Research Design Part II

RESEARCH DESIGN
PART 2
Experimental Research Design
Purpose
The aim of the experimental research is to investigate the possible cause-andeffect relationship by manipulating one independent variable to influence the other
variable(s) in the experimental group, and by controlling the other relevant
variables, and measuring the effects of the manipulation by some statistical means.
By manipulating the independent variable, the researcher can see if the treatment
makes a difference on the subjects.
If the average scores of two groups prove to be significantly different, and if
there are not any explanations for this difference, then it can be concluded that the
effect of the treatment caused this difference. This is where experimental research
differs from correlational research, For instance, correlational studies only describe
or predict the strong relationship between socioeconomic level and the academic
achievement but cannot prove the direct cause-and-effect relationship between these
two variables. It is the experimental research which can demonstrate that by
changing the independent variable, a change is possible on the dependent variable.
In educational research the most frequently studied dependent variables are
achievement, motivation, attention, interest in learning, participation and attitudes.
The common independent variables that are manipulated are teaching methods,
types of assignments, types of teaching materials such as text books and visual aids,
types of rewards, types of questions used by the teacher, and evaluation techniques.
There are however, some independent variables such as age and gender that cannot
be manipulated. When the independent variable that is chosen cannot be
manipulated, either a comparative research is conducted, or a second independent
variable is chosen for manipulation in order to conduct an experimental study.
The following are some examples to illustrate the experimental research
(Isaac & Michael, 1977):

To investigate the effects of two methods of teaching a twelfth-grade

history program as a function of class size (large and small) and levels
of student intelligence (high, average, low), using random assignment
of teachers and students-by-intelligence-level to method and class size.

To investigate the effects of a new drug abuse prevention program on

the attitudes of junior high school students using experimental and
control groups who are either exposed or not exposed to the program
respectively, and using a pretest-posttest design in which only half of
the students randomly receive the pretest to determine how much of an
attitude change can be attributed to pretesting or the educational
program.
To investigate the effects of two methods of pupil evaluation on the
performance of children in the twenty-three elementary schools of a given

Research Manual for Social Sciences

suburban district. 'N' in this study would be the number of classrooms,

rather than children, and the method would be assigned by stratified
random techniques such that there would be a balanced distribution of the
two methods to classrooms across grade levels and socio-economic
locations of schools. (p. 24)
Types
Different types of experimental research can be conducted depending on
the nature of subjects and the instruments, and the way data are collected and
analyzed. Answers to the following questions would determine what type of
experimental design to follow:
-

Will there be a control group?

How many subjects will there be?
Will the subjects be randomly selected?
Will each group be pretested?
How will the obtained data be analyzed?
What factors may affect the internal validity?
What factors may affect the external validity

Experimental investigations can be conducted on groups or individuals.

Accordingly, the structure of the design changes as group experimental design, or
single-subject experimental design.
Group Experimental Design
Group experimental designs can be of different forms If there is only one
independent variable that can be manipulated, then a single-variable design is used.
If there are two or more independent variables, and at least one can be manipulated,
then a factorial design should be chosen.
Single-variable designs. These studies are classified under three main
headings depending on the degree of control maintained on other variables:
1.
2.
3.
1.

Pre-experimental designs (low degree of control)

True experimental designs (high degree of control)
Quasi-experimental designs (medium degree of
control)
Pre-experimental designs are classified depending on whether there is an
involvement of one or two groups, and whether the groups are posttested only,
or both are pretested and posttested:
-

One-shot case studies: One group is exposed to the treatment, and only a
posttest is given to observe or measure the effect of the treatment on the
dependent variable within the experimental group. Since it is applied on
a single group, there is no control group involved in this design. As seen
in Table 6.2, first of all, the chosen group is exposed to the treatment ,

Research Design Part II

and then it is tested only once for the purpose of measuring the degree of
change on the dependent variable after the treatment.
-

One-group pretest-posttest design: One group is pretested and exposed

to the treatment, and then posttested. This is called a one-group pretestposttest design because the two tests are administered to the same group.
As seen in Table 6.3, the first one is administered at the beginning of the
treatment and the second one at the end. This design is better than the
one-shot case study design but it still has some weaknesses because no
proper selection is made taking the other possible dependent variables
into consideration.

