Experimental Research Methods and Designs

Vocabulary practice flashcards covering experimental research characteristics, control of extraneous variables, poor designs, true experimental designs, quasi-experimental designs, and factorial designs.

Experimental Research

The only type of research that attempts to influence a particular variable, and the best type for testing hypotheses about cause-and-effect relationships.

Independent Variable

The treatment or factor that is manipulated and directly controlled by the researcher.

Dependent Variable

The outcome variable that researchers observe or measure after the treatment has been administered to see if groups differ.

Manipulation of the Independent Variable

An essential characteristic of experimental research where the researcher deliberately and directly determines what forms the independent variable will take and which group will get which form.

Random Assignment

A procedure occurring before an experiment begins that ensures every individual who participates in the study has an equal chance of being assigned to any of the experimental or control groups.

Random Selection

A sampling method in which every member of a population has an equal chance of being selected to be a member of the sample.

Extraneous Variable

Any variable not being investigated that is not considered an independent variable but has the potential to affect the dependent variables or controlled conditions.

Holding Variables Constant

A technique used to eliminate the possible effects of an extraneous variable by completely removing it from the study (for example, restricting subjects to females and excluding all males).

Building the Variable into the Design

A technique to minimize threats where the extraneous variable is included directly as distinct groups in the study design in order to assess its effects and interactions on outcomes.

Matching

A technique of pairing subjects on specific variables of interest (such as age) and then assigning one member of each pair to each of the comparison groups.

Using Subjects as Their Own Controls

A design method where the same participants experience every treatment or condition, and their results under one condition are compared with their own results under another to reduce individual differences.

Analysis of Covariance (ANCOVA)

A statistical procedure used to equate groups statistically on the basis of a pretest or other variables, adjusting the posttest scores of the subjects in each group accordingly.

One-Shot Case Study

A poor experimental design in which a single group is exposed to a treatment or event ($X$) and a dependent variable is subsequently observed or measured ($O$).

One-Group Pretest-Posttest Design

A poor experimental design in which a single group is measured or observed on the dependent variable both before ($O$) and after ($O$) being exposed to a treatment ($X$).

Static-Group Comparison Design

A poor experimental design (also termed a nonequivalent control group design) that uses two already existing, intact groups where one receives treatment and the other does not, followed by posttest measurement.

Static-Group Pretest-Posttest Design

A poor experimental design that differs from the static-group comparison design only in that a pretest ($O$) is administered to both intact groups prior to treatment.

True Experimental Designs

Experimental designs that include random assignment ($R$) to equate groups and control for threats to internal validity.

Randomized Posttest-Only Control Group Design

A true experimental design where subjects are randomly assigned ($R$) to an experimental group receiving treatment ($X$) or a control group ($C$), and both groups are measured only after treatment ($O$).

Randomized Pretest-Posttest Control Group Design

A true experimental design where subjects are randomly assigned ($R$) to experimental and control groups, both take a pretest ($O$) and posttest ($O$), and only the experimental group receives treatment ($X$).

Randomized Solomon Four-Group Design

A true experimental design using four randomly assigned groups (two pretested, two unpretested, two treated, two untreated) to explicitly assess and control for the interactive effects of pretesting.

Mechanical Matching

A matching process of pairing two individuals whose scores on a particular variable (such as mathematics aptitude scores and test anxiety scores) are similar.

Statistical Matching

A matching procedure that avoids subject loss by giving each subject a predicted score on the dependent variable based on correlation, using the difference between predicted and actual scores to compare groups.

Quasi-Experimental Designs

Research designs that do not use random assignment, relying instead on other techniques to control or reduce threats to internal validity.

Matching-Only Design

A quasi-experimental design in which researchers match subjects in the experimental and control groups on certain variables without the use of random assignment.

Counterbalanced Design

A quasi-experimental design for equating groups where each group is exposed to all treatments, but each group receives the treatments in a different order.

Time-Series Design

A quasi-experimental elaboration of the one-group pretest-posttest design involving repeated measurements or observations over a period of time both before and after a treatment ($X$).

Factorial Design

A modification of posttest-only or pretest-posttest control group designs that permits the simultaneous investigation of two or more independent variables.

Moderator Variable

An independent variable (which may be a treatment or a subject characteristic) studied in a factorial design to examine its interaction with another independent variable.

Interaction

A condition in a factorial design where the effect of one independent variable on the dependent variable depends upon the specific level of another variable (moderator variable).