Introduction to Statistics

The main criteria for selecting a sample will be that the sample is representative of the population and that there is no or very little subjectivity in the choice of the sampling units.

Sampling is not only conducted by survey researchers on human populations, but also by auditors on a company’s accounts, by agricultural researchers on different pieces of land, and by quality control inspectors on products in a factory, to name only a few examples.

The easiest form of data is called categorical, or qualitative data.

Categorical variables and data can be either nominal or ordinal.

Quantitative variables and data can be either interval-scale and ratio-scale continuous data.

Statistics in medical research is often called Biostatistics.

Data Science is a field that includes Statistics as well as areas such as Computer Science, Database Management and Data Visualization, for example, and has come into being mainly as a result of the spectacular growth in the amount of available data in this new information world that we live in.

• Commerce, especially online electronic commerce
• Finance, for example share prices on stock markets, all managed electronically
• Insurance, all the premiums, incidents, actuarial transactions in an insurance company
• Biomedicine, especially in genetics, where information is literally exploding as gene-sequencing reveals and codes the total genetic profile of a person
• Transport, for example in the airline industry, all the flights, all the passengers
• Climate data, measurements from tens of thousands of weather stations across the world

This new field has developed out of the availability of very large data sets, often referred to as “big data”.

Data are observations made on anything worth studying, such as people, organizations, products, animals, in fact just about everything that constitutes the world we live in.

The categories of nominal categorical variables have no inherent order (e.g., region of residence), whereas the categories of ordinal categorical variables do have an order (e.g., examination grade).

Data values on continuous interval-scale variables are compared by taking differences, whereas values on continuous ratio-scale variables are compared by taking ratios, i.e. percentage increases or decreases.

Statistical investigations can be observational or experimental.

An observational study involves collecting data from a group of individuals, companies, or other sampling units, and then making deductions about relationships between the observed variables or differences between groups.

An experiment is designed to reveal a causal relationship for example to arrive at a conclusion that a treatment cures a disease, or that having more money leads to taking riskier decisions when faced with investment decisions.

Statisticians are trained to interpret and analyse data, and so are indispensable in many industries, companies and administrations, among others.

The first role of Statistics is to reduce this mass of complex data to a simpler form in order to facilitate understanding and learning from the data. This is often done by making graphical  representations of the data.

Statistics provides the necessary tools in Data Science to analyse and interpret extremely large sets of data, facilitating decision-making and forecasting.

The main skill of a data scientist is knowledge of Statistics, which is the key discipline for producing results from data, supported by technical knowledge such as programming and database skills. 

Analytics, this is a term now used in business circles as a substitute for the word Statistics, but it really means the same thing. The word Statistics is considered by some people, especially businessmen, as a bit old-fashioned, and sometimes even difficult to pronounce! But don’t be fooled: Analytics is a fancy word for Statistics.