Descriptive Statistics: Fundamental and Necessary in Research

Descriptive statistics offers simple summaries of the chosen sample and about the different observations that the researcher makes. The summaries that are made in descriptive statistics can be quantitative as well as visual. They can be of use at different stages of research. In some researches, descriptive statistics form the fundamental basis for initial description and as pre requisite for more extensive research. In some researches, which are simpler in nature, they may by self-sufficient for a specific investigation.

The history of descriptive statistics goes way back and it first technically appeared as a basic tabulation of the population and economic data. In the recent times, it is a collection of summarized techniques and has been formulated under the heading of exploratory data analysis. The box plot is a good example of this kind of a technique.

In the field of business, it offers a useful means of offering summary of varied types of data. For instance, often investors and brokers are seen putting to use historical account of return behaviour by performing analysis that is empirical and  analytical so that more wiser investing decisions can be made for the future.

The most commonly used descriptive statistics is the mean. It is an informative measure of the finding out the central tendency of a variable if it is reported along with its level of coefficient. The confidence intervals that are given to us by the mean help us to have the range of values that are there around the mean where the true mean can be expected to be located. The width of   the confidence interval is dependent upon the size of the sample and the variations of the data values.  It is proven that a larger sample size means more reliable and accurate the mean. If there is more variation in the data, the reliability of the mean goes down. An important element is descriptive statistics is the description of the variable and the shape of the distribution and it tells you the frequency of the values from different ranges of the variable. The researcher is more interested to know how close the distribution to being a normal distribution is. Even a simple descriptive statistics provides information which has relevance in this concern. The skewness and kurtosis would tell the distribution of the data and its closeness to being a normally distributed data.

%d bloggers like this: