Data analysis is a logical follow-up to the data collection process. Collected data has to be analyzed with regards to how it impacts the research questions and/or hypothesis. It is from your data analysis that your findings emerge.
What is data analysis?
The Cambridge Dictionary describes data analysis as the process of examining information, especially using a computer, in order to find something out or to help with making decisions.
Data analysis can also be defined as a process of inspecting, cleaning, summarizing, transforming, and modeling data with an intention of unraveling useful information that can help in arriving at logical conclusions and decisions.
Thus, the main purpose of data analysis is to interpret data meaningfully and use those interpretations to make informed decisions.
Types of data in research
Data can be classified in many ways. Let’s look at some common ways to categorize data.
Primary data refers to all data personally collected by the researchers themselves. It can be obtained through quantitative measurements from lab experiments or surveys or qualitatively through interviews, focus groups or other means.
This is a kind of data that comes from external (secondary) sources. It is comprised of generated data that researchers collect for research purposes. Some sources of secondary data include government and private sector publications, reports and databases, as well as other sources such as books, journals, etc.
Quantitative data is numerical in nature and uses mathematical measurements and statistics. This type of data is rigid and properly defined. For example, variables such as GDP, height, etc can be collected and then analyzed quantitatively. Quantitative data can be analyzed with the help of software like SPSS and Eviews, among others.
Qualitative data are non-numerical, non-statistical, and mostly semi-structured or unstructured in nature. This implies that qualitative data are not collected and processed numerically but mostly in words or literature. Qualitative data can be used to analyze people’s comprehension of their social reality or diverse facets of human life, for example, culture, beliefs, morality, etc. Computer-aided qualitative data analysis software (CAQDAS) such as Nvivo, MAXQDA, Quirkos, Atlas.ti, etc can be used to analyze qualitative data.
Below are some tips on how to effectively write a data analysis chapter.
Understand your data measurement instruments
You cannot analyze what you don’t understand. It may seem a trivial point but it’s an important starting point. It is therefore imperative that you have a thorough understanding of all your data analysis techniques and procedures. This means that you have to take the time to understand the methods of analysis that you are using for your study, especially if you are working with quantitative data (although this is equally important for qualitative studies as well).
For many types of studies, a strong knowledge of statistics will strengthen your data analysis enabling you to determine the statistical significance of your results, among others. It will also help your discussion of findings. Moreover, you need a sound knowledge of statistics to help see select the most suitable of the various measurement techniques available for your study.
Learn how to use statistical software
Having sound statistical knowledge is a necessary condition for writing an analysis effectively but it is not sufficient in today’s computer-driven world. From SPSS to STATA and Eviews, several statistical software have evolved and continue to evolve to help students and researchers with computations that may not have been possible manually or with calculators.
If you have opted for any software for your data measurement, needless to say that you must comprehensively study the operational mode of such software in relation to the techniques the software will be measuring with the data fed into it. This will help you analyze the data precisely.
Present your data and highlight your measurement techniques
Before analysis, the data that is about to be analyzed must be presented in a legible format, usually in a structure list/table/spreadsheet (according to the nature of the data). If you are dealing with secondary data in tabular format, be sure to indicate the source(s) of the data under the table. Presentation of your data makes it easy for other researchers to measure them with the same techniques so as to validate or dismiss your findings.
Start with descriptive statistics
The process of data analysis usually commences with descriptive statistics which can be used to summarize, organize and present the characteristics or properties of data. Then you have measures of central tendency that describe mathematical techniques for makings sense of data sets. These include the mean, median, and mode. Similarly, a data set can be described by looking at its variability / dispersion (how much the data varies from the mean) by using the mathematical variance, standard deviation, and so forth.
After introducing the data via descriptive statistics you can now move to the more rigorous inferential statistics which uses sample data to make generalizations and predictions about the larger population of subjects. Some measurement techniques for inferential statistics include ANOVA, regression analysis, etc.
Include computer printouts
It may be necessary to insert the printouts of all the computed outputs of the various measurement techniques used by the software to process collected data. These can include tables, graphs, charts, etc. Ensure that you accurately analyze every computer output based on your knowledge of both the measurement technique and the software used. When in doubt about any aspect of the analysis, try and seek advice from your peers or colleagues to clarify things further.
Be sure to indicate both the software and software version used under any diagram that explains data. If you have surplus data diagrams that you think will help offer more insight into your thesis, attach them in the appendix section.
Data analysis is an important component of most academic papers. It helps you to identify and understand the patterns, trends, and themes in your data which can then help you when discussing your findings.