Answers to Test Yourself
(1) c. (2) c. (3) a. (4) d. (5) Manipulation of an independent variable and control for extraneous variables that can be done with random assignment. (6) The primary difference is that in a time series design, several pretest and posttest measures are recorded and the pattern of scores before and after a treatment or event is compared. In a simple pretest–posttest design, only one pretest and one posttest measure is recorded and the difference between these two scores indicates the effect of the treatment that occurred between the two measures. (7) Answers will vary, but the design should include a pretest and posttest on the same behavior, nonrandomized experimental and control groups, and a treatment or condition implemented for the experimental group between the pretest and posttest measures. (8) Designs will vary, but the study should include surveys of approval ratings each month or so for a period of several months before August 2011 and several months after August 2011. (9) The main sources of bias in a quasi-experiment are history effects (effects of a historical event on the results where the historical event is not the treatment of interest in the study), maturation effects (subjects in the study may develop or age over the course of study if it takes place over a longer period of time and this can influence or affect the results), and attrition or mortality (subjects drop out during the course of the study, possibly biasing the sample that completes the study).