The word Statistics can be interpreted in two sense. In plural sense Statistics refers to numerical statements of facts related to each other. In singular sense Statistic refers to statistical methods or the theory of statistics. These methods relate to the collection, presentation, analysis and interpretation of data.
Statistics as a discipline is considered indispensable in almost all spheres of knowledge. Statistics present facts in simple form, reduces the complexity of data, facilities comparison, helps in forecasting and drawal of inference. Statistic is of immense use in planning, business, economics, administration etc. But statistics has got several limitations. It does not deal with qualitative phenomena and individuals. It can also be misused if it is handled by inexpert hands.
Collection of data is the first step among the statistical methods. Internal data are collected from within the organization whereas external data are collected from outside. External data are again of two types (i) Primary Data (ii) Secondary Data. Primary data are first hand information. Secondary data are already collected and processed by some agency or persons and are currently used as second hand information. Primary data can be collected either by direct personal observation, indirect oral interview, mailed questionnaire method, scheduled method or from local agents. Primary data may be collected through census enquiry or sample enquiry. Census enquiry refers to complete enumeration of each unit of the ‘universe’ or ‘population’ samples survey is the technique to study the universe on the basis of a sample. A sample is a finite subset of the population.
Census and sample studies have their relative merits and demerits. The selection of the method depends upon the nature and extent of the enquiry, degree of accuracy desired, cost and time factor etc. Sampling can be either probability sampling or Non-probability or purposive sampling. Probability sampling can be of five types (i) Single Random sampling (ii) Systematic sampling (ii) Stratified sampling (iv) Cluster Sampling (v) Multi-stage sampling. Secondary data can be collected either from published sources or from unpublished sources.