Educational Research: Experimental Research Gay, Mills, and Airasian

You might also like

Download as ppt, pdf, or txt
Download as ppt, pdf, or txt
You are on page 1of 33

Educational Research

Chapter 9
Experimental Research

Gay, Mills, and Airasian


Topics Discussed in this Chapter
 Defining characteristics of experimental
research
 Conducting experimental research
 Manipulation and control
 Threats to validity
 Internal validity
 External validity
 Group designs
Defining Characteristics
 Research designed to investigate cause and effect
relationships through the direct manipulation of an
independent variable and control of extraneous
variables
 Independent variable – the variable being manipulated
 Dependent variable – the variable in which the effect of the
manipulation of the independent variable is observed
 Researcher manipulation and control – choice of treatments,
choice of a research design, use of specific procedures, etc.

Objectives 1.1 & 1.2


Experimental Process
 Six steps to conducting experimental research
1. Selection and definition of the problem
 Statement of a hypothesis indicating a causal
relationship between variables
2. Selection of participants and instruments
 Random selection of a sample of subjects from a larger
population
 Random assignment of members of the sample to each
group
 Selection of valid and reliable instruments
Objectives 2.1 & 2.2
Experimental Process
 Six steps to conducting experimental research
(cont.)
3. Selection of a research plan
 Three types of comparisons
 Comparison of two different approaches
 Comparison of new and existing approaches
 Comparison of different amounts of a single approach
4. Execution of the research plan
 Two concerns
 Sufficient exposure to the treatment
 Substantively different treatments

Objectives 2.1, 2.3, & 2.6


Experimental Process
 Six steps to conducting experimental
research (cont.)
5. Analysis of data
6. Formulation of conclusions

Objective 2.1
Manipulation and Control
 Manipulation
 The researcher’s decisions related to what
constitutes the independent variable
 Active and assigned variables
 Active variables are those the researcher actively
manipulates
 Choice of an instructional strategy
 A particular counseling approach
 Assigned variables are those that cannot be manipulated
by the researcher but are of interest
 Gender
 Race
Objectives 3.1 & 3.2
Manipulation and Control
 Control
 The researcher’s efforts to remove the influence of any
extraneous variables that might have an effect on the
dependent variable
 The goal is to be assured the only differences between
groups is that related to the independent variable
 Participant variables – characteristics of the subjects
 Pre-existing achievement levels
 Differences in attitudes
 Environmental variables – characteristics of the context
 Learning materials
 Differences in the time available for treatment between groups

Objectives 3.3 & 3.4


Experimental Validity
 Internal validity – the degree to which the
results are attributable to the independent
variable and not some other rival explanation
 External validity – the extent to which the
results of a study can be generalized
 Population validity – generalizations related to
other groups of people
 Ecological validity – generalizations related to
other settings, times, contexts, etc.
Objectives 4.1, 4.2, 6.1, & 6.2
Experimental Validity
 Relative importance of internal and external
validity
 Internal and external validity are related
reciprocally
 Controlling internal validity decreases external validity
 Controlling external validity decreases internal validity
 First demonstrate an effect in a highly controlled
environment (i.e., prioritize internal validity)
 Second replicate the study in a more realistic,
natural setting (i.e., prioritize external validity)
Objective 4.3 & 4.4
Threats to Internal Validity
 History
 Maturation
 Testing
 Instrumentation
 Statistical regression
 Differential selection of participants
 Mortality
 Selection-maturation interaction, etc.
Objectives 5.1 – 5.8
Threats to External Validity
 Pre-test treatment interaction
 Multiple treatment interference
 Selection treatment interaction
 Specificity of variables
 Participants
 Operational definition of the treatment
 Operational definition of the dependent variable
 Specific times
 Specific circumstances
 Treatment diffusion

Objectives 6.3 – 6.7


Threats to External Validity

 Experimenter effects
 Reactive arrangements (Participant Effects)
 Artificial environment
 Hawthorne effect (attention)
 John Henry effect (compensatory rivalry)
 Placebo effect
 Novelty effect

Objectives 6.8 & 6.9


Controlling for Extraneous Variables
 Extraneous variables must be controlled
to be able to attribute the effect to the
treatment
 Group equivalency must be assured
 Four major means to achieve control
 Randomization
 Selection – controls for representation
 Assignment – controls for group equivalency
Objectives 2.5, 7.1, & 8.1
Controlling for Extraneous Variables
 Matching
 Identifying pairs of subjects “matched” on specific
characteristics of interest
 Randomly assigning subjects from each pair to different
groups
 Difficulty with subjects for whom no match exists
 Comparing homogeneous groups
 Restricting subjects to those with similar characteristics
 Restricting subjects results in problems related to
generalization

