Descriptive and inferential statistics and Functions and Limitation of statistic ?

Descriptive and inferential statistics:

Statistics as a subject may be divided into two branches i.e. descriptive and inferential statistics.

1.                 Descriptive statistics:

Descriptive statistics includes the methods and procedures used in the collection, analysis and interpretation of data and expressing the data into various from such as tables, graphs, charts, diagrams and finding averages and other measures which would describe the data.

The purpose of descriptive statistics is to present data in such a way that one can easily draw conclusion.

2.                 Inferential statistics:

Inferential or inductive statistics includes the methods and procedures used to draw inferences (results, conclusion or decisions about the population on the basis of sample data. This branch of statistics includes the methods of estimation and testing of hypothesis.

Functions or uses of statistics:

1.                 Statistics simplifies complex data:

The main function of statistics is to simplify the large mass of data, as it is inconvenient to draw any conclusion from such data. The human mind is not as fertile as to memories all facts (figures) about the problem e.g. it is not possible for a research worker to remember the wages of 500 workers of a firm. If he finds the average wage of the 500 workers, then it is easy to remember.

2.                 Statistics are used for comparison:

Statistics facilitate comparison between two or more variable relating to different time or places. Suppose we are told that whole sale price index in Pakistan is 254 in 2006, then this single figure does not show whether price levels high or how. On the other hand if we are supplied the whole sale price index for 1995 also, then we can compare the two quantities and can decide whether the price level has increased or decreased.

3.                 Statistics studies relationship among different facts:

Statistics establishes relationships between two or more variable relating to different fields e.g. the yield of crop is affected by many factors like soil fertility, quality of seed, amount of fertilizer, amount of rainfall, temperature etc. statistics enable us to study the effect of such factors. Statistics tools may be employed to established the relationship among these factors.

4.                 Statistics provides techniques for drawing inferences:

In most of the enquiries it is not possible to conduct the census survey to study the characteristics of the population e.g. To study the average age of the population of a country. Then due shortage of time, resources and lack of trained investigators, we are unable to study the age of each and every person. In such a case sampling methods may be employed to get the deserved information about the population under study. Thus statistics inference has become the most important branch of statistics.

5.                 Statistics helps in fore casting of future events:

Forecasting of future events has become an important function of statistics. In most organizations plans of development are prepared in advance for future and formulate economic policies by analyzing the past and present conditions. E.g. the government make estimates of the domestic production of food grain and requirement of food grains and then decides the quantum (i.e. quantity) of imports if necessary. For making fore cast statistics provides techniques such as regression (simple, multiple, partial) etc.

Limitation of statistics:

1.                 Statistics does not study individuals it deals with aggregates:

Statistics deals with aggregates. Statistics methods are employed to study the characteristics of the pop as a whole and not of individuals. Suppose we compute average height of the population of students as 68 inches. From this figure it’s not possible to know the height of a particular student named “X”.

2.                 Statistics does not study the qualitative phenomena:

Statistics methods cannot be employed in qualities problems such as honesty, intelligence etc. these characteristics cannot be given numerical expressions e.g. a statement of the type that ”sir syed ahmad khan was a great scholar” cannot be given qualitative expression and cannot be analysed by statistics methods.

3.                 Statistics of results are true on the average or long run:

Statistics laws are not exactly like the laws of physics or chemistry, therefore conclusion drawn from statistical studies are not universally true, but they are true on the average. E.g. when we say that per capita income in Pakistan is RS 1450 per capita income is computed by dividing the total income of the country by the total population. The per capita income therefore represents the average income of the people.

4.                 Statistics is liable to be misused:

Statistics methods are liable to be misused and mishandled. If less export people are appointed for the collection and analysis of the data, it will surely give inaccurate results. Only the persons who have an export knowledge of statistics can efficiently handle the statistical methods.