Data Analytics How to Use Graphs to Present Your Data Smartly
“Data by itself is useless. It is only useful when you apply it.”
In Data Analytics, when we say data, these involve numbers or texts or symbols that represent some pieces of information. More often than not, we can see the numbers. Because numbers are involved, it is easier to think that it has some values of quantitative or qualitative variables. It must be taken note of that the term “values” is broad enough to cover everything to which value can be ascribed to.
Variables have different types. It is necessary to understand these types so as to know how to make a graphic presentation that really suits its content, nature and treatment of their values. First, there is a quantitative variable. This is also called numerical in the sense that it has a significant meaning as a measurement.
For example are persons height and weight. The best way to contain this kind of variable is thinking of the basic forms of arithmetic such as adding, subtracting, multiplying and dividing. Of course, symbols may be used to denote a value of a certain number in its place and stead. This can further be classified as continuous and discrete.
A continuous variable is a specific kind of a quantitative variable that describes data in a measurable way. If your data deal with measuring a height, weight, or time, then you have a continuous variable. Here there is the interval, and within this interval, any value can be possible. A discrete variable has a finite number of possible values and does not have the inherent order.
Here, every value is not possible. For example, in grading a performance of a product, you may use qualitative values such as 1,2,3 and 4 for the rating but this does not specifically show the real value in its strictest quantitive sense. Statistically speaking, only integer values are possible. Look the example below that talks about a family size.
Second, there is what is called a categorical or qualitative variable. This is not subject to any quantification because it is descriptive or labels. It describes what it means to measure. The examples are gender, colour and name of places.
Because of its definition and nature, it is always discrete. It can be nominal or ordinal. Nominal is used for labelling scales such as gender, the colour of one’s hair and birthplaces. In ordinal, by definition, there is the order which is significant and important although the value is not really known.
For example is the colour of happiness, hatred and so on. It is not concerned with order although there is some kind of order. For example, the red for happiness is greater than its green counterpart depending on how one categorizes its level.
This is the biggest question: How are you going to present your data? Descriptive statistics now comes in. It is the idea of presenting and describing the features of your data. It can be done through various means: graphical representation, tabular representation and summary statistics. First two are called visualization technique. For better understanding the dichotomy of the presentation, it is better to tackle the overview of descriptive vs. inferential statistics.
Descriptive statistics are used to present quantitative descriptions in a manageable form. This is a way to see something meaningful of data at hand. In short, you make a statement based on, about, and derived from these data. As a limitation, you are not allowed to make conclusions beyond the data at hand. You cannot make inferences or generalizations.
On the other hand, inferential statistics go beyond the figures. By its name, one can make inferences and conclusions. In descriptive statistics, more that one variable may be involved, that says, that the point of interest is the relationship between or among different variables. One should ask: How does one variable change with respect to other variables?