Methods of Research Experimental

If we are to advance beyond the dark ages of educational pre-science, we must emulate the experimental proficiency and zeal of colleagues in other behavioral sciences.
JULIAN C STANLEY
Significance of Experimental Work
Experiment has proved to be an essential and powerful tool in leading man Cowards progress. There would have been very little progress, as we find it today, without systematic experimentation.
The goal of all experiments is progress and good life. In so far as good education is recognized as the basis of adequate individual and social development, need for experimentation in education to improve educational practices and being is beg realized increasingly. The educationists are constantly searching for. more effective methods of instruction, more satisfactory techniques of evaluation, richer learning materials, more efficient systems of administrative organization. This search is assuming greater urgency in India because of the very rapid expansion and democratization of education: A number of new educational problems have arisen. For a successful solution of the multitude of problems and for the full realization of the educational aims set up, experimental work, adequate both in quality, should be carried out in varying degrees by all Those who are engaged in the teaching profession.
Meaning of Experimentation—Experimentation differs from normative or descriptive survey Methods and from other techniques of research in that the experimenter has some degree of control over some variables involved and the conditions under which variables are observed.
According to J. W. Best, "Experimental research is the descrip­tion and analysis of what will be, or what will occur under carefully controlled condition."
Basic Assumption Behind Experimentation
The basic assumption behind the experimentation is the law of the single variable formulated by John Stuart Mill. The principle reads as "If an instance in which the phenomenon Under investi2ation occurs, and an instance in which it does not occur have every circumstance in common save one, that one occurring only in the former, the circumstances in which alone the two instances differ is the effect. or the cause, or an indispensable part of the cause, of the phenomenon."
Stated in a simple language,(if two situations are alike in every respect, and one element is added to or removed from one but not the other, any difference that develops is the result of the operation of that element added or removed.;
`Experimental research describes 'what will be' when all relevant conditions are carefully-controlled.
Educational Experimentation
Meaning--W. S. Monore and M. D. Engelhard . describe as "Experimentation" is the name given to the type of educational research in which the investigator controls the educative factors to which a child or group of children is subjected during the period of inquiry and observes the resulting achievement...In the simplest type of educational experiment the investigator seeks to evaluate the influence of some one educative or "experimental" factor on .a single group of children. He must start the experiment with some measure­ment of the initial attainment of the children in the trait or ability to be influenced. He then .Subjects the group to the experimental factor, such as particular type :Of drill material in arithmetic, for the duration of the experiment. At the end, the investigator applies a final test for the purpose of determining the gain in achievement that has resulted from the application of the experimental factor"
Jahoda and others,.. regard social experiment as a method of testing hypothesis. Greenwood suggests that an experiment is the proof of a hypothesis which seeks to take up two factors into causal relationship through the- study of contrasting. situations which have been controlled,' on all factors except the one of interest the latter. being either the ,hypothetical cause or the hypothetical effect.. According to ',Festinger, "The essence of an experiment may be described as observing .the effect on a- dependent variable of the 'manipulation of an independent variable.")
Five Types of Experiments
Greenwood has mentioned following five types of experiments:
1. Trial and Error Experiment.
2. Observational Study.
3. Natural Experiment
4. Ex-post facto Experiment
5. Laboratory Experiment.
Experimentation and Laboratory Method
According to Travers, "Experimentation may occur within the laboratory or outside it Laboratory studies usually require, relatively small numbers of subjects and the careful control of many factors that can not be controlled in other .situations.. When experimentation requires equipment or complex apparatus, it may be necessary to work .within the. laboratory. Of course, the laboratory itself introduces variables, which it may be desirable to control but which - cannot be -controlled easily. For instance, human subjects who are introduced into a laboratory except to behave in a certain Way, or at leastleel that the situation calls for certain kinds of responses."
Walter R. Borg. thinks The experiment is the ultimate form of research design; providing the most rigorous test of hypothesis that is ihe
According to John V, Best, "Experimental research is the description and analysis of what will be, or what will occur; under care tally controlled situation.."'
