Tips and ResourcesIn the 21st Century, it appears that most of modern society is faced with the challenged of interpreting and using an abundance of statistical information. It is from this perspective that StatsBusters believes it is critically important for clients to know about statistics.
The word statistics is derived from the Latin word status (meaning ‘state’). The early use of statistics was to compile data about maps and graphs for countries. The Egyptian, Roman, and Roman Catholic empires were using statistics to compile census data on people. In 1662, John Graunt started using statistical information to publish information about births and deaths, and following his work, studies of mortality, disease rates, population sizes, income, and unemployment rates became commonplace.
Today, statistics are necessary to evaluate the effectiveness of pharmaceutical products, transportation safety, and medical procedures. It is also true that scholars in the social sciences attempt to link individuals’ perspectives with social phenomenon through survey work. The Polish Peasant, a work dating back to 1918, was the first substantive sociological engagement with ‘the individual and the social’. [citation] The study of Polish immigrants revealed that for one to understand culture, he or she must understand the individual’s motivations, objectives, and personality. Jurgen Habermas further highlighted the link between the individual and society. Building on the works of Durkheim and Mead, he argued “Individuals owe their identities as persons exclusively to their identification with, or internalization of, features of collective identity; personal identity is a mirror image of collective identity.” [citation]
The point here is to suggest that corporate leaders, researchers, lecturers, and students need to understand statistics for the following three reasons:
- To properly present and describe information (Descriptive Statistics);
- To understand how to make estimates about the characteristics of a population and generate decisions about a given population based on sample results (Inferential Statistics); and
- To be able to predict some phenomenon into the future (Forecasting).
Inferential statistics are significantly more important than descriptive statistics. This is because descriptive statistics can often be subjective, whereas inferential statistical is more objective in analyzing data. Inferential statistics is an outgrowth of probability theory. The foundations of inferential statistics can be traced back to the middle of the 17th Century in a correspondence between the mathematician Pascal and the gambler Chevalierde Mere. [citation] Other mathematicians such as Bernoulli, DeMoivre, and Gauss were also instrumental in the development of inferential statistics.
Inferential statistics can be broken down into four categories as follows (a) population, (b) sample, (c) parameter, and (e) statistic. The population is the totality of items or things under consideration. A sample is a proportion of the population.
Inferential statistical methods are necessary because of the need for sampling. It would be too time consuming and costly to study entire populations. As a result, decisions about a population’s characteristics are based on information derived from a sample. Probability theory indicates whether the results from the sample reflect the results in the population. Formulas such as the proportion for sample size determination are typically used to estimate the adequacy of a research project’s sample size in reflecting the population.
A parameter is a summary measure that is computed to describe a characteristic of the population. And a statistic is a summary measure that is computed to describe a characteristic from only a sample of the population. Parameter estimation refers to statistical models, such as the linear regression model, logistic regression model, and the Cox regression model. Each model has a different procedure, or formula, for estimating the parameters. Hypothesis testing refers to a number of procedures such as the t-test, Chi-square test, and analysis of variance (ANOVA) to examine association between variables
It is hoped that this information has provided the reader with a brief summary of the field of statistics. For further tips and resources about statistics, please do not hesitate to contact one of our analysts at info@statsbusters.com.
© 2006 StatsBusters, LLC. StatsBusters is an American and British cooperative. The United States and United Kingdom member firms are separate and independent legal entities and each describes itself as such. All rights reserved.
StatsBusters Online Privacy Statement and Disclaimer