CHAPTER 12
EXPERIMENTAL RESEARCH
TEACHING ACTIVITIES FOR CHAPTER 12
Activity 1. Do experiments in education have value?
Point out to your students that the U.S. Department of Education, through its Institute of Education Sciences
and the No Child Left Behind Act, is strongly advocating high-quality experiments to determine which educational
programs are effective and which ones are not. The assumption is that school systems will adopt programs validated
by experiments and require their teachers to use them.
Point out, too, that teachers traditionally have valued the freedom to form their own philosophy of teaching and
to experiment in their own classrooms. Action research, the topic of Chapter 18, supports this approach, based on
the assumption that teachers will discover what works best for them by doing informal experiments in their own
Activity 2. Doing an experiment in class.
We find that most students have never participated in an experiment. Therefore, they might find it interesting to
participate in at least a brief experiment, and the experience also might help them develop a deeper understanding of
the features of experiments described in Chapter 12.
Here are the procedures for a simple, quick experiment:
1. Give each student three scrap pieces of paper, or have them tear a sheet of paper into three
pieces. Tell them that their responses will be anonymous.
2. Tell the students that you will read the names of 10 objects and you want them to remember as
many of these objects as possible. When the students are ready, say each object slowly and
clearly. Here is one possible list: apple, key, computer, president, armchair, airplane, zebra,
movie, candle, garage.
8. Ask students to write down the names of as many of the objects as they can remember on the first piece
of scrap paper. Give them a time limit of 30 seconds. Then repeat the list aloud and ask them to count
the number of words that they wrote down that were correct. Ask them to write on the second scrap
piece of paper the number of correctly recalled words and write “Posttest” next to this number. Finally,
ask students to write the number indicating the group to which they were assigned on the second scrap of
paper (1, 2, or 3). Remind students that the second scrap of paper should have 3 items of information on
Activity 3. Analysis and evaluation of the article reprint.
You can ask students to read the reprinted article about a particular after-school parenting program at the end of
Chapter 12 and evaluate it using the criteria in Appendix 3 and those specified on page 317. In class, you can call
on students to state what they consider strengths and weaknesses of the study. Also, you can ask them to form small
MULTIPLE-CHOICE TEST ITEMS FOR CHAPTER 12
1. The feature of experimental research that most differentiates it from other types of quantitative research is its
a. incorporation of multiple dependent variables into the research design.
b. incorporation of multiple independent variables into the research design.
c. manipulation of the independent variable.
d. random selection of research participants from a defined population.
2. The main difference between manipulation of an independent variable and observation of naturally
occurring variations in an independent variable is that
a. naturally occurring variations provide more persuasive evidence of a causal relationship
involving the independent variable.
b. manipulation provides more persuasive evidence of a causal relationship involving the
independent variable.
c. manipulation is typically done in artificial laboratory settings, whereas observation of naturally occurring
variations can be done in schools.
d. manipulation requires the involvement of fewer research participants than observation of naturally
occurring variations.
3. In experiments, random assignment means that research participants are
a. selected at random from a defined population.
b. assigned at random to receive different posttests.
c. free to select the experimental or control condition in which they will be involved.
d. assigned at random to different experimental and control conditions.
4. Random assignment of research participants to experimental and control groups guarantees that
a. the experimental and control groups are equivalent in all respects.
b. there is no systematic bias in the assignment of the participants to the two groups.
c. the experimental and control groups will have approximately the same mean scores on the posttests.
d. the experimental and control groups will have approximately the same mean scores on the pretests.
5. Pretests in an experiment
a. measure variables that existed prior to the beginning of the experiment.
b. can be used to assess gains or losses among the research participants as a result of the experimental
intervention.
c. can be used to make statistical adjustments that equate the experimental and control groups on the
variables measured by the pretests.
d. All of the above.
6. The primary characteristic of quasi-experiments is that
a. research participants are not randomly assigned to the experimental and control groups.
b. they do not include pretests.
c. they include a pretest only for the experimental group.
d. analysis of covariance cannot be used to analyze the research data.
7. A factorial experiment
a. yields data that are analyzed by the statistical technique of factor analysis.
b. includes only measures that constitute ratio scales.
c. has more than one independent variable.
d. cannot be used to study interaction effects involving two independent variables.
8. In experiments, the term extraneous variable refers to
a. the intervention assigned to the control group.
b. a factor other than the treatment variable that might affect the outcome variables.
c. a factor that underlies two or more of the independent variables.
d. a variable that increases the experiment’s internal validity.
9. An experiment has high internal validity if
a. an observed difference between an experimental and control group is solely attributable to the
treatment variable.
b. there is a strong effect of the treatment variable on the outcome variable.
c. the research participants have been randomly assigned to the experimental and control conditions.
d. All of the above.
10. Match the following examples with the type of threat to internal validity that each appears to reflect.
___ 1. When the investigator began her three-week program designed to reduce racial prejudice, she
was unaware that “Roots” was to be rerun on television during the third week.
___ 2. Due to overcrowding, six students from Mrs. Brown’s class were assigned to another home
room. This happened during the fourth week of a five-week experimental program being
conducted in Mrs. Brown’s home room class.
___ 3. Mrs. Funk’s entire third grade class scored well below the national norm on the mathematics
portion of an achievement test. Following a month–long intervention, the class was retested,
and virtually everyone’s mathematics score improved.
___ 4. A posttest was administered six months after a training session aimed at increasing students’
manual dexterity, and both the experimental and control groups showed significant gains over
their pretest scores.
___ 5. A civics teacher plans an experiment to evaluate a simulation game designed to heighten students’
awareness of national political issues. Prior to introducing the game in class, she administers a
measure of students’ knowledge of these issues to all students involved in the experiment.
a. Statistical regression
b. History
c. Testing
d. Maturation
e. Experimental mortality
11. Statistical regression in experiments is the tendency for
a. research participants who are near the 50th percentile on the pretest to show smaller gains on the
posttest than research participants between the 1st and 20th percentiles.
b. research participants who are at the 95th or higher percentile on the pretest to show smaller gains on
the posttest than research participants who are near the 50th percentile on the pretest.
c. the posttest to have lower validity for research participants in the control group than for research participants
in the experimental group.
d. analysis of covariance to over-estimate the pretest-posttest gains for the experimental and control
groups.
12. A selection-maturation interaction occurs in an experiment if the research participants
a. are not randomly selected and the experiment lasts more than one year.
b. are not randomly assigned to the experimental and control groups.
c. in the experimental and control groups are at different developmental levels.
d. are at a higher developmental level than the population for which the experimental intervention was
designed.
13. An experiment has high treatment fidelity if
a. the intervention is sufficiently strong to create an effect on the outcome variable.
b. the intervention was developed by individuals with substantial expertise. .
c. each of the research participants receives exactly the same intervention.
d. the intervention is implemented according to the specifications of the researchers.
14. An experiment has population validity if
a. the sample of research participants is similar to the population for which the intervention is
intended.
b. educators can generalize the results of the experiment to their local community.
c. the experimental and control groups have similar characteristics.
d. research participants have been randomly assigned to the experimental and control groups.
15. An experiment has ecological validity if
a. it was conducted in a school system that is considered high-performing according to state or
national standards.
b. the individuals who delivered the experimental intervention were well-trained.
c. the conditions that were present during the experiment are similar to the conditions
in the setting to which the research results will be applied.
d. All of the above.
16. A baseline condition in a single-case experiment is similar to the ______ in a group experiment.
a. posttest
b. pretest
c. control condition
d. experimental condition
17. Experiments with an A-B-A-B design generally have ____________ than experiments with an A-B-A design.
a. less internal validity
b. greater internal validity
c. greater experimental mortality
d. less treatment fidelity
18. Multiple-baseline single-case experiments
a. use the same type of baseline as an A-B-A experiment.
b. use the same type of baseline as an A-B-A-B experiment.
c. use a naturally occurring condition as the baseline.
d. use situations other than the naturally occurring condition as the baseline.
19. Treatment effects in single-case experiments generally are determined by
a. a pretest and posttest.
b. one pretest and multiple posttests.
c. one measure that is administered repeatedly.
d. a different measure for the baseline and treatment conditions.
SHORT-ANSWER TEST ITEMS FOR CHAPTER 12
1. Name and explain three threats to internal validity that should be considered in designing an experiment.
2. Suppose that you have located a research report that provides evidence about effective teaching of ninth–grade
algebra, and you wish to apply the research results to your local algebra classes. List two factors that you must
consider to determine the degree of external validity of the research results.
3. Name two procedures used in single-case experiments to achieve control over, and precise description of, the
experimental situation.
4. A researcher administers a spelling test to all the students in 20 college freshman English classes, and identifies
175 students who score less than 50 percent on the test. She asks these students to volunteer for a remedial
spelling program from a commercial publisher. A total of 120 students volunteer, and she randomly assigns 60
of them to the experimental group and the remaining 60 to the control group. She teaches the spelling program
to the experimental group one hour a week for 10 weeks. After 10 weeks, 28 students have completed the
program. She administers another spelling test to the 28 students remaining in the experimental group and to
the 58 students remaining in the control group. The difference between the mean posttest score of the
experimental group (M = 80) and the mean posttest score of the control group (M = 66) is found to be
significant at the .01 level.
a. Indicate the type of experimental design (true experiment, randomized field trial, quasi-experiment,
or single-case experiment) and the specific research design that was used.
b. Indicate the major threat(s) to internal validity associated with that design.
c. Explain how each threat is likely to have affected the results.
CHAPTER 12
EXPERIMENTAL RESEARCH
Multiple-Choice Test Items
Short-Answer Test Items
