What is Descriptive Analytics |Explained Descriptive Analytics with Example| 9 Points of Descriptive Analytics

There are four types of analytics segment broadly. Those are Descriptive analytics, Diagnostic analytics, Predictive Analytics & Prescriptive Analytics. So, we’ll discuss what is descriptive analytics.

Among all four types of analytics, Descriptive Analytics is the starting point for any kind of analytics to find out what has happened in the past. Also, descriptive analytics is the key towards storytelling with data. Descriptive analytics helps to established cause and effect relationship.

On the other hand, predictive analytics is about to predict the future depending on the past data and diagnostic analytics is to find out the reasons of an incident by using data measurement and KPI. Prescriptive analytics is to make decision depending on the past and future data.

What is Descriptive Analytics?

Descriptive analytics is all about to working with historical data. Through descriptive analytics, we are trying to find out the past trends or insights through statistical tools. There are many use cases for descriptive analytics which are being used for all industries to take right decision based on the historical occurrence.

There are few tests being performed while doing descriptive analytics.

Mean– To find out the mean or average. In case of Excel, AVERGAE function should be used. In case of R, use mean() function to get mean of a data set.

Median– Median is useful instead of average if the data set contain extreme values. Also, Mean and Median should be deferent. In such cases, median should be considered which is not affected by the extreme values. To find out the median. In case of Excel, MEDIAN

function should be used. Use median() function in R to find out median.

Maximum– To find out the maximum value from data set. In case of Excel, MAX function should be used. Max() function is being used to get the maximum value in R.

Minimum– To find out the minimum value from data set. In case of Excel, MIN function should be used. In case of R min() function is being used to get minimum value.

Standard Deviation– Standard Deviation is to find out the deviation of the data points in a data set. In case of Excel, STDEV function should be used. Sometimes, variance can be performed instead of standard deviation. The main difference between standard deviation and variance is standard deviation is the square root of variance. In excel, VAR function is being used to check the variance of data.

Range– Range is to find out the difference between maximum and minimum value in data.

1^st, 2^nd and 3^rd Quartile– Quartile tool is to break the data into four parts for better understanding of the data. In case of Excel, QUARTILE function should be used. Quartile data can be found in R by using summary() function.

Outlier- Outlier finding and removing it from data set is a crucial part to prepare the data for further analysis in terms of predictive analysis. Outlier can be found through quartile calculation and simplest way is to plot the data through boxplot.

Skewness and Kurtosis- Skewness and Kurtosis is on of the important part of data to be checked while performing descriptive analytics. If data is being skewed, the necessary action can be taken in terms of cleaning the data by the data analyst to perform any further analysis. In case of excel, SKEW and KURT functions are being used to check skewness and kurtosis. For skewness and kurtosis in R, moments library is required. Moreover, to get skewness and kurtosis, skewness and kurtosis functions are being used. Also, hist() function is useful to visualize histogram of a data set.

In R summary() function is useful to get basic descriptive statistics results.

What is an example of Descriptive Analytics?

Here is the dummy data set of employee salaries as an example to perform descriptive analytics.

Salary Data		Descriptive Analytics
89,678		Mean	78,735
66,703		Median	76,634
74,109		Max	141,392
79,159		Min	54,433
141,392		Standard Deviation	24,706
54,433		Range	86,959
80,553		Quartile 1	65,847
56,081		Quartile 2	76,634
79,683		Quartile 3	80,336
65,561

Descriptive analytics with data visualization?

There are multiple factors which are being tested while performing descriptive analytics as discussed before. So, Boxplot is such data visualization solution which helps to find out most of the factors like maximum, minimum, average, median, spread of data, skewness of the data.

Here is the example:

Also, Bell curve is being used to check the skewness and kurtosis of the data which are part of descriptive analytics. bell curve is basically formed with the help of probability distribution method and to check the distribution of a data set. Bell curve works better in case of normal distribution of a data set. It is also called density curve or 68-95-99 as it is showing the data based on 1/2/3 standard deviation.

Histogram is another graphical method to check the distribution of a data set, in other words, frequency distribution. Histogram also talks about skewness and of the data. All the analytics or statistical tools have the option to create histogram as data visualization. In excel, histogram chart insert option is readily available in insert chart option. In R and Python, it is part of data visualization packages.

The trendline is helpful if it is time series data to check the trend. Though it is all about historical data, past business trend and seasonal impact on business can be identified. Also, it is helpful for ROI analysis, price changes, external and internal factors. If you are using excel to check the historical trend of data, simply create scatterplot and add trendline into it. It will give the clear visibility on the trend. Additionally, regression equation can be added into it which will add more value.

So, we have discussed that what is descriptive analytics and how to perform the same. Also, nine points of descriptive analytics explained with example to make the concept clear. Apart from that, many times correlation parameter also checked to understand the relationship among two or more variables with the help of historical data.