Experimental Psych 2

matched group design

Matched group design (also known as matched subjects design) is used in experimental research in order for different experimental conditions to be observed while being able to control for individual difference by matching similar subjects or groups with each other.

treatments

To manipulate your independent variable, you must expose your participants to at least two levels of that variable. The specific conditions associated with each level are called the treatments of the experiment.

deception

When you either actively mislead participants or purposely withhold information from the participant, you are using deception.

parametric design

When you manipulate your inde- pendent variable quantitatively, you are using a parametric design. The term parametric refers to the systematic variation of the amount of the independent variable.

confounding

Whenever two or more variables combine in such a way that their effects cannot be separated, a confounding of those variables has occurred.

Random Assignment

With random assignment, you assign participants to treatments randomly by picking their names out of a hat, for example. A way to get rid of extraneous variables.

representative sample

A representative sample closely matches the characteristics of the population.

biased sample

A sample having 25 white and 5 orange would not be representative (the ratio of colors does not approximate that of the population) and would constitute a nonrepresentative or biased sample.

Sample

A sample is a small subgroup chosen from the larger population.

directionality problem.

A second reason why it is hazardous to draw causal inferences from correlational data is that, even when a direct causal relationship exists, the direction of causality is sometimes difficult to determine. This difficulty is known as the directionality problem.

external validity

A study has external validity to the degree that its results can be extended (general- ized) beyond the limited research setting and sample in which they were obtained.

multistage sampling

A variant of cluster sampling is multistage sampling. You begin by identify- ing large clusters and randomly selecting from among them (first stage). From the selected clusters, you then randomly select individual elements (rather than selecting all elements in the cluster).

independent variable

An independent variable is a variable whose values are chosen and set by the experimenter.

interaction

An interaction is present when the effect of one independent variable changes across the levels of another independent variable.

Systematic sampling

is a popular technique that is often used in conjunction with stratified sampling. For example, once you have randomly chosen the page of the telephone book from which you are going to sample, you then might pick every fourth item (where k 4). Systematic sampling is much less time consuming and more cost effective than simple random sampling.

nonresponse bias (for mail survey)

nonresponse bias occurs when a large proportion of participants fail to complete and return your questionnaire.

operational definition

operational definition involves defining a variable in terms of the operations required to measure it. Defining variables operationally allows you to measure precisely the variables that you include in your study and to determine whether a relationship exists between them.

primary source

primary source is one containing the full research report, including all details necessary to duplicate the study. A primary source includes descriptions of the rationale of the study, its participants or subjects, materials or apparatus, procedure, results, and references.

volunteer bias

Because a sample made up entirely of volunteers is biased, the external validity of your experiment may be affected; this is known as the volunteer bias.

Carryover effects

Carryover effects occur when a previous treatment alters the behavior observed in a subsequent treatment. The previous treatment changes the subject, and those changes carry over into the subsequent treatment, in which they change how the subject performs.

Cluster sampling

Cluster sampling differs from the other forms of sampling already discussed in that the basic sampling unit is a group of par- ticipants (the cluster) rather than the individual participant

nonparametric design

Manipulating your independent variable qualitatively results in a nonparametric design.

Debriefing

During a debriefing session, you inform your participants about the nature of the deception used and why the deception was necessary.

Error variance

Error variance is the variability among scores caused by variables other than your independent variables (extraneous variables or subject-related variables such as age, gender, and personality). Reducing EV: The principal way to reduce error variance is to hold extraneous variables constant by treating subjects within a group as similarly as pos- sible.

Experimental research

Experimental research has two defining characteristics: manipulation of one or more independent variables and control over extraneous variables.

Generalization

Generalization is the ability to apply findings from a sample to a larger population.

randomized two- group design.

If you randomly assign your subjects to two groups, expose the two groups to different levels of the independent variable, and take steps to hold extraneous variables constant, you are using a randomized two- group design.

correlational relationship

In a correlational relationship, changes in one variable accompany changes in another, but the proper tests have not been conducted to show that either variable actually influences the other.

Random Sample

In a random sample, every person in the population has an equal chance of being chosen for the study. A true random sample allows for the highest level of generality from research to real life.

telephone survey

In a telephone survey, you contact participants by telephone rather than by mail or via the Internet. You can ask some questions more easily over the telephone than you can in written form.

correlational research

In correlational research, your main interest is to determine whether two (or more) variables covary and, if so, to establish the directions, magnitudes, and forms of the observed relationships.

counterbalancing

In counterbalancing you assign the various treatments of the experiment in a different order for different subjects. The goal is to distribute any carryover equally across treatments so that it does not produce differences in treat- ment means that could be mistaken for an effect of the independent variable.

nonrandom sample

In practice, most psychological studies use a nonrandom sample, usually of individuals from a highly specialized subpopulation—college students.

proportionate sampling

In proportionate sampling, the proportions of people in the population are reflected in your sample. In the population example, your sample would consist of 10% Hispanics (500/5,000 10%), 30% Blacks (1,500/5,000 30%), and 60% Whites (3,000/5,000 60%).

face-to-face interview.

In this method, you talk to each participant directly. This can be done in the participant's home or place of employment, in your office, or in any other suitable place.

Experimental/Control group

In this most basic of experimental designs, the group receiving the treatment is called the experimental group and the other group the control group. The control group is treated exactly like the experimental group except that it is not exposed to the experimental treat- ment.

Internet surveys

Internet surveys can be distributed via e-mail or listserves or posted on a Web site. E-mail surveys are easy to distribute but do not permit complex navigational designs

Open-ended items

Open-ended items allow the participant to respond in his or her own words.

simple random sampling

Randomly selecting a certain number of individuals from the population is a technique called simple random sampling.

multiple control group design.

This design is used when a single control group is not adequate to rule out alternative explanations of your results, and it is known as the multiple control group design.

Restricted items

Restricted items (also called closed-ended items) provide a limited number of specific response alternatives. -"never" "always" "rarely"

Sampling error

Sampling error refers to the amount of inaccuracy that is estimated to exist within a sample population of the trait being measured. Therefore, a sampling error is calculated to reflect how accurate the results of a study actually are.

Stratified sampling

Stratified sampling provides one way to obtain a representative sample. You begin by dividing the population into segments, or strata. For example, you could divide the population of a particular town into Whites, Blacks, and Hispanics. Then select at random within each segment.

internal validity

The ability of your research design to adequately test your hypotheses is known as its internal validity. Essentially, internal validity is the ability of your design to test the hypothesis that it was designed to test.

matched-pairs design

The matched-pairs design is the matched-groups equivalent to the randomized two-group design. The hypothetical reaction time experiment just described uses a matched-pairs design.

third-variable problem.

The possibility that correlational relationships may result from the action of an unobserved "third variable" is called the third-variable problem. This unobserved variable may influence both of the observed variables (e.g., cell phone use and the likelihood of having a motor vehicle accident), causing them to vary together even though no direct relationship exists between them

causal relationship,

causal relationship, one variable directly or indirectly influences another. In other words, changes in the value of one variable directly or indirectly cause changes in the value of a second.

main effect

the separate effect of each independent variable is termed a main effect.

factorial design

you may be able to gain more information at less expense by using a factorial design that incorporates two or more independent variables in a single experiment.