Chapter 10 FITB

Question 1

In a true experiment, the experimenter has complete control over the experiment. A quasi experiment is one in which the experimenter lacks some degree of control. The most important difference is that in a true experiment, the subjects are — to conditions, whereas in a quasi experiment, the subjects are — for conditions from —

Answer 1

assigned, selected, previously existing groups.

Question 2

The independent variables of an experiment are sometimes called —, each of which has at least — levels

Answer 2

factors, two

Question 3

In one-factor experiments, the levels of the variables are sometimes called – or —

Answer 3

treatments or conditions

Question 4

The two basic elements of good experimental design are:

Answer 4

1) the existence of a control group or a control condition
2) and the random allocation of subjects to various conditions

Question 5

Within-subjects control of order and sequence effects may be achieved by

Answer 5

randomization, block randomization, or reverse counterbalancing

Question 6

When it is not possible to control for order and sequence effects within subjects, the — technique may be used

Answer 6

Latin square

Question 7

The simplest possible true experiment has — conditions tested within subjects. All subjects experience both conditions in — order

Answer 7

two, counterbalanced

Question 8

— experiments are conducted when the hypothesis is not a simple yes or no question, when determining the shape of a function is desirable, or when multiple rival hypotheses must be ruled out.

Answer 8

Multiple-conditions

Question 9

Multiple-conditions,— experiments are common in perception research, as when one scales the brightness of different intensities of a light

Answer 9

within-subjects

Question 10

— design is used when a significant interaction between conditions would occur if tested –

Answer 10

A between-subjects, within subjects

Question 11

Some designs to avoid are the:

Answer 11

one-group posttest-only design,

the posttest-only design with nonequivalent control groups,

and the one group pretest-posttest design