The study of the nature of variables and their conceptualization is fundamental to statistical analysis. The results of such identification determine the possibility of selecting appropriate tests and descriptive statistics instruments and, thus, the outcome of the study itself. Strictly speaking, the entire pool of variables can be divided into two groups, including categorical and quantitative variables (Lio & Hu, 2021). Categorical variables are measured at ordinal or nominal levels and define non-numeric, qualitative values.
For example, racial groups, groups of people based on their pizza topping preferences, or groups of patients based on the speed of postoperative recovery are categorical variables. In contrast, if variables explore numerical relationships and are defined at the ratio/interval level, they are quantitative variables. Quick examples include the distribution of student height in a class or the age of employees in a work group.
In healthcare and clinical sciences, conceptualizing and identifying variables is even more critical. Because the outcomes of decisions in the healthcare industry affect both the health and lives of patients and the success and reputation of healthcare providers, the proper conceptualization of variables is particularly important. For a more detailed study, two clinical variables can be selected as parameters of interest: systolic blood pressure and patient ethnicity. On the one hand, systolic blood pressure is a determinant of blood pressure and a predictor of pathologies associated with hypo- or hypertension (Kocyigit et al., 2020). It is a quantitative variable with a precise numerical scale, an apparent zero, and equal distances between neighboring values — meaning it is measured at the ratio/scale level.
On the other hand, patient ethnicity is used to personalize treatment because many diseases are known to be associated with a patient’s genetic and racial background (Javed et al., 2022). It is a categorical (qualitative) variable that is defined at the category (ethnic group) level. There is no zero in this distribution, and distances between groups do not make mathematical sense. The variable is also nominal because the distribution of ethnic groups has no hierarchical order. I prefer to use systematic sampling for data collection because it is relatively easy to implement (select every third member of a random sample), and it reduces the likelihood of systematic errors.
References
Javed, Z., Haisum Maqsood, M., Yahya, T., Amin, Z., Acquah, I., Valero-Elizondo, J., & Nasir, K. (2022). Race, racism, and cardiovascular health: Applying social determinants of health framework to racial/ethnic disparities in cardiovascular disease. Circulation: Cardiovascular Quality and Outcomes, 15(1), 1-15.
Kocyigit, S. E., Erken, N., Dokuzlar, O., Gunay, F. S. D., Bulut, E. A., Aydin, A. E., & Isik, A. T. (2020). Postural blood pressure changes in the elderly: Orthostatic hypotension and hypertension. Blood Pressure Monitoring, 25(5), 267-270.
Lio, C., & Hu, L. (2021). Types of data involved in gender statistics: Qualitative and quantitative variables. UN Stats.