Table 6.2
Example of a One-shot Case Study Design

STEPS

PROCEDURE

AIM

(on single group)

Step 1

TREATMENT
Ten weeks of
instruction on reading
with a new technique

Step 2

POSTTEST

To influence the dependent

variable
(reading comprehension skill)
To measure the degree of change
on the dependent variable

a reading test or a
questionnaire

Table 6.3
Example of a One-group Pretest-posttest Design

STEPS PROCEDURE

AIM

(on single group)

Step 1

PRETEST

Step 2

(Reading test)
TREATMENT

To measure the degree of the

dependent variable before the
treatment
To influence the dependent
variable

(Ten weeks of instruction

on reading with a new
technique)

Step 3

POSTTEST
(Reading test)

To measure the degree of change

on the dependent variable

Research Manual for Social Sciences

Static-group comparison design: At least two groups are involved. After

one group receives the treatment, all groups are posttested. For example,
two groups that have already been formed are chosen, one is given a
treatment different from the regularly applied one. As seen in Table 6.4,
an experiment is conducted on Group 2; therefore, Group 2 is given the
treatment before being posttested. Although Group 2 receives no
treatment, the posttest is administered to these subjects as well in order
to see the relative effect of the treatment. This design has better control
over most of the variables.
Table 6.4
Example of a Static-group Comparison

STEPS
Step 1

PROCEDURE
GROUP 1
No new
treatment

GROUP 2
Treatment

(Regular
Instruction )

(A new technique)

POSTTEST

POSTTEST

Step 3
(Reading
test)

(Reading test)

AIM
To influence the
dependent variable

To measure the degree

of change on the
dependent variable

2. True experimental designs have the highest level of control among the three
single-variable experimental designs because the subjects within the groups are
randomly assigned for each group. When subjects are randomly assigned,
there is higher control of the internal validity as well as the external validity.
Moreover, there is always a control group to compare the results of the subjects
in the experiment with other subjects of similar status that have not been
exposed to the treatment.
True experimental research may be designed with or without a pretest on at
least two groups of randomly assigned subjects. The classification of true
experimental designs are made accordingly (see Tables 6.5 - 6.7):
1. The posttest-only control group design
2 .The pretest-posttest control group design
3. Solomon four-group design

Table 6.5
The Posttest-only Control Group Design

STEPS

P R O CE D U R E

AIM

Research Design Part II

Step 1

Random
assignment
for Control
Group

Step 2

No treatment

Random
assignment for
Experimental
Group

Treatment
Step 3

POSTTEST

POSTTEST

To control subject
characteristics threat to
internal validity
To observe any possible
change on dependent
variable
To measure the degree of
change

Table 6.6
Pretest-posttest Control Group Design

STEPS

P R O CE D U R E

Step 1

Random
assignment
for Control
Group

Step 2

PRETEST

Step 3

No treatment

Step 4

POSTTEST

AIM

Random
assignment for
Experimental
Group

PRETEST

Treatment
POSTTEST

To control subject
characteristics threat to
internal validity
To measure the degree of
the dependent variable
before the treatment
To influence the
dependent variable
To measure the degree of
change

Table 6.7
Solomon Four-group Design
Step

G1 Exp.

G2 Control

Random
assignment

Random
assignment

2
3
4

PRETEST
PRETEST
Treatment
-----POSTTEST POSTTEST

G3 Exp.
Random
assignment

Treatment
POSTTEST

G4 Control
Random
assignment

--------POSTTEST

Solomon four-group design takes the effect of

pretest and
posttest sensitivization into consideration.
It is the combination of pretest-posttest control group (G1 and G2) and
posttest only control group (G3 and G4) designs. In other words,
- subjects are randomly selected and placed into four groups;
- the first and the second groups are pretested;
- the first and the third groups are exposed to the treatment, and the
second and the fourth groups are taken as control groups;
- all four groups are posttested.

