Essential Terms and Concepts in Scientific Investigation

A comprehensive collection of vocabulary terms related to scientific investigation and statistics, useful for exam preparation.

Hypothesis

An informed statement of what might be true; a tentative explanation of some phenomenon that can be tested by further investigation.

Theory

A mature, coherent body of interconnected statements, based on reasoning and evidence, that explains a variety of observations.

Prediction

What you expect to see in a data set from an experiment if a given hypothesis is true.

Independent variable

The variable whose effect is being studied, usually manipulated.

Dependent variable

The variable being acted upon by the independent variable, usually measured.

Experimental treatment

Manipulation of the independent variable in a randomly selected set of individuals.

Control group

A randomly selected set of individuals treated exactly like the experimental group in every way except for the independent variable.

Correlation

A measure of the relation between two variables; does not equal causation.

Positive correlation

When one variable increases, the other also increases.

Negative correlation

When one variable increases, the other decreases.

No correlation

When one variable increases, the other shows no significant change.

Mean

The numerical average; sum of all values divided by the number of observations.

Mode

The most frequently occurring value in a frequency distribution.

Median

The middle-most value in a frequency distribution that divides ordered observations into two equal parts.

Frequency histogram

A graphic representation of how frequently different values occur.

Normal distribution

A bell curve frequency distribution.

Standard deviation

A measure of variation in a population.

Standard error (of the mean)

The standard deviation of an estimate of the mean; decreases with sample size.

Sample size

The number of observations in a data set; generally, bigger is better.

Null hypothesis

The hypothesis stating that the independent variable does not affect the dependent variable.

p-value

The probability of getting the observed result by chance, assuming the null hypothesis is correct.

Statistically significant

Generally means the p-value is less than five percent (p < 0.05) and the null hypothesis is rejected.

Not significant

Generally means p > 0.05, often abbreviated as N.S.

Nominal data

Classified in categories, such as blue or brown eyes.

Continuous data

Variable can take any value within the observed range.