Objectives 8.4 & 8.5


Controlling for Extraneous Variables
 Using subjects as their own controls
 Multiple treatments across time
 Problem with carry-over effects
 Analysis of covariance (ANCOVA)
 Statistically adjusting the posttest scores for
the subjects in each group for pretest
differences that existed at the beginning of the
study
 Creates statistically equivalent groups

Objectives 8.6 & 8.7


Controlling for Extraneous Variables
 Other ways to control extraneous
variables
 Holding variables constant
 Using only males rather than males and
females
 Selecting teachers with only similar levels of
experience
 Selecting only one grade level
 Stipulating the specific length of a treatment
Objective 8.2
Group Designs

 Two major classes of group designs


 Single-variable designs – one independent
variable
 Factorial designs – two or more independent
variables
 Three types of experimental designs
 Pre-experimental designs
 Experimental designs
 Quasi-experimental designs
Objectives 9.1, 9.2, 10.1, & 10.2
DESIGNS
Experiments
If X, then Y

or, in more colloquial terms:

If the program is given, then the


outcome occurs
If X, then Y
and
If not X, then not Y

Or, once again more colloquially:

If the program is given, then the outcome occurs


and
If the program is not given, then the outcome does not
occur
NOTATION
Pre-Experimental Designs
 Three types
 One-shot case study
X O
 One-group pretest-posttest design
O X O
 Static group comparison
X1 O
X2 O
 Threats to internal validity – see Figure 9.1
Objective 9.2
True Experimental Designs
 Three types
 Pretest-posttest control group design
R O X O
R O O
 Posttest only control group design
R X O
R O

Objective 9.4
True Experimental Designs
 Three types (cont.)
 Solomon four-group comparison
R O X O
R O O
R X O
R O
 Threats to internal validity – see Figure
9.2
Objective 9.4
Quasi-Experimental Designs
 Three types
 Non-equivalent control group design
O X O
O O
 Time series design
O O O O X O O O O
 Counterbalanced design
O X1 O X 2 O X 3 O
O X3 O X 1 O X 2 O
O X2 O X 3 O X 1 O
 Threats to internal validity – see Figure 9.2
Objective 9.5
Factorial Designs
 Two independent variables and one
dependent variable
 The effect of teaching strategy and gender on
students’ achievement
 The effect of a particular counseling technique
and the clients’ ethnicity on the success of the
treatment
 The effect of a specific coaching approach and
children in three age groups on the ability to
perform certain physical tasks
Objective10.1
Factorial Designs
 This design increases explained variance and
reduces unexplained variance
 Explained variance is that which can be accounted
for by the independent variable(s)
 By adding an additional variable into the design the
explained variance is likely going to increase
 Unexplained variance is that which cannot be
accounted for by the independent variable(s)
 Because it is unexplained, the researcher assumes this is
due to error
 This is also known as error variance

Objectives 10.1 & 10.2


Factorial Designs
 Interaction
 The degree to which changes in the dependent
variable are different depending on the levels of
each of the independent variables
 A particular instructional strategy is more effective
for males than females
 A particular counseling technique is more effective
when the ethnicity of the counselor and client are
similar

Objective 11.1
Factorial Designs
 Interaction
 Visually explained by a graph of
performance of all levels of both
independent variables
 Parallel lines indicate no interaction
 Non-parallel lines indicate an interaction
 Statistically examined with an F-test of the
statistical significance of the interaction

Objective 11.1
Factorial Designs
 Interaction
 Visual presentation of a significant interaction
14

12

10

TEST
T est Scores

Small Tes ting Room

4 Large Testing Room


Small Lec ture R oom Large Lecture Room

LECTURE

 Statistical analysis of this significant interaction:


F1,16 = 4.88 (p = .0420)
Objective 11.1
Factorial Designs
 Interaction
 Visual presentation of a non-significant interaction
18

16

14

12
T est Scores

10
TEST

Small Tes ting R oom

8 Large Testing R oom


Small Lec ture R oom Large Lec ture Room

LECTURE

 Statistical analysis of this non-significant


interaction: F1,16 = 0.97 (p = .340)
Objective 11.1
GROUP WORK
Group of 3 to 4 students
Answer the following questions

1. Explain the similiraties and differences


between pre-experiment, true
experiment, and quasi-experiment
2. Explain 3 designs of pre-experiments, 3
design of quasi-experiments, and 3
designs of true experiment.

You might also like