Research Manual for Social Sciences

This design provides the best result but it requires a large sample so that
enough subjects could be assigned to four groups. When the sample is
large, administering the tests becomes difficult, time and energy
consuming.
3. Quasi-experimental designs may be conducted in three different forms:
- Non-equivalent control groups design
- Time-series design
- Counterbalanced design
All three categories of quasi-experimental design are similar to some
extent to the true experimental design. However, they differ in the
following aspects:
-

Non-equivalent control group design is similar to pretest-posttest control

group design except that no randomization is required. The groups are
generally chosen from clustered units such as classrooms or counseling
groups. The major drawback of this design is that since there is no
randomization, variables related to history, maturation or testing may
interfere with the effect of the treatment.

Time-series design is an elaborated version of the one-group pretestposttest pre-experimental design in the sense that subjects are repeatedly
pretested and posttested before and after the treatment rather than being
tested once at the beginning and a second time at the end of the treatment.

Counterbalanced designs are used to equate the experimental and control

groups. In this design, the treatments are given to all the groups in a
different order, and the number of groups should be equal to the number of
treatments (see Table 6.8). In other words, if there are two different
treatments, there should be two groups.
Table 6.8
An Example of a Counterbalanced Design
Week

1-4

5-8

Session One
Group 1 Group 2

Session Two
Group 1 Group 2

Technique X

Technique Y

Technique X

TEST

TEST

TEST

TEST

Mean = 70
Technique Y

Mean = 40
Technique X

Mean = 75
Technique Y

TEST
Mean = 50

Mean = 55
Technique X

Technique Y

TEST

TEST

TEST

Mean = 75

Mean = 50

Mean = 85

Factorial designs. These are applied to determine if the effects of the

independent variable can be generalized. A factorial design, in this sense, indicates

Research Design Part II

relationships between variables. It may be that while one variable increases, the
other one decreases, or an increase in one gives rise to the other variable.
These designs are the modified forms of either the posttest-only control
group or pretest-posttest control group designs with or without random assignment.
They also allow the investigation of additional independent variables. Moreover,
they enable the researcher to study the interaction of the independent variable with
one or more other variables, which are called moderator variables. A moderator
variable could be a treatment variable or a subject characteristic variable. For
instance, aside from finding the effect of different methods of teaching on students'
success level, the researcher might want to look into the effect of the class size in
the application of these two methods. Under these conditions, the researcher has to
design a 2 by 2 factor design (see Table 6.9.)

Table 6.9
An Example of a Factorial Design
Step

G1 Exp. G2 Control

2
3

PRETEST

PRETEST

PRETEST

PRETEST

Treatment 1
Class size 1

Treatment 2
Class size 2

Treatment 1
Class size 2

Treatment 2
Class size 1

POSTTEST

POSTTEST

Random
assignment

G4 Control

Random
assignment

Random
assignment

G3 Exp.

POSTTEST

Random
assignment

POSTTEST

Single-subject Experimental Design

These designs are applied when only one subject is involved in the study.
They aim at studying the behavioral changes of different individuals as a result of
some intervention or treatment. Gay, (1987, p. 323) mentions three types of singlesubject experimental design:
1. A-B-A withdrawal design
2. Multiple-baseline design
3. Alternating treatments design
A-B-A withdrawal design. These designs are basically of three types: (a)
the A-B design, (b) the A-B-A design, and (c) the A-B-A-B design. In the A-B
design, similar to counterbalanced design, the subject is exposed to a non-treatment
condition (A) as well as to a treatment (B) in order to study the difference of
behavior within the subject. Clinical studies frequently make use of this technique.
In order to find out if the improvement is really due to the treatment or due to some
other factors, the A-B-A design is applied. In making final judgements on the effect
of the treatment, this time the A-B-A-B design is applied.

