Introduction To Behavioral Research, Chapter 8 Of Paul C. Cozby's Text

Front 1

Describe the ralationship between a confounding variable and the internal validity of an experiment.

Back 1

A confounding variable is a variable that varies along with the independent variable so that you cannot determine which variable is responsible for the observed effect. This means that there is a possible alternative explanation. When an experimenter eliminates all alternative explanations, the results of the experiment can confidently be attributed to the independent variable, and the experiment is said to have internal validity.

Hint 1

alternative explanation(s)

Front 2

Name the two forms of the basic, simple experimental design.

Back 2

The simplest possible experimental design has two variables (I.V. and D.V.) and the I.V. has two levels: an experimental group and a control group. There are two forms of this design:
1. Posttest-Only Design
2. Pretest-Posttest Design

Hint 2

Think about when you might apply the "test" in an experiment. Before or after the manipulation of the I.V.?

Front 3

Describe the posttest-Only design of a basic experiment.

Back 3

A researcher using a Posttest-Only design must:
1. obtain two equivalent groups of participants through randon assignment(e.g. experimental and control),
2. introduce the I.V.
3. measure the effect of the I.V. on the D.V by administering a test. (Same test is given to both groups)

Front 4

Describe the Pretest-posttest design of a basic experiment.

Back 4

A researcher using a Pretest-posttest design must:
1. obtain two equivalent groups of participants,
2. administer a pretest,
3. introduce the I.V
4. measure the effect of the I.V. on the D.V. by administering a test.

Front 5

Disscuss the advantages and disadvantages of using a pretest as part of your simple experimental design.

Back 5

Advantages:
1. Tests for quivalency in the groups befor the I.V. is introduced/manipulated. It is not usually necessary to use a pretest for for randomly assigned groups, but it is possible that in a small sample size, randomly assigned groups will not be equal.
2. Experimenter may need the pretest to find the lowest or highest scores of the sample so that they may be randomly assigned to a group.
3. Pretests can be used to measure the extent of change in each individual.
4. Prestests can be used to identify if "mortality rate" (number of subjects that drop out of experiment) is an alternative explanation for the results (did a certain type of people drop out for a specific reason?)

Dissadvantages:
1. It is time-consuming,
2. It is difficult to administer,
3. It can sensitize people to what you are studying so that they may be able to figure out your hypothesis.

Front 6

Contrast the independent group design with the repeated measures design.

Back 6

Independent design - participants are randomly assigned to the various conditions so that each participates in only one group.

Repeated measures design - participants are assigned to all conditions. (E.g. both the experimental and the control levels of the I.V.)

Front 7

Order effect

Back 7

A disadvantage with repeated measures design in wich the order of presenting the tests affects the dependent variable.

Performance on the second task improves merely because of the practice gained on the first task.

Order effects concerned with the passage of time:
1. Practice effect - improvement in performance as a result of repeated practice
2. Fatigue effect - a deterioration in performance as the participant becomes tired, bored or distracted.

Order effects that occur when the effects of the first trestment influence the response of the second treatment:
1. Contrast effect - when the response to the second condition is altered because the two conditions are contrasted to one another.

Solutions to the order effect:
1. Counterbalancing
2. Latin squares

Front 8

Counterbalancing

Back 8

A technique to control for order effect in which the order of the condition is counterbalanced; important in a repeated measures design.

With complete counterbalancing, all possible orders of presentation are included in the experiment.

Front 9

Latin Squares

Back 9

A technique for controlling for order effects withous having all possible orders is to construct a Latin Square.

A limited set of orders is constructed to ensure that:
1. Each condition appears at each ordinal position
Each condition precedes and follows each condition one time.