Measurement levels or scales of measure are classifications that define the nature of the information within the values assigned to the variables. Knowing and having a comprehensive understanding of the level of measurement of one’s variable is critical because each measurement level gives different levels of details. For instance, while nominal provides the least details or description of the data, ordinal shows a more accurate and in-depth characterization. However, ratio and interval provide the highest amount of detail or description of the information (Data Levels and Measurement, 2020). Therefore, they each determine how one analyzes their data as there is a hierarchy in the intricacy and precision of the measurement level ranging from low to high.
Continuous Variables
Continuous variables allude to numeric variables, which have an infinite number of values between two given values. It is a specific type of quantitative variable utilized in statistics in a bid to describe measurable data. For instance, continuous variables include height, eye color, gender, age, weight, number of siblings, or time. Continuous variables are quantitative in nature, meaning that they involve numbers or quantities. This is because they have a fixed interval between the adjacent values and can thus be manipulated mathematically.
Delegate your assignment to our experts and they will do the rest.
Descriptive statistics for continuous variables typically fall into three major categories including shape statistics (for example, kurtosis, skewness), location statistics (for example, mode, median, mean, quantiles), and dispersion statistics (for example, range, standard variation, variance, interquartile range) (Larson, 2006).
Categorical Variable
Categorical variables are representative of the type of data that may be portioned into groups. Instances of categorical variables include age group, sex, race, as well as educational level. These types of data are considered to be categorical because they may be grouped according to the variables that are present in the data. There are two major types of categorical data, and they include nominal data and ordinal data. Nominal data are data types that are utilized to name variables without the provision of numerical values. This type of data is mostly collected using questionnaires and surveys and is descriptive in nature. Instances of nominal data include sex, hair, color, name, etc. Ordinal data, on the contrariwise, are data types with a set order or scale. However, they lack a standardized scale on which the difference in variables is to be measured. Some of the instances of ordinal data include interval scale, Likert scale, etc.
The basic statistics that are available for categorical variables are percentages and counts. Descriptive statistics that are utilized to analyze data for a single categorical variable include relative frequencies, frequencies, fractions, and percentages. However, categorical variables can be represented or displayed in frequency distribution tables, column graphs, or bar charts. These can be some of the best ways to quickly visualize the data to show how the samples are divided up between the different categories.
Conversion of Continuous Variable to A Categorical Variable
Continuous variables can be converted into categorical variables by the process of quantization. The process usually entails creating a partition of the range of the continuous variables, that is, the creation of the disjointed intervals, which conjointly cover the scope of the continuous variable and assigning each data point to the bin where it belongs.
When it comes to the conversion of a continuous variable to a categorical variable, there is a myriad of reasons why one would opt to do so. For instance, one may do so because sometimes the quantitative scale reflects meaningful qualitative differences. In another instance, it makes sense to convert continuous variables to categorical ones when there are few variables of the quantitative variables. Finally, one can also convert a continuous variable to a categorical one when the relationship is not linear. Therefore, it may make more sense to convert the continuous variables to categorical variables, especially when it helps one in understanding the relationship between the variables.
References
Data Levels and Measurement. (2020). Statistics Solutions. Retrieved 26 December 2020, from https://www.statisticssolutions.com/data-levels-and-measurement/
Larson, G. M. (2006). Descriptive Statistics and Graphical Displays. Circulation . Retrieved from https://www.ahajournals.org/doi/full/10.1161/CIRCULATIONAHA.105.584474#:~:text=Descriptive%20statistics%20for%20continuous%20variables,eg%2C%20skewness%2C%20kurtosis).