Research Manual for Social Sciences

Multiple-baseline designs. These are used when the A-B design is used,
and the conditions cannot be reversed. In this design, the treatment can be
withdrawn but the effects of the treatment continue, and it is impossible to go back
to the nontreatment condition.
Alternating treatments. These are used to determine the relative effects of
two or more treatments within a single-subject. This design is preferred because no
withdrawal is necessary. The treatments are alternated on a random basis with no
interval in between. However, it has one major drawback. The effect of one
treatment may persist while the second treatment is being given.
In the A-B-A withdrawal design, the analysis and interpretations are made
by studying the graphic representations of the results or simply by visual inspection.
The primary criterion of making generalizations is the "clinical significance of the
results rather than the statistical significance" (Gay, 1987, p. 325). The statistical
results obtained from tests (e.g. t-test) may support the clinical generalizations.
In arriving at a reliable decision, the results need to be replicated. Direct
replication is done by the same investigators using the same or different subjects in
a specific setting. Simultaneous replication is done at the same time and at the same
location on a number of individuals that have the same problem. When the
investigators, behaviors, or settings are differed, then it is called "systematic
replication." "Clinical replication involves the problem of a treatment package,
composed of two or more interventions which have been found to be effective
individually, designed for persons with complex behavior disorders" (Gay, 1987, p.
326).
During the manipulation of the independent variable, it is difficult to keep
the other variables under control. This control can be maintained by eliminating or
removing their influence that would affect the performance. For instance, in
investigating the effect of one teaching technique on students, two groups of
students are chosen, and the experimental group is exposed to the treatment but the
control group is not. Meanwhile, these students continue with the instruction. If only
one teacher teaches both the experimental group and the control group, the teacher
factor as a possible threat to the internal validity of the study would be eliminated.
Otherwise, any positive or negative effect might be attributed to the teacher as well
as the teaching technique itself.
When the difference of behavior (dependent variable) observed in the
experimental group is directly the result of the application of the technique
(manipulation of the independent variable), and not of any other variable, this
condition is referred to as internal validity. It is difficult to maintain internal
validity, because, during the period of treatment, similar factors mentioned below
may affect the performance of the dependent variable (see also Table 6.10):
The occurrence of any event which is not part of the treatment (history threat)
The mental and physical changes that occur within subjects (maturation threat)
Subjects getting familiarized with test items after having taken the pretest
(testing threat)
Assessment of performance by unreliable measures (instrumentation threat)
Selection of subjects on the basis of their extreme scores (This leads the
subjects with the highest scores on a pretest to score lower on a posttest or
vice-versa.) (regression threat)

Research Design Part II

Not choosing groups of the same level (The initial difference between the
groups may still be reflected on the posttest results.) (subject characteristics
threat)
The loss of subjects for some reason during the study (mortality threat)
The unintended outcomes of the location the treatment of the study takes
place (location threat)
The treatment of the experimental group in an unplanned way
(implementation threat)
Collecting data without giving sufficient orientation to the data collectors (data
collector characteristics)
Table 6.10
Threats to Internal Validity and the Degree of their Control

Threats 1
S
M
L
I NA
D
T NA
H
Mat.
A
R
Im.
-

Different Types of Experimental Designs

2
3
4
5
6
7
8
**
**
*
**
**
*
*
*
*
**
**
*
*
*
*
*
*
*
*
*
**
*
*
*
*
*
*
*
**
**
*
**
**
*
**
*
** **
**
-

9
**
*
*
**
*
**
-

10
**
**
**
*
*
**
**
-

11
**
**
*
*
**
**
*
*
-

12
**
**
*
*
**
**
*
*
-

Key:
Degree of control over the threats.:
(**)
= Strong control, threat is not likely to occur
(*)
= Some control, slight possibility of threat
(-)
= Weak control, high possibility of threat
(N/A)
= Not applicable
Threats : S = Subject, M = Mortality, L = Location,
I = Instrument Decay, D = Data collector characteristics,
T = Testing, H = History, Mat. = Maturation, A = Attitudinal,
R = Regression, Im. = Implementation
Designs: 1 = One-shot case study, 2 = One-group pretest-posttest, 3 = Static group
comparison, 4 = Posttest-only control group, 5 = Pretest-posttest control group,
6 = Matching only pretest-posttest control group,
7 = Solomon four-group, 8 = Counterbalanced, 9 = Time-series, 10 = Factorial
design, 11 = A-B-A-B, 12 = Multiple baseline

10

Research Manual for Social Sciences

Some suggestions can be offered as how to avoid some of these threats.