Georg. G. IVIouly describes this method as, "Experimentation can he considered a "technique of deliberately staging a situation designed to force nature to provide a 'yes' or !fio" answer to Specific hypothesis concerning the phenomenon under discussion. "The same writer thinks that an experiment 'must be self-contained. and this in `turn. calls ter the satisfaction of three basic interrelated conditions :
(i) Control.
(ii) Randomization,.
(iii) Replication.
An experiment cannot be interpreted accurately unless these conditions are fulfilled,
Main Features of the Experimental Method
1. Experimentation involves an attempt to control all essential factors except a single variable,. which is manipulated with a view to determine and measure the effects of its operation.
2. It has been applied with considerable success in the: classroom by controlling significant-factors within certain limits.
3. In educational research, the basic condition of other things being equal' is difficult 'for fulfillment. In education: we deal with those subjects i.e., human beings who are complex.
4. The or roi group and experimental group are never as den-tic:1i as they ought to be tbr an exact experiment.
5. Experimental studies in education though never strictly empirical can yet approximate strictly' empirical research in many areas.
6. All experiments in education are ultimately experiments with children. It is, therefore, veiy important from ethical point of, view that they are not subjected to conditions that may harm them.
History of Educational Experimentation
The second half of the nineteenth century' saw the beginning of , educational experimentation. Wundt set up a psychology laboratory at Leipzig in 1879 where he studied experimentally the learning process and tried to clarify and define learning panel:11s. In the 1880's Ebbinghaus invented methods :of measuring association and memory. which affected teaching 'techniques. Thereafter, the experimental work of Thorndike, Judd and Freeman paved the way for further experimental research in education. The work of Rice and Cornmad during the last decade of the 19th century that was done on spelling achievement among students may be called the beginning of strict educational experimentation.-
Uses of Experimentation in Education
Main uses of experimentation in education are:
1. Determining and evaluating the adequacy and effectiveness of educational aims and objectives through the measurement of outcome.
2. Serving as a basis for the formulation, execution' and modification of educational policies and programmes.
3. Ascertaining the effect , of any change in the normal educational practices and programmes.
Major Steps in the Experimental Method
1. Selecting and delimiting the problem.
2. Reviewing the literature.
3. Defining the population.
4. Planning the experiment.
5. Conducting the experiment.
6. Measuring the outcomes.
7. Analyzing and interpreting the outcome.
8. Drawing up the conclusions.
9. Reporting the results.
I. Selecting and Delimiting the Problem. This involves;
i. Investigating the needs - in the field of action and deciding upon a problem.
ii. Conversion of the problem into a hypothesis that can be verified or refuted by the experimental data.
II. Reviewing the Literature- This implies the study of the literature related to similar problems.
III. Defining the population- It is necessary to define the population precisely so that there can be no question about the population to which the conclusion are to apply.
IV. Planning the Experiment. This includes;
(i) Determining the methods of experimentation.
(ii) Place of the experimentation.
Duration of the experiment
(iv) Determining the materials of the experiment.
(v) Conducting a: pilot study.
(vi) Selecting the subjects and groups.
V. Conducting the Experiment- This includes the following sub steps:
(a) Control of variables and non-experimental factors.
(b) Keeping a careful record of steps in the procedure.
(c) Applying the experimental-factor o—
VI. Measuring the Outcome---This implies giving a careful consideration to the selection of the criterion on the basis of which the results are to be measured.
VII. Analyzing and Interpreting the Outcomes—This implies the need for competence in statistical procedures.
VIII. Drawing up the Conclusions----Care must be taken to restrict the conclusions to the conditions actually present in the experiment. The conclusions of the study must be restricted to the population actually investigated and care must be exercised not to over generalize the results.
IX. Reporting the Results—The study should be reported in such a way that the reader can make a judgment as to its adequacy.
