Glossary Items Starting with "E"

ecological validity: Experiments achieve ecological validity when they reproduce accurately real-life situations, thus allowing easy generalization of their findings. See external validity.
effect size: An index of the size of the statistical difference between groups that is expressed in standard deviation units.
effective range: Characteristic of any dependent measure; the range over which the dependent measure accurately reflects the level of the dependent variable.
electronic mail: A mechanism for sending messages via the Internet to anyone that has an electronic mail address.
E-mail: See electronic mail.
empirical: Based on observed data. For example, a relationship between variables is empirically established if it has been observed to occur.
empiricism: System of knowing that is based solely on observation of the events.
enumerative data: Synonymous with nominal data.
equal intervals: A characteristic of the abstract number system in which the differences between units are the same anywhere on the scale.
equipment subsection: See apparatus subsection.
error bar: An addition made to either histograms or frequency polygons that indicates the size of the standard error of the mean.
error term: Generic term used in many different statistics; it provides a basis for comparing observed differences between groups. The error term is usually based on a measure of the variability of scores within each group.
error variance: A function of the variability of scores within groups.
ethical checks: A series of questions that a researcher must ask about the research and the specific procedures included to safeguard participants.
evaluative biases of language: Language has a tendency to blend description and evaluation, which can distort the perceptions of objective behavior.
ex-post-facto design: Nonexperimental research design in which the current situation of the participant is observed and related to previous events. Because there are no manipulations of variables, confounding variables cannot be controlled and alternative hypotheses cannot be ruled out. Therefore, it is a weak design and causal inferences cannot be drawn from it.
ex-post-facto fallacy: Error in reasoning in which one assumes that the observed relationship between current events and some historical events represents a causal relationship.
ex-post-facto reasoning: See ex-post-facto fallacy.
ex-post-facto study: See ex-post-facto design.
exact replication: Repeating a study by using exactly the same procedure used in the original study. See also replication.
experiment: High-constraint research procedure in which participants are randomly assigned to two or more conditions and compared on a dependent measure. Experimental designs provide adequate control over virtually all possible confounding variables.
experimental analysis of behavior: Procedures for the controlled study of single individuals or small groups, which are based on B. F. Skinner's operant conditioning concepts.
experimental design: In experimental design, participants are assigned to groups or conditions without bias, such as with random assignment, and all appropriate control procedures are used.
experimental group: Group of participants assigned to one or more conditions defined by a specified level of the independent variable. Contrasted with a control group.
experimental level of constraint: Research in which two or more groups or conditions are compared on at least one dependent measure. Experimental research provides adequate controls for most confounding variables and, therefore, allows the researcher to draw causal inferences.
experimental research: See experimental level of constraint.
experimental variance: Variability among the group means in a research study.
experimentation: Process by which a researcher studies the relationship between independent and dependent variables by systematically manipulating the independent variable and observing the effects of the independent variable on the dependent variable.
experimenter bias: Any effect that the expectations of the researcher might have on the measurement and recording of the dependent variable. Uncontrolled experimenter bias threatens the validity of research.
experimenter effects: Any behavior of a researcher that might affect the behavior of participants or the measurement of dependent variables. Experimenter expectancies can create powerful experimenter effects.
experimenter expectancies: Expectations of the researcher about the relationship between the variables being studied. Experimenter expectancies may affect the accuracy of the observations, especially when judgments are required.
experimenter reactivity: Any action by the researcher other than the manipulation of the independent variable that tends to influence participants' responses. A type of experimenter bias.
exploratory research: Low-constraint research designed to investigate feasibility and to generate, rather than test, hypotheses.
external validity: Extent to which the results of a study accurately indicate the true nature of a relationship between variables. If a study has external validity, the results are said to be generalizeable.
extraneous variable: Any variable other than the independent variable that might affect the dependent measure. Extraneous variables are potentially confounding and must be controlled.
extraneous variance: Variability in scores on the dependent measure that can be accounted for by the effects of extraneous variables.