Chapter 7 – Experimental Research

a controlled test of a cause-and-effect relationship that researchers suspect exists
Universal Laws
“suggests that a consequent event or condition will always follow a given antecedent”
Statistical Laws
“assert that a specified antecedent condition will be followed by a given event a certain percentage of the time, and they predict the percentage within specific limits”
Placebo Group
a condition in an experiment where research participants believe they are getting the independent variable being studied, but do not.
Placebo Effect
a change in any placebo group
Control Group
a condition (group) in an experiment that does not get the independent variable manipulation or any other manipulation (such as a placebo); a no-treatment comparison group
Spurious Relationships
(nonsense correlations) patterns of statistical, but meaningless occurrence
Alternate Causality Argument
(alternate hypothesis) occurs when some variable other than the one researchers study in an experiment causes the changes observed in the dependent variable
in experimental research, when a researcher “tries systematically to rule out variables that are possible causes of the effects he or she is studying other than the variable that he has hypothesized to be ’causes'”
Manipulated Variable
(active variable, controlled variable) an independent variable that is manipulated by a researcher in an experiment by controlling when or how much of it research participants receive
(groups) groups in an experiment that receive differential exposure to the independent variable
Treatment Group
(experimental group) a group in an experiment that receives a manipulation of the independent variable.
Comparison Group
any group in an experiment against which another group is compared, such as two treatment groups
Attribute Variables
a trait or character of people, such as age or gender
Natural Experiment
a naturally occurring experiment involving no manipulation of an independent variable by the researcher in which one group is exposed to one level of the independent variable and another group is exposed to another level or does not receive it
Observed Variable
an independent variable in an experiment that is observed, rather than manipulated, by the researcher
Random Assignment
(randomization) a procedure used in experimental research in which each research participant has an equal chance of being assigned to any particular condition of an experiment
in experiments, a measurement of research participants on relevant variables that need to be accounted for before exposing the treatment group to the manipulation of the independent variable
Difference Score
(gain score) the score found by subtracting a pretest score from a posttest score
in experiments, a measure of relevant variables that occurs after the manipulation of an independent variable
Threshold Effects
an effect that occurs when changes in a dependent variable may not occur until the independent variable reaches a certain level (threshold)
Experimenter Effects
a potential internal validity threat that can occur in an experimental research when different experimenters consistently administer different manipulations of the independent variable
Double-Blind Procedure
a procedure used in experimental research to ensure that those who receive them do not know which participants are getting which manipulation
Blank Experiments
introducing an irrelevant treatment in experimental research to keep participants from guessing the true purpose of the experiment
John Henry Effect
a type of Hawthorne effect that occurs when research participants in a control group take the experiment as a challenge and exert more effect than they otherwise would
Intervening Variable
intervenes between the independent and dependent variables to explain the relation between them or provide the causal link
Confounding Variables
obscure the effects of another variable; when the separate effects of two or more variables cannot be determined
Suppressor Variable
conceals or reduces a relationship between an independent and dependent variable
Reinforcer Variable
one that increases a causal relationship between variables
Lurking variables
a kind of confounding variable—unpleasant surprise when discovered
Extraneous variables
variables that are not the main focus of attention in an experiment but which can have an effect on the variables being studied and potentially compromise any causal relationship found between the independent and dependent variables
Control variable
a variable that researchers try to control
Matched-Pairs Design
participants are matched in pairs on some important characteristic and then one member of each pair is randomly assigned so the first condition and the other partner is assigned to the second condition
any variable controlled for statistically in this manner
step-by-step procedures
Full experiments
demonstrate the highest degree of control because the independent variable is manipulated by the researcher and research participants are randomly assigned to create two or more equivalent conditions; Demands two or more conditions
either manipulate or observe the independent variable and may have one or more conditions; only one condition and uses multiple pretests and posttests as baseline measures to assess changes within those same participants before and after an experimental treatment
demonstrate the least amount of control of the three types of experiments; manipulate or observe the independent variable and may have one or more conditions; research participants are not assigned randomly to them; when there is only one condition, preexperiments use a single pretest; do not establish any baseline comparison
One-group posttest-only design
a single treatment group is exposed to the independent variable and then assessed on a posttest
One-group pretest-posttest design
pretest allows researchers to compute a different score between the pretest and posttest scores
Posttest-only Nonequivalent Groups Design
nonrandomly assigns research participants to a treatment or a control group and then measures them on a posttest; there was no random assignment nor pretest given, so we have to assume that the conditions did not start off equivalent with respect to communication skills or anything else for that matter
Single-Group Interrupted Time Series Design
3 pretests -> Treatment -> 3 posttests; the multiple pretests help to establish an intragroup baseline comparison, which is a way of comparing the same group over time prior to the experimental manipulation
Pretest-Posttest Quasi-Equivalent Groups Design
nonrandomly assigns research participants to a treatment or control condition, measures them on a pretest, exposes on group but not the other to the treatment, and then measures both groups again on a posttest; pretests can only rule out initial differences on the variables assessed; there still could be lost of unmeasured and unknown variables that make a difference for the treatment, but not the control, group on the posttest
Full Experimental Designs
The highest level of experiment in terms of control because the independent variable is manipulated by researchers and there are two or more conditions to which participants are randomly assigned, which rules out initial differences and creates equivalent conditions
Pretest-Posttest Equivalent Groups Design
randomly assigns participants to a treatment or a control group and administers a pretest and posttest; R -> Pretest -> treatment -> posttest; R -> Pretest -> no treatment -> posttest; Because random assignment was used, the researcher can rule out the selection bias and all the many known and unknown initial differences between the conditions that might have plagued the quasi-experimental and preexperimental multiple-group designs; Provides a high degree of confidence that the findings are due to the treatment and not to initial differences between the conditions
Posttest-Only Equivalent Designs Group
Same as the Pretest-Posttest Equivalent Groups design except there is NO pretest; very powerful design because it takes care of the potential selection threat but it does not risk the sensitization threat as that design does
Solomon Four-Group Design
Combines the pretest-posttest equivalent groups and the posttest-only equivalent group designs
what independent variables are called when more than one is being studied
Design Statement
a series of numbers, one number for each independent variable in the study, separated by a multiplication sign
Crossed factor designs
involve having every level of one factor appear with every level of the other factor
Nested Factor Design
the levels of one factor only appear (are “nested”) within a single level of another factor
Mixed Design
when the number of levels for each factor aren’t equal
Design Diagram
a box with one independent variable represented on the abscissa (horizontal or x) and the other on the ordinate (vertical or y)
a design diagram shows researchers all the possible combinations of the independent variables; the cell is each possible combination
Unbalanced Design
when there are unequal numbers
Between-group Designs
those in which one group of participants receives one level of an independent variable and are compared to another group that receives another level
Within-Group Design
one in which a single group of people is tested two or more times
Repeated-Measures Design
the same participants are given multiple treatments of the independent variable and measured after each exposure
Treatment Order Effect
occurs if the order in which the treatments are presented makes a differences; earlier treatments sensitize participants to later treatments
randomize the treatment order
Treatment carryover effects
the effects of each treatment have passed before exposing participants to subsequent treatments
Laboratory experiments
take place in a setting created by researchers
Field experiments
conducted in participants’ natural setting