In order to control the possible threats, the researcher tries to
control the research conditions strictly;
collect more information on subjects;
check the details more carefully;
find a suitable design (see Table 6.11)
Table 6.11
Suggested Techniques to Avoid Threats to Internal Validity
Suggested
Threats
S
M
L
I
T
H
Mat.
R
Im.

Control the
conditions
strictly

Techniques

Collect
more
information

Check the details Find a suitable

carefully
design

*
*
*
*

*
*
*

*
*
*
*
*
*
*
*

.
The extraneous variables in a group experimental design can be controlled
if
-

the subjects are selected randomly;

the pairs are matched from experimental and control groups;
homogeneous groups or subgroups are compared;
analysis of covariance is applied when necessary;
the same group is exposed to two different treatments.

In experimental research, if the obtained results can be generalized or

applied to other groups or environments, the treatment is said to have external
validity. Aside from the above mentioned factors, there are other issues that affect
the external validity:
-

Exposure of the group to the treatment prior to the intended experiment

(thus it is difficult to observe the effect of the later treatment.)
The characteristics, personality or bias of the experimenter (One
experimenter cannot conduct the same experiment in all settings.)
Hawthorne effect (Students may be negatively or positively involved,
thinking that they are involved in an experiment or getting special
attention.)
Novelty effect (Students are interested and motivated due to getting
involved in a different activity, which indirectly affects their performance
positively.)

Research Design Part II

11

Characteristics of the subjects which are termed as organismic variable

(For instance, subjects may perform differently depending on their sex. )
Intervening variables such as boredom and anxiety

Basic Steps
Isaac and Michael (1977) report the seven steps in experimental research as
outlined by Van Dalen and Meyer (1966) :
1.
2.
3.
4.

5.
6.
7.

Survey the literature relating to the problem.

Identify and define the problem.
Formulate a problem hypothesis, deducing the consequences, and
defining basic terms and variables.
Construct an experimental plan.
a. Identify all non-experimental variables that might contaminate the
experiment, and determine how to control them.
b. Select a research design.
c. Select a sample of subjects to represent a given population, assign
subjects to groups, and assign experimental treatments to groups.
d. Select or construct and validate instruments to measure the
outcome of the experiment.
e. Outline procedures for collecting the data, and possibly conduct a
pilot or "trial run" test to perfect the instruments or design.
f. State the statistical or null hypothesis.
Conduct the experiments.
Reduce the raw data in a manner that will produce the best appraisal of
the effect which is presumed to exist.
Apply an appropriate test of significance to determine the confidence
one can place in the results of the study. (p. 25)

Whatever research method is chosen, it is generally adviced by scholars

(Allen 1987; Gay 1987; Isaac & Michael 1977) to evaluate the work undertaken at
every stage (introduction, review of literature, data analysis etc.) of the research in
order to avoid or detect any shortcomings or faulty applications that may appear in
the study. Researchers are adviced to ask themselves questions, such as the ones
presented below in order to be able to conduct the study according to the required
norms (see also Appendices 3 and 4).
TITLE
1. Have I precisely identified the problem area with sufficient clarity?
2. Have I included the specifications of the independent and the dependent
variables?
3. Have I arranged the words in the title effectively?
4. Have I tried to avoid using irrelevant expressions, unqualified nouns, quotations
and underlinings?
5. Have I identified the target population?
6. If I am writing a thesis or dissertation, have I typed the required information on
the title page?

