What is Statistics;
The Online Etymology Dictionary refers that statistics was defined by Barnhart in 1770s as the “science dealing with data about the condition of a state or community”. In Latin, the word statisticus is related to the word Status and thus stata/state. Sir John Sinclair was a politician and was involved into finance and thus he tried to define statistics as“the study of collecting, analyzing, interpreting, presenting, and organizing data” which was included in his 21 volumes he wrote about financial and statistical facts (1791).
The data collection for statistical analysis can be done with multiple ways but these ways can be summarized mainly into four categories/topics:
Many times, this method can be very time and cost consuming because one researcher must find all the people of a e.g. city , the whole population, no one must be left out! This can be a significant problem when data must be collected from city with high numbers of population such as China. Therefore, this method can be very time and cost consuming.
This method can be more useful and is used more often. Here, the sample is significantly smaller than in the previous method. The sample can consist of 20, 100 or 1000 participants. Using such samples, useful conclusions and results can be reported. This method is less time and cost consuming. For example, if you like to explore the music preference on 20 years old persons, you do not need to find the whole population of these persons e.g. on a city but you can take a sample (randomly, non-randomly, stratified etc) of this population, lets say 20, 100 or 500 participants of that age.
The big question here can be what sampling method you must choose in order to have the best result which will be representative of the true result of the total population in that study. Another example can be exit polls that try to find the election result by asking voting preferences. Here, sample size is much smaller than the total population.
Data collection using research methods can take severe time on deciding the design of this research. There are multiple research methods and each research method can require considerably different amounts of time/cost/resources. A longitudinal or cross-sectional/country study can require a lot of time in comparison with a questionnaire type study that will require only university student to be conducted. The latter one is not time and cost consuming.
However, the results of a research method can be very useful in almost all scientific fields which can involve the producing of new drugs, new methods of measuring company performance, of new theories or to decline old ones etc.
The method of observation can be of low cost but it can require time. Also, the researcher/observer must be very careful in order not influence the environment that he / she would like to observe. For example, a researcher / observer may want to observe systematically the eating / diet behavior of a group of wild animals (or people or plants etc.). Using these observatory data, then, he / she can run many statistical analysis in order to reach a useful conclusion on the e.g. eating behavior and the loss of weight. However, he / she cannot manipulate experimental conditions: he / she cannot divide wild animals into groups which one will be given a specific type of diet and the other one a different diet. He / she cannot intervene at all!
So, what is statistics? Statistics has influenced all scientific fields. Some type of questions that Statistics can answer are: