Experimental research Method

Assignment on Experimental Method Submitted as a part of Ph.D. course work

Submitted by: PRASANNA DEVARAMATHA ANILKUMAR Research Scholar

Supervisor: Dr. N Parvathamma Professor and chairman Department of P G Studies and Research in Library and Information Science Gulbarga University, Gulbarga.

Department of P G Studies and Research in Library and Information Science Gulbarga University, Gulbarga. 2014

Experimental Research:

• Introduction: • Experimental Research • Areas of experimental research application • Essential Characteristics of Experimental Research • Tools of Control • Experimental Research design • Types of Experimental Research  Pre-Experimental Design  True Experimental Design  Quasi-Experimental Design  Correlation and Ex Post Facto Design • Steps involved in Experimental Research • Advantages of experimental Research 

Disadvantages of experimental Research



References

Experimental Research: Introduction: Research designs are either experimental or non-experimental. Experimental research is conducted mostly in laboratories in the context of basic research. The principle advantage of experimental designs is that it provides the opportunity to identify cause-and-effect relationships. Non-experimental research, e.g., case studies, surveys, correlation studies, is non-manipulative observational research usually conducted in natural settings. While laboratory-controlled experimental studies tend to be higher in internal validity, non-experimental studies tend to be higher in external validity. One major limitation of experimental research is that studies are typically conducted in contrived or artificial laboratory settings. Results may not generalize or extrapolate to external settings. Two exceptions to this rule are natural experiments and field experiments. Natural experiments document and compare the behaviors of subjects before and after some natural event; e.g., floods, tornadoes, hurricanes. Field experiments involve manipulating conditions in the natural setting for the purpose of determining their influence on behavior.

The field

experiment is unique insofar as it tends to be moderately high on both external and internal validity. In experimental research, the investigator manipulates conditions for the purpose of determining their effect on behavior. Subjects should be unaware of their membership in an experimental group so that they don’t act differently (Hawthorne Effect). In the simplest experimental design, investigators administer a placebo to the control group and a treatment to the experimental group. Experimental designs vary in terms of subjects’ assignments to different groups, whether subjects were pre-tested, whether different treatments were administered to different groups, and the number of variables being investigated. Experiments are typically structured in terms of independent, organism, and dependent variables. The independent variable is a manipulated environmental stimulus dimension, the organism-variable is some dimension (e.g., sex, race) of more or less stable characteristics of the organism, and the dependent variable is a behavioral dimension that reflects the influence of the independent and organism-variables. The general objective in experimental research is to define

the relationship between the antecedent (independent and organism) variables and the consequent (dependent) variables. Experimental Research: It is a basically a collection of research designs, guidelines for using them, principles and procedures for determining statistical significance, and criteria for determining the quality of the study. Experimental research is a research in which we use logic and symbols found in natural science. An experiment means, “Modifying something in a situation, then comparing an outcome to what existed without the modification.” Experiments are generally conducted in order to test the strength of relationship between variables. Also saw that when the researcher is testing influence of one variable on another. The variable doing the influencing called independent variable. While the other being influenced called dependent variable. Areas of experimental research application: Experimentation is an important approach in most branches of science. It is one of the most useful and powerful approach in LIS. It can be used for the following purposes.  To identify relationship between variables.  To test hypotheses.  To test theories.  To establish principles.  To test new techniques for acquiring, classifying, storing and retrieving location of documents/ information.  To test new library and information services.  To explore conditions under which a library and information phenomenon occurs.  To satisfy the curiosity of a researcher for a library and information phenomenon.  To test a new LIS education programme

 To test new LIS curriculum  To test new method for organizing classes. Essential Characteristics of Experimental Research : A. Comparison of Groups : 

2 groups (Experimental Groups)



Experimental group receives treatment



Control group does not receive treatment.

B. Manipulation of Independent variable : 

The independent variable may be established as omne form of the variable versus another.



The independent variable may be established as the presence versus obscene of a particular form.



The independent variable may be established as varying degrees of the same form.

C. Randomization : 

Subjects are randomly assigned to groups.



Random assignment. -

Individual participants have the chance of being assigned to any of the two groups.

-

Elemenates extraneous variables.

Independent Variable: The independent, or manipulated variable, is a factor that’s intentionally varied by the experimenter. –

Any variable that can be manipulated, or altered, independently of any other variable

–

Hypothesized to be the causal influence

Dependent Variable: The dependent, or responding variable, is the factor that may change as a result of changes made in the independent variable. –

Criterion by which the results of the experiment are judged.

