Science encompasses the systematic study of the structure and behaviour of the physical and natural world through observation and experiment, and technology is the application of scientific knowledge for practical purposes. Oxford Reference provides more than 210,000 concise definitions and in-depth, specialist encyclopedic entries on the wide range of subjects within these broad disciplines.

Our coverage comprises authoritative, highly accessible information on the very latest terminology, concepts, theories, techniques, people, and organizations relating to all areas of science and technology—from astronomy, engineering, physics, computer science, and mathematics, to life and earth sciences, chemistry, environmental science, biology, and psychology. Written by trusted experts for researchers at every level, entries are complemented by illustrative line drawings, equations, and charts wherever useful.

SAMPLE RESOURCES

Discover science and technology on Oxford Reference with the below sample content:

A timeline of life science: from single-celled water creatures to sequencing the human genome

Quotations about science and technology from Oxford Essential Quotations

'The Universal Genetic Code' from A Dictionary of Plant Sciences

A biography of Lise Meitner from The Oxford Encyclopedia of Women in World History

A list of mathematical symbols from The Concise Oxford Dictionary of Mathematics

'The planets: orbital and physical data' from A Dictionary of Astronomy

What is the one term or concept that everyone—from students to everyday web users—should be familiar with? Why?

I wish that everyone understood the scientific method, and in particular the unique importance of the controlled experiment as a method of scientific discovery. Children should be taught at school what an experiment is and why it is such a powerful way of discovering the truth. Psychology uses various research methods, but the most powerful is undoubtedly controlled experimentation, not because it is more objective or precise than other methods, but because it is uniquely capable of providing evidence of causal effects.

The defining features of an experiment are manipulation of a conjectured causal factor, called an independent variable because it is manipulated independently of other variables, and examination of the effect of this on a dependent variable, while simultaneously controlling all other extraneous variables that might otherwise influence the dependent variable. In psychological experiments, extraneous variables can seldom be controlled directly, partly because people differ from one another in ways that affect their behaviour. You may think it’s impossible to control for all individual differences and other extraneous variables, but in fact there is a remarkable solution to this problem.

In 1926, the British statistician Ronald Fisher discovered a powerful method of control called randomization. By assigning subjects or participants to an experimental group and a control group strictly at random, and then treating the two groups identically apart from the manipulated independent variable (applied to the experimental group only), an experimenter can control, at a single stroke, for all individual differences and other extraneous variables, including ones that no one has even considered. Randomization does not guarantee that the two groups will be identical but rather that any differences between the groups will follow precisely the known laws of probability.

This explains the purpose and function of statistical significance tests in psychology. For any observed difference, a significance test enables a researcher to calculate the probability that a difference at least as large as the observed difference could occur by chance alone. The researcher then knows what the probability is of such a large difference under the null hypothesis – the working hypothesis that the independent variable has no effect. If the probability under the null hypothesis is sufficiently small (by convention, usually less than 5 per cent, often written p < .05), then it is reasonable to conclude that the observed difference is probably not due to chance, and if it is not due to chance, then it must be due to the independent variable, because all other variables that could explain it have been controlled by randomization.

If this immensely powerful idea were more widely understood, then people would be less vulnerable to illusory correlation, more sceptical about merely anecdotal evidence, and capable of interpreting findings from any survey research, case study, correlational study, observational study, or quasi-experiment with appropriate caution.