Data Selection in Research Methodology

Once you have your research question and hypothesis ready, the next step is to identify the data to prove or disprove your hypothesis.

In a business research study, you examine the effect of one variable on another, resulting in a cause-effect relationship between them. The two variables include the following:

1. Independent (explanatory) variable: The variable that is not affected by the other factors that you are attempting to measure is an independent variable. 
2. Dependent (response) variable: The variable that is influenced by the other variables that are being measured is a dependent variable. 

For research on ‘How does digital usage before bedtime affect the sleeping and dreaming pattern of a person?’, the amount of digital usage is the independent variable, and the number of hours of sleep and the quality of sleep are the dependent variables. 

Let us learn more about data collection from Chris in the next video:

You learnt that in order to select the best identifying data, it is necessary to follow some best practices in your research study. A few of the best practices are as follows:

1. The potential sample data should relate to the dependent and independent variables to prove the hypothesis. For example, in the previous example on digital usage before bedtime, if you use the data during the daytime and not before bedtime, it may not be useful in your research study.  
2. Do not select the sample data based on the dependent variable, as it results in selection bias, which can invalidate the hypothesis. For example, in the previous example on digital usage before bedtime, if you select the sample data based on the quality of sleep (dependent variable), then you might not be able to gauge the effect of digital usage on the quality of sleep. It is probable that the person was tired from the day’s activity, which resulted in poor sleep quality. However, this might not prove or disprove your hypothesis, which is based on studying the effects of digital usage before bedtime. 
3. Be aware of cases where information availability is scarce, as it may introduce its own bias. For example, in the previous example on digital usage before bedtime, if you choose to collect data only for top 100 world athletes, you may encounter high cases of information unavailability for your research study. ​​​​​​

Additional readings

• You can read this blog published by The Office of Research Integrity to know more about the data selection process.

In the next segment, you will learn about the data collection process of the research.