–

Variable that is expected to be dependent on the manipulation of the independent variable

Control Group: In a scientific experiment, the control is the group that serves as the standard of comparison. The control group may be a “no treatment" or an “experimenter selected” group –

Group of subjects exposed to the control condition

–

Not exposed to the experimental treatment

Experimental Group Group of subjects exposed to the experimental treatment Experimental Validity: •

Internal Validity –

•

Indicates whether the independent variable was the sole cause of the change in the dependent variable

External Validity –

Indicates the extent to which the results of the experiment are applicable to the real world

Tools of Control

Internal Sources

Pre-Test/

Control

Post Test

Group

History

X

Maturation

X

Randomization

Additional Groups

Pre-Testing

X

Measuring Instrument

X

Statistical Regression

X

Differential Selection

X

Experimental Mortality

X

Interaction of Factors

X X

X

X

External Sources Pre-Testing Differential Selection Procedures Multiple Treatment

X X

X X

EXPERIMENTAL DESIGNS

Experimental Research design: The experimental method includes many different research design to counteract the possible confounding variables that could influence the internal and external validity of a study. The various designs have different strength and weaknesses. Anyone who chooses to do an experiment must balance the focus of the research question against the time resources available for conducting the study in order to choose the best design. Types of Experimental Research 1. Pre-experimental designs (low degree of control) 2. True experimental designs (high degree of control) 3. Quasi-experimental designs (medium degree of control) 4..Correlation and Ex Post Facto Design 1. Pre-experimental designs (low degree of control)  They are not characterized by random selection of participants from a population.  They do not include a control group.  They allow little or no control over extraneous variables that might be responsible for outcomes other than what the researcher intended.  There is no comparison between an equivalent non- treatment group is made Types of Pre- Experimental Design  The One-Shot Case Study  The One- Group Pretest posttest Study

 The Static Comparison/ Posttest only Design with Nonequivalent Groups 2. True experimental designs (high degree of control)  True experimental design is regarded as the most accurate form of experimental research, in that it tries to prove or disprove a hypothesis mathematically, with statistical analysis.  For some of the physical sciences, such as physics, chemistry and geology, they are standard and commonly used. For social sciences, psychology and biology, they can be a little more difficult to set up.  For an experiment to be classed as a true experimental design, it must fit all of the following criteria.  The sample groups must be assigned randomly.  There must be a viable control group.  Only one variable can be manipulated and tested.  The tested subjects must be randomly assigned to either control or experimental groups. Types of True Experimental Design (Classical Experimental)  Pretest-posttest Control Group Design  Posttest Only Control Group Design  The Solomon Four Group Design 3. Quasi-experimental designs (medium degree of control)  Quasi-experimental design is a form of experimental research used extensively in the social sciences and psychology.

 Designs for which it is impossible randomly assign participants to all groups are called quasi experimetal designs because they are not truly experimental. Types of Quasi- Experimental Design  Pretest-Posttest Nonequivalent Group  Time Series Design  Nonequivalent Before- After Design

4. Correlation and Ex Post Facto Design  To seek for cause-effect relationships between two sets of data  To search backward from consequent data for antecedent causes

Pre-Experimental Design - loose in structure, could be biased Aim of the Research

Name of the

Notation

Design

Paradigm

To attempt to explain

One-shot

X» O

a consequent by an

experimental

antecedents and consequences. The least

antecedent

case study

reliable of all experimental approaches.

To

evaluate

the

influence of a variable

One

group

O»X»O

pretest-

Comments

An

approach

that

prematurely

links

An approach that provides a measure of change but can provide no conclusive results.

posttest To

determine

the

influence of a variable on one group and not on another

Static

group

comparison

Group 1: X »

Weakness lies in no examination of pre-

O

experimental

Group 2: - » O

Conclusion is reached by comparing the

equivalence

of

groups.

performance of each group to determine the effect of a variable on one of them.

True Experimental Design - greater control and refinement, greater control of validity

Aim

of

the

Name of the

Notation

Research

Design

Paradigm

To study the effect

Pretest-posttest

R--[O»X»

This design has been called

of an influence on a

control group

O

"the

carefully controlled

[O»-»O

sample

Comments

old

workhorse

of

traditional experimentation." If effectively carried out, this design controls for eight threats of internal validity. Data are analyzed

by

analysis

of

covariance on posttest scores with the pretest the covariate. To minimize the

Solomon four-

R--[O»X»

This is an extension of the

effect of pretesting

group design

O

pretest-posttest control group [O»-»O

design and probably the most

[- » X » O

powerful

[-»-»O

approach. Data are analyzed by

experimental

analysis of variance on posttest scores. To

evaluate

situation

a that

cannot be pretested

Posttest

only

control group

R--[ X»O [-»O

An adaptation of the last two groups in the Solomon fourgroup design. Randomness is critical. Probably, the simplest and best test for significance in this design is the t-test.

Quasi-Experimental Design - not randomly selected

Aim

of

the

Research

Name

of

the

Notation

Comments

Design

Paradigm

Nonrandomized

O»X»O

One of the strongest and most widely

situation in which

control

O»-»O

used

random

pretest-posttest

To

investigate

a

selection

group

quasi-experimental

designs.