General Description of an Experimental Method
Control Factor--The customary method or device of doing a job is known as the control factor.
Experimental Factor---This is the new method, arrangement or device, which is being introduced to develop pupil' growth in skill's and knowledge in a new setting.
A comparison is made between the results of these.. situations and the difference between the control pupil mean and the experithent pupil mean would determine the relative superiority –of the method or factor showing the greatest pupil gain.

The various steps in a simple experiment of this type can be represented thus
Experimental Group Control Group
1. Pretest 1. Pretest..
2 Application of experimental factor 2. Application of control factor:
3. Final test. 3. Final test.
4. Measurement of pupil mean 4. Measurement of pupil mean gain
gain (Final test scores minus (Final test scores minus pretest
pretest scores.) scores.)
Experimental Designs
1. The single individual or single group design:
2. The parallel-group design (two or more).
a. Equated by random sample.
b. Equated by mean score.
c. Equated by matched pairs
d. Co-twin method.
3. Rotational method.
a. Single group.
b: Parallel, equated. groups
1. The Single It dividual or single 'group design--This type is probably the most elementary and less rigorous design. This type of experiment is carried on in comparing the growth of a single individual or growth under, two-sets of conditions. The group is subjected to an experimental and, a control factor for equivalent periods of time and then outcomes are compared. The: procedure may be listed as follows,
a. Test the group.
b. Introduce Method A.
Test and group again.
d: Note the gains.
e. Allow for a period of transition.
1. Test the, group again.
g. Introduce Method B.
h. Test the group again and note the gain.
i. Compare the games in (d,) and (h).
Advantages
I. It is simple to plan and operate.
2. lt-requires no equation or rotation of groups.
3. It is well suited to classroom use and provides a stimulus to better classroom teaching.
Limitations
1. The teacher may not necessarily be equally effective and enthusiastic about both the methods.
Gains Due to practice efforts are generally greater in the second, testing than the first one. •
3. In the second phase of this experiment, students involved are a little older and a little mature than they were when they entered the first phase.
4. The very novelty of the new method may produce greater gain, irrespective of its real merit
Some Important Points to be Considered. while Applying this Method
1.• The unit of work should be of equal difficulty.
2. The unit of work should be equally interesting.
3. Each unit of work should be given the same time.
4. The tests produce scores of equate scale value.
Parallel or Equivalent Group Method
This method of grouping is more complex than the one-group experiment but more accurate. In this, two or more groups of subjects equivalent in all significant respects are selected One of these parallel groups serves as the control group and the experimental factor or factors are applied to the other groups, one by one for a specified period of time. The difference observed at the end of the period between. The control and the experimental groups is expected to be due to the introduction of experimental factor.
The main steps in this method are:
ii. Securing equivalent groups.
iii. Applying the experimental factor.
iv. Comparing the results.
v. Interpreting and reporting the results.
Factors Leading to the Formation of Equating Groups
i. Chronological age.
ii Intelligence quotient.
iii. Sex.
iv. Race.
v. Physical conditions.
vi. Previous achievement.
vii. Study habits.
viii. Personality traits.
a)-43y random selection--This consists of selecting groups by some system of random selection
b) By equating on the basis of mean scores and standard devia­tions—Groups are selected Whose mean ages, intelligence scores and other factors considered significant are the same.
c) By matched pairs—Pairs of students are formed whose characteristics are as nearly alike as possible in respect of age, sex, intelligence, home back ground, race, etc.
d) By co-twin- method-- Pairs of identical twins are placed in the control and the experimental factor.
Limiting-Factors
i. Equating groups,
ii. Tremendous administrative problems involved in
recognizing classes.
in. Teacher's competency and enthusiagro.
iv. Problem of satisfactory validity and measuring instruments.
Rotational Methods--Groups are rotated at periodical interval. For example, Groups A and B may use methods A and B. respectively for a particular period and then exchange methods for the same period. A comparison is then made of the relative gains of each of the groups under two methods.
