مقاله انگلیسی رایگان در مورد آمار توصیفی – الزویر 2019

Statistics is a mathematical science for collecting, analyzing, interpreting and drawing conclusions from a set of data. The first instance of descriptive statistics was given by the “bills of mortality” collected by John Graunt’s in 1662. Around 1749, the meaning of statistics was limited to information about “states” to design the systematic collection of demographic and economic data. In the early 19th century, accumulation of information intensified, and the definition of statistics extended to include disciplines concerned with the biology and the biomedicine. The main important areas of statistics are the descriptive statistics and the inferential statistics. The first statistical method gives numerical and graphic procedures to summarise a collection of data in a clear and understandable way without assuming any underlying structure for such data, the second one provides procedures to draw inferences about a population on the basis of observations obtained from samples. Therefore, the use of descriptive and inferential methods enables researchers to summarize findings and conduct hypothesis testing. The descriptive statistics is the primary step in any applied scientific investigation to simplify large amounts of data in a sensible way. Indeed, the goal of descriptive statistics is to give a clear explanation and interpretation of the information collected during an experiment. For instance, in medicine and biology, the observations obtained by a phenomenon are large in number, dispersed, variable and heterogeneous preventing the researcher from directly understanding it. To have a full knowledge of the under investigated phenomenon, it is first necessary to arrange, describe, summarize and visualize the collected data (Spriestersbach et al., 2009). In this article, we present two statistical descriptive methods: graphical and numerical. The graphs and tables are used to organize and visualize the collected data. Numerical values are computed to summarize the data. Such numbers are called parameters if they describe population; they are called statistics if they describe a sample. The most useful numerical value or statistics for describing a set of observations are the measures of location, dispersion and symmetry. Generally, graphical methods are better suited than numerical methods for identifying patterns in the data, although the numerical approaches are more precise and objective. Since the numerical and graphical approaches complement each other, it is wise to use both. In the following sections, we first introduce the statistical data types (quantitative or qualitative), and then, the way to organize and visualize collected data using tables and graphs. Several kinds of statistics measures (location, dispersion and symmetry) are also discussed to provide a numerical summary of data (Manikandan, 2011a,b,c; Wilcox and Keselman, 2003). Finally, clinical data are elaborated and discussed as an illustrative example.