Differs from experimental designs

and assignment are

because test and control groups are

not possible

not equivalent. Comparing pretest results

will

equivalency

indicate between

degree

of

experimental

and control groups. To

determine

influence

of

the a

Time

series

experiment

O»O»X»

If

substantial

change

follows

O»O

introduction of the variable, then the

variable introduced

variable can be suspect as to the cause

only after a series of

of the change. To increase external

initial

validity, repeat the experiment in

observations

and only where one

different

group is available

conditions.

To

bolster

the

validity of the above design

with

places

under

different

Control group time

O»O»X»

A variant of the above design by

series

O»O

accompanying it with a parallel set of

O »O »- »

observations without the introduction

O»O

of the experimental variable.

[X1 » O1]

An on-again, off-again design in

»[X0 » O2]

which the experimental variable is

» [x1 » O3]

sometimes present, sometimes absent.

the

addition of a control group To control history in

Equivalent

time designs with a

samples

variant of the above design

time-

Correlation and Ex Post Facto Design

Aim

of

the

Name of the

Notation

Comments

Research

Design

Paradigm

To seek for cause-

Causal-

-»

A very deceptive procedure that

effect relationships

comparative

Oa ¥ Ob

requires much insight for its use.

between two sets of

correlation

«-

Causality cannot be inferred merely

data

studies

because

a

positive

and

close

correlation ratio exists. To

search

backward

from

Ex post facto

This approach is experimentation in

studies

reverse. Seldom is proof through

consequent data for

data substantiation possible. Logic

antecedent causes

and inference are the principal tools of this design

Steps involved in Experimental Research

 State the research problem  Determine if experimental methods apply  Specify the independent variable(s)  Specify the dependent variable(s)  State the tentative hypotheses  Determine measures to be used  Pause to consider potential success  Identify intervening (extraneous) variables  Formal statement of research hypotheses

 Design the experiment  Final estimate of potential success  Conduct the study as planned  Analyze the collected data  Prepare a research report

Advantages of experimental Research: –

Strength with which causal relationships can be inferred.

–

Ability to manipulate one or more variables.

–

Proven to be a very useful and robust scientific method (i.e., withstood the test of time).

–

Best establishes cause and effect relationships

–

Adjustability/ flexibility

Disadvantages of experimental Research: –

Tight controls often produce artificial conditions that could limit the generalizability of the findings (i.e., internal vs. external validity trade-off).

–

Time consuming.

–

Expensive.

–

Lack of “ reality”

–

Unrepresentative samples

–

Human behavior is very complex and cannot be fully studied using experimental methods.

Conclusion: Experimental research is generally regarded more important than other type of research. This type of method used to study casual relationships. This is mainly due to use of true experimental designs, which use equivalent comparison groups, allowing manipulation of experimental variables and controlling influences on the dependable variables.pre experimental

design is the weakest experimental design, having a high external validity and low internal validity. Quasi experimental design is compromise between two. Finally it is difficult to to have experiment in LIS than other sciences.

References: 1. Myers, J. L. (1972). Fundamentals of experimental design. Boston: Allyn & Bacon. 2. Shedish, W. R., Cook, T. D., & Campbell, D. T. (2001). Experimental and quasiexperimental designs for generalized causal inference. Boston: Houghton Mifflin, 3. Kirk, R. E. (1968). Experimental design: Procedures for the behavioral sciences. Belmont, CA: Brooks/Cole. 4. Campbell, D. T., & Stanley, J. C. (1966). Experimental and quasi-experimental designs for research. Chicago: Rand McNally. 5. Cook, T. D., & Campbell, D. T. (1976). The design and conduct of quasi-experiments and true experiments in field settings. In M. Dunnette (Ed.), Handbook of industrial and organizational psychology (pp. 228-293). Skokie, IL: Rand McNally. 6. Cook, T. D., & Campbell, D. T. (1979). Quasi-experimentation: Design & analysis issues for field settings. Boston: Houghton Mifflin. 7. Aronson, E., & Carlsmith, J. M. (1968). Experimentation in social psychology. In G. Lindzey & E. Aronson (Eds.), Handbook of social psychology: Vol. II. Research methods (2nd ed., pp. 1-79). Reading, MA: Addison-Wesley. 8. http://www.slideshare.net/uroojshafqat/experimental-research-11401013 9. Leedy, P.D. (1997). Practical research: Planning and design (6th ed.). Upper Saddle River, NJ: Prentice-Hall, Inc., p. 232-23 10. http://www.okstate.edu/ag/agedcm4h/academic/aged5980a/5980/newpage2.htm 11. www.suu.edu/faculty/white/researchdesign 12. http://allpsych.com/researchmethods/quasiexperiment 13. http://webs.wofford.edu//reidak 14. http://www.uic.edu/classes/socw/socw560/EXPERMT/sld027.htm 15. www.experiment-resources.com 16. Neil J. Salkind “ Exploring Research 6th Edition’’ Experimental Research, 217-245. 17. Ref : Joseph Luzzi, Ph.D., Educational Psychologist