12

Research Manual for Social Sciences

INTRODUCTION
1. Have I defined the problem accurately, clearly focussing on the basic difficulty?
2. Have I presented the background of the study by giving a careful analysis of
what is known about the problem, and what the suspected facts are? In other
words, have I demonstrated the relevance of the existing knowledge to the
problem?
3. Have I stated the educational significance of the problem in either basic or
applied research by stating clearly the major points to be resolved and/or the
objectives to be investigated?
4. Have I mentioned the variables or factors to be studied or investigated; and
have I indicated the relationship between these variables or factors?
5. Have I directly or operationally defined the technical terms that may sound
unfamiliar to my readers?
6. Have I stated the problem, the question, and the approach to be adopted in a
logical and internally consistent manner?
7. Have I defined the scope of the study by mentioning what aspects to include,
and what to exclude, regarding the topic and the problem area? Have I given
justification for the inclusions and the exclusions I have made ?
REVIEW OF LITERATURE
1. Are the investigations and studies cited in this section adequate and relevant to
the problem under investigation?
2. Have I compiled a comprehensive review of literature?
3. Have I critically analyzed the studies I have cited, and have I compared and
contrasted these results?
4. Have I avoided reviewing these studies only in the form of abstracts and
annotations?
5. Have I developed an appropriate rational or theoretical framework from these
cited research studies by concluding the chapter with a brief summary on the
implications of these studies for the problem I have been investigating?
HYPOTHESES
1. Have I clearly specified the questions that I am seeking answers for? In other
words, have I clearly stated my assumptions and the specific hypotheses that I
aim at testing?
2. If I have deduced any consequences of the research or have made predictions
regarding the outcome of the research, are
they consistent with my
hypotheses?
3. Am I sure that the hypotheses I have specified are testable?
PROCEDURES
1. Have I given the size of the sample and the major characteristics of my
subjects?
2. Have I indicated whether I have studied the entire population?
3. Have I clearly explained my method of selecting the sample?
4. Have I clearly explained my method of sample selection (e.g. random
assignment, matching, voluntary, etc.)?

Research Design Part II

5.
6.
7.
8.
9.
10.
11.

12.
13.

13

Am I sure that I have not been biased in the selection of my subjects?

Is the number of my subjects sufficient to apply the research method I have
adopted?
Do my subjects really represent the whole population?
Have I given sufficient information about the reliability, validity, and
standardization properties of the instruments I have used for testing purposes?
Have I described the procedure I have applied in a logical sequence?
Have I clearly demonstrated the relationship between the method adopted and
the research questions formulated?
Have I appropriately discussed the way I will conduct the method and the way I
will keep the procedure under control so that another researcher who wants to
replicate my study can make use of it properly?
Have I specified how the null (statistical) hypotheses relate to my research
hypotheses?
Were there any uncontrollable instances? If so, have I clearly stated them; and
have I indicated my assumption for the appearance of such uncontrollable
factors?

These questions serve as confirmation prompts to guide researchers while

conducting their research. Questions not relevant to the content of the chapters are
excluded in this volume. For questions relevant to data analysis and the other steps
to follow, reference can be made to Volume 2 of this book.

EXERCISES
A.

Match each aim indicated in Column A with a suitable research design cited in
Column B. Then write the corresponding number in the blank provided.

A
a. __ The aim of the study is to identify associations
between/among variables
b. __ The aim of the study is to describe the
characteristics of a population
c. __ The aim of the study is to seek out associations
between two quantitative variables for each
subject
d. __ The aim of the study is to test the hypothesis
that there is a positive relation between the
independent variables.
B.

B
1. Experimental
2. Historical
3. Descriptive
4. Causal-comp.
5. Correlational

Which research method discussed in this chapter would you use for each of the
following research topics?

14

Research Manual for Social Sciences

1.
2.
3.
4.
5.
6.

The effects of the Communicative Approach on the development of

communicative competence of second graders.
____________________
The feelings of high school teachers about the eight-year compulsory
education in Turkey. ____________
The factors that cause delinquent behavior among university students.
_____________________
The acquisition of hierarchical terms in Turkish.
___________________
Females have a greater amount of linguistic ability than males.
___________________
Predicting students' outcomes from their perceptions of the psychosocial
environment of their classrooms. ___________

C.

As a researcher, you want to compare the effectiveness of two different

techniques of grammar teaching (deductive and inductive) in enhancing the
productive skills.
1. Formulate your hypothesis and indicate the type of hypothesis chosen.
2. Decide on the variables you will focus on.
3. Choose your research design and give reasons for your choice.
4. Indicate the possible threats of the study and think of a way of
eliminating the possible threat(s).

D.