This method enables us to minimize the influence of uncontrolled factors and to provide a more convincing test of the superiority of the particular method under investigation.
Need for a Cautious Approach to Experimentation
1. The subjects should be representative as to the age, sex, reliability of the grade, intelligence. etc.
2. Such subjects should be selected as are reasonably expected to be available throughout the period of the experiment.
3. The number of subjects be adequately large to allow for losses for elimination on account of absences from one or more tests
4. The materials must be appropriate to the subjects and the experimenter.
5. The place in which the experiment-is to be conducted must be typical of the situations to which the results of the experi­ments are expected to apply.
6. Appropriate measuring instruments should be used.
7. Since the effectiveness of teaching learning in any classroom situation can rarely be attributed to a single factor is it is very essential to exercise great caution in interpreting the "resulting experimental conclusions.',, The basic conditions of 'other things being equal' is difficult of fulfillment in educational research. Besides the control group and the experimental group are never as identical as they ought to be for an exact experiment.
Advantages of Experimental Method
1: Establishing direction of causality4--The most important advantage of the experiment is that the relationship that you actually observe is clear in its casual direction.'
2. Low cost —Compared to that of alternative. research methods, the laboratory cost of a laboratory experiment can often be low.
3. Convenience—An experiment can be done whenever you like, to suit your convenience.
4. Adjustability pofyariahtes and parameters--Unlike a survey, an experiment permits us to arrange the parameters and vary the variables in whatever fashion we desire, to look for whatever effects interest us.
In experimentation we can systematically refine the relationship we are investigating..
5. Replication:— By repeating (replicating) an experiment, we ail obtain an average result; our conclusion is not based on a single observation that might be unusually high or unusually low., Replication is one of the most useful tools in obtaining valid results:
6. Unraveling Multivariable Causation----If two in independent variables are closely related in the world outside the laboratory, a survey cannot easily determine which of them causes variation in the dependent variable. A laboratory experimenter can track down the extent to which each is responsible for change in - the independent variable because the laboratory experiment holds some -factors, constant as it varies others and vice versa.
Disadvantage of Experiments.-
1. Lack of Reality—The most important -disadvantage of the laboratory experiment is that we c n never be sure that the analogy between the experiment and the real World' really holds.
2. Unrepresentative Samples--Subjects on whom the experiment is run in the laboratory may be very unlike the people in the real world about whom we wish to draw conclusions.
3. Expensive—Some experiments may be very costly than other research methods.
4. Hazardous Outcomes,---Most of the sex experiments may be- unethical or objectionable to good taste.
Limitations of Experimental Method
Experimental method suffers from a score of limitations:
Extraneous Variables---In educational experiments, a number of extraneous variables are present in the situation or are generated by the experimental design and procedures, which influence the results of the experiment in ,ways, which 'are difficult to evaluate.
Concept of Causation—Causation means an invariant one4o-one relationship. between cert4irrOtecedents and certain consequents. It is almost impossible to equate into situations in 'all respects except for the factor whOse effect is being investigated.
Experimental Control-Imposed control in an educational situa­tion tends to make it artificial.
Maturation - There is every. possibility that between initial and subsequent observations, children may become tired or wiser or influenced by incidental. learning's and experiences . that they encounter. through normal maturation.
Testing-Testing may sensitize -children by Making them more '.aWareOrtlie concealed purposes of the researcher and may serve as a stimulus to change.
Unsuitable Instruments---Use of unsuitable instruments or techniques to describe and measure various aspects of behavior of children is likely to affect the validity of an experiment.
Differential Selection—Selection bias Es represented by the non­equiValenee of control and experimental groups.
Contamination --Contamination is a type of bias introduced when a. researcher has some previous knowledge about the children involved in an experiment.
Problems for 'Discussion
1. Bring out clearly the need for experimental research. What are its chief characteristics and limitations?
2. What precautions should he taken in an experiment in education to obtain reliable results?