CSUF Department of Psychology
Unit Banner

Scientific Explanations are:

Assume cause & effect



As diverse as sciences are, all scientific explanations have a lot in common. Psychological research, just like all other scientific research, adheres to these principles.

Scientific Explanations are Empirical

"Empirical" means "based on the senses." All scientific explanations must be based on empirical observations or experiments (or at the very least, be directly inferred when direct observation cannot be achieved).

Scientific Explanations are Tentative

This means that all findings are subject to change should there be enough evidence to necessitate a change. There is always a chance that the scientific theory is wrong, no matter how much supporting evidence there is for the theory.In other words, nothing is ever proven in science.

Scientific Explanations are Probabilistic

Nothing can be measured precisely; there is always a degree of uncertainty in one's measurement. This is especially true in psychology because humans are amazingly complex. In psychology, keep in mind that any finding is what's likely to be true but because of the impreciseness and variability in human observation, human behavior, and measures we say there's only probability of a finding being true.

Scientific Explanations are Testable

There must exist an outcome that does not support the theory. An example of a non-testable theory is Freud's theory of personality. You can't test the id, the ego, or the superego. Thus, Freud's theory is not scientific. (The same applies to dream interpretation -- certainly not scientific.)

Scientific Explanations are Parsimonious

If there are multiple theories that explain the data, the simplest one is usually correct. This is sometimes called the principle of simplicity. Here's a silly example that shows the principle: If I put on a blindfold and tell you how many fingers you're holding up, you can either explain this by a) telepathy (i.e., mind-reading), or b) that I cheated and saw your fingers. Which is the most parsimonious explanation? Clearly, the 'cheating' explanation is much simpler and requires the fewest assumptions.

Scientific Explanations Assume Cause & Effect

For every effect, there is a cause. If we assume no cause, the explanation is not scientific. In fact, much of psychological research is about finding the causes of human behavior.

Scientific Explanations are General

Scientific findings must be applicable to other situations, to other people, to other scenarios, in other locations, at other times, etc. This is called "external validity." If the finding only happens once and does not apply to anyone or anything else, it's not scientific.