Read the information related to three different studies. Then decide whether
there are any threat(s) to the internal validity of each study. Explain how you
would avoid these threats.
Study 1. In this study, a new vocabulary teaching technique is investigated.
The subjects are randomly assigned to an experimental group and
a control group. The teacher of the experimental group is an
accomplished instructor with ten years of experience. The other
instructor, however, has just graduated .
Study 2. A researcher decides to interview students studying counselling and
special education with the aim of comparing their attitudes toward
their instructors regarding their gender. The first group is
interviewed in the researcher's office and the other in a coffee
shop.
Study 3. A teacher wants to compare the effect of two different textbooks and
thus uses these books in the English classes at two high schools
for a semester. At the end of the semester, the teacher analyzes
the test results of the two groups. In Group 1, 30 percent; in
Group 2, 15 percent of the students have been absent on the day
of the test.

E.

Which threats discussed in this chapter do you think are the most important for
a researcher to consider? Why? Which do you think would be the most
difficult to control? Explain.

15

Research Design Part II

F.

Match the types of experimental design (a-f) with the corresponding diagrams
(1-5).
X = Treatment, O = Test/Observation,
R = Randomization, G = Group
a. ___
b. ___
c. ___
d. ___
e. ___
f. ___

Time-series design
One-shot case
Posttest-only control group design
Pretest-posttest control group design
Solomon four-group design
Static-group comparison

1.

RG1 X
O
_________________
RG2
O

2.

O1 0 2 O3 O 4 X O 5 O6 O7 O8

3.

4.

RG1
O
X
O
_________________
RG2
O
O

5.

RG1
RG2
RG3
RG4

O
O

X
X

O
O
O
O

G.

Study the information given in Tables 6.2-6.9 and then draw a chart indicating
the characteristics of each experimental design.

H.

Following are some problems teachers are faced with during their instruction.
If you were the teacher, what type of questions would you ask yourself to
explore each of the problems?
Problem:
Questions:

Students do not learn by error

correction.
_______________________________
_______________________________

Problem:

Students do not participate in class.

16

Research Manual for Social Sciences

Questions:

_______________________________
_______________________________

Problem:
items.
Questions:

Students do not know very many

vocabulary

_______________________________
_______________________________

K. Below is a definition of a sample as reported by Zimmerman and Kinsler 1979)

(in Moore, 1983). After reading the definition try to answer the following
questions.
Subjects
Fifty-four boys and 54 girls were selected in equal numbers from
kindergarten and first-grade classes at a New York City public
school located in a lower-class area of Harlem. The first readers
ranged in age from 6 years 0 months to 7 years 6 months, and
were comparable in age to the 6-year-olds in the Walters and
Parke (1964) study. The kindergarteners ranged in age from 5
years 1 month to 6 years 0 month and were comparable in age to
the 5 year-olds in the Walters et al. (1983) study. The children
were of either black or Hispanic origin. Within each age group,
the children were randomly assigned to treatment condition (p.
129) .
1.
2.
3.
4.
L.

What kind of selection has been made?

What is the selection based on?
How were the subjects assigned to the treatment? In other words,
how were the subjects divided into control and experimental groups?
What criterion has been used in selecting the two samples?

As a term project think of a topic for research and try to elaborate on it. A
worksheet is provided for you to facilitate your work.

A WORK SHEET
Investigator______________Reviewer_____________Date_____
General area of interest _________________________________
________________________________________________________
________________________________________________________
Unanswered questions related to sub-area of interest
1. ._____________________________________________________
________________________________________________________
2. ______________________________________________________
________________________________________________________
3. ______________________________________________________
________________________________________________________

Research Design Part II

17

Preliminary statement of the problem _____________________

________________________________________________________
________________________________________________________
Scope of investigation ___________________________________
________________________________________________________
________________________________________________________
Specific objectives (based on unanswered questions)
1. ______________________________________________________
________________________________________________________
2. ______________________________________________________
________________________________________________________
3. ______________________________________________________
________________________________________________________
Hypotheses or tentative solutions (match with forgoing objectives)
1. ______________________________________________________
________________________________________________________
2. ______________________________________________________
________________________________________________________
3. ______________________________________________________
________________________________________________________
Study procedures
1. ______________________________________________________
________________________________________________________
2. ______________________________________________________
________________________________________________________
3. ______________________________________________________
________________________________________________________
Kinds of data required
1. ______________________________________________________
________________________________________________________
2. ______________________________________________________
________________________________________________________
3. ______________________________________________________
________________________________________________________