scientific knowledge is continuous and social. These data may be important for further investigations.

3. Analysis

Finally, after two laborious steps, where the information is condensed, it is time to analyse the data itself. At this stage, the researcher will interpret the data, giving meaning to them from the theoretical framework on which the work is based. Combining data and theory, an analysis is structured to generate a better understanding of the studied phenomenon.

The Obstacles to Effective Data Analysis

It is important to note that challenges of data analysis are numerous. In this session, we examine what are the main problems that researchers face when carrying out their data analysis.

  • The illusion about clarity of results

The first obstacle faced is the researcher’s illusion with conclusions drawn “at first sight.” Academic research is a complex process, with rigorous analytical methods, processes, and techniques. Thus, every analysis and conclusion must be supported by results and theoretical concepts. Quick conclusions tend to simplify the data, dealing only with the surface of the data. It is also a clear sign of immature or rushed research work. Hence, to prevent this oversimplification of data, it is necessary that researcher’s backup their data analysis with valid theories and arguments.

  • Forgetting the Meaning of Data Analysis

Take note that, being conscious of the theoretical foundation and justification of the methods and techniques used in the research is fundamental. However, they do not replace the importance that data and analysis have for the research to achieve its objectives. Forgetting the meaning and importance of the results of data collection and analysis is another obstacle that a researcher faces throughout his research. Indeed, one might wonder how possible it is to forget the meaning of results of data collected, however, this is a scene that many researchers can attest to because of the rigours of the work.

  • Difficulty in articulating concrete results with broad and abstract knowledge

Finally, the third obstacle for a researcher is the difficulty of bringing together the theoretical and practical foundations of research. This happens mainly with young researchers, who are going through their first research experience. Improving the theoretical foundation of the research is the best way to overcome this obstacle. One must be able to link the theory and practice of the data. With greater understanding of the researched subject, knowledge tends to narrow and get closer to the results found. Research practice is also an important point to overcome this obstacle.

Tips to achieve effective data analysis

  • Examination of descriptive statistics

A lot of people tend to just dive into complex analyses before they spend time examining the data from a basic perspective. However, the reality is that many times the descriptive statistics provide critical context for your complex analysis, allowing them to be much more interpretable and clearer to understand.

  • Trim your data prior to analysis, making it easier to focus on analysis.

You can either manually delete your unneeded variables (after saving your dataset as a separate set; or by using the data variable set.

  • Never perform analysis on the master copy of your data.

In general, there is nothing to be afraid of while doing analysis, as it is very hard to actually “mess-up” your data while running analysis.  However, with that said, never use your master copy.