Just like picking a project topic, some young researchers have issues when it comes to selecting the appropriate variables that are suitable for their research work. Forgive me but I will like to highlight that this problem is common with people who do not have statistics background. But that is not a problem; I will try as much as possible to simplify this article to help you through your research work.
You need to understand that there are different types of analysis ranging from Chi-square, ANOVA, MANOVA, Correlation, regression and so many other types of analysis. The choice of which analysis to chose for your research is dependent on the nature of research. For some researchers working with primary data, especially for undergraduate level, they prefer to use, simple percentage, and chi square.
You need to understand while discussing from the introduction down to the problem statement and making your decisions for the objective of the study, you must have it in mind that to answer your research objectives you will need your analyzed variables since it is the coefficient of these variables that will be used for the analysis. Thus, the first thing to learn here is how to identify your dependent and independent variables.
At the time you made your choice to pick the topic; for example “taxation as a tool for economic development in Nigeria”
In cases like this, when the dependent variable is not defined, we proxy it, and even in cases where you are dealing with variables that cannot be estimated, you also proxy. Let me give you an example;
You cannot estimate economic growth directly, (by estimate we mean, you cannot measure economic) so we usually proxy it to the nearest macroeconomic variable that best describe economic growth as it relates to the project topic or the problem identified: in most cases, gross domestic product (GDP) is used to proxy economic growth. You can proxy taxation either with domestic income tax (DIT), capital expenditure (CE), and even total revenue (TR).
Now the idea here is that, since your regression line is Y = a + bx, we can then stretch “bx” into bx1, bx2, bx3 depending on how many independent variables you wish to work with.
So when thinking about the variables to use for your analysis, you need to consider the macroeconomic variables that are related to the topic and of course variables you can derive from the proposed topic.
Please note that the reason why it is important for these variables to be derived from the proposed topic is that, if you select variables that are not related, you will end up answering questions that the researcher did not ask; by this I mean, your analysis will not answer you objective and the goal of the research will be defeated, and your supervisior will research the work that you have done. This is why it is important to pick a researchable topic, topic that you can easily identify your dependent and independent variable and topic that you can easily source data to analyze.
NOTE for student: make sure you have data for analysis before submitting your topic for approval.
Make sure the topic you have in mind is researchable before you proceed, otherwise you will get stuck and may have to start from the beginning.
If you have questions please feel free to join us in the comment section and if you wish to contribute your opinion, the comments are open for you too.