Statistics is the profession concerned with collecting, analyzing and presenting data. It encompasses everything from simple counting methods to highly complex mathematical systems. The history of statistics can be traced back to the fifteenth century when the theory of probability and the study of chance were first considered. At that time, international commerce and trade were of vital importance to many European nations. The loss of valuable freight to storms, piracy or theft led to the concept of insuring cargo against loss. Wealthy individuals, in return for a premium, agreed to compensate merchants in the event the cargo was lost. The insurers began estimating the number and types of risks involved in insuring the ships, and the first studies on the theory of chance were undertaken.
The first mathematician to make an organized study on chance was the sixteenth-century Italian mathematician Girolamo Cardano. An avid gambler, Cardano applied his knowledge and skills at dice games to write Liber de ludo alaea ( Book on Games of Chance), which contained a mathematical basis for figuring odds in a variety of games. The book introduced the concept of probability, asserting that aside from mere luck, certain laws and rules govern any issue. In 1654 the study of probability was continued by two French mathematicians, Blaise Pascal and Pierre de Fermat. A certain French count, the Chevalier de Méré, was fond of gambling and often posed gaming problems to his friend, Pascal. Pascal's and Fermat's solutions to these problems created a new branch of mathematics: probability theory. By the late 1700s, probability and statistics had become increasingly important. Napoleon Bonaparte (1769-1821), himself intrigued by numerical data, established a Bureau de Statistique. By 1801 the first censuses of France and England were taken.
Pierre Laplace was an important pioneer in the study of probability, and his research influenced many later mathematicians; his 1812 paper, Théorie analytique des probabilitiés, was particularly significant in the development of the field. After his death, Laplace's student, Siméon Denis Poisson, carried on Laplace's work. In 1837 he developed what became known as the Poisson distribution while investigating the form of the binomial distribution for large number of trials. He also generalized the Bernoulli law of large numbers, which states that in a large number of trials, the ratio of successes to total outcomes will equal the probability of success.
Statistical studies continued under English mathematician Augustus De Morgan (1806-1871). Although De Morgan's main interest was logic, his use of statistics to predict life expectancies formed the basis of approaches still used in the insurance profession. By 1800 several new statistical sampling methods had been developed and information on populations, vital statistics and life expectancies continued to be collected. Concerned over errors in measurement, Carl Friedrich Gauss directed his mathematical talents to the solution of that problem. He developed procedures for statistical analysis and created an error function used to describe deviation from the norm. Named in his honor, the term Gaussian distribution, also called the normal distribution, refers to a fundamental function of statistics. Influenced by the earlier work of Laplace was another French mathematician, Lambert Adolphe Jacques Quételet (1796-1874). For ten years beginning about 1825, Quételet wrote several papers on social statistics. His book Sur l'homme et le développement de ses facultés, essai d'une physique sociale, is considered to be the first true statistical work ever produced. He introduced the concept of the "average man," a statistic that remains very important to modern day public health and insurance professions.
In the late 1860s, Sir Francis Galton began researching the distribution and inheritance of individual characteristics in man, demonstrating the importance of applying statistical methods to biology. Unfortunately, some of the subjects Galton chose for his statistical analysis did not lend themselves to objective treatment. His attempts to map the distribution of good looks in England and to test the effectiveness of prayer were two such examples of his poor subject choices. However, Galton's ideas were renewed by British mathematician, Karl Pearson, who was interested in developing statistical methods for the investigation of evolution and heredity. His invention of standard deviation, the Pearson coefficient of correlation, the coefficient of variation and the theory of random walk were just a few of his accomplishments. However, Pearson is best remembered for his 1900 invention of the chi-square test, which is a highly significant statistical tool.
The subject of errors in statistical research was addressed by William Sealy Gossett (1876-1937), an English brewery worker. Hoping to solve problems encountered during various brewery processes, the company sent Gossett to study under Pearson for a year. Upon returning Gossett introduced several successful improvements at the brewery. In 1908 he published his most important achievements: a study of probable errors and a small sampling test, which became known as Student's t-test of statistical hypotheses--so named for Gossett's pseudonym, "Student." With the turn of the century, statistical analysis research methods were revolutionized by the work of English biologist Ronald Fisher. Besides improving several existing functions and tests, Fisher's achievements in the areas of random sampling and the analysis of variances remain valuable in current statistical work. Statistics continues to play an important role in many professions. It is employed by physical, biological and social science workers; engineers, business managers, government officials, market analysts, and many others.
This is the complete article, containing 875 words
(approx. 3 pages at 300 words per page).
