Distribution of Data in quantitative analysis and research

In order to apply an appropriate statistical analysis, it is important that you know the distribution of data. Suppose you want to analyse your data in order to check its capability. There can be chances where you may have applied statistical analysis assuming your data to be Normal or Not normal, but in reality it can be vice versa. Then your whole analysis can be costly and inaccurate. Therefore,  to avoid such mistake you need to determine the distribution of your data.
There are several ways to check or test the distribution of your data, which we will study in later posts.

Distribution of data

Shape of the data is called distribution of data.

Normal distribution

If the data or set of observation is divided into two equal / similar parts and each part being the mirror of the other is called normal distributed data or normal distribution. It is also known as bilateral symmetrical distribution.

Normal distribution has a symmetric, bell-shaped curve. To define any normal distribution, two most important parameter needs to be defined, that is, the mean of the distribution where most of the population lie’s and secondly is the standard deviation which tells the spread or deviation of the data from the mean position in horizontal axis.

The 68-95-99.7% Rule

All normally distributed data must satisfy following properties:

68% of data lies between one standard deviation from the mean. Between and

95% of data lies between two standard deviations from the mean. Between and

99.7% of data lies between three standard deviations from the mean. Between and

The above rule of thumb for are also known as the Empirical Rule.

Non normal distribution

The distribution of data in which two half are not mirror image of each other or one another, are known as asymmetrical distribution of data. Skewed distribution is one of the example of asymmetrical data.

They are of two types: