Quasi – Experimental Designs

Quasi – Experimental Designs

            These designs provide control of when to and to whom the measurement is applied but because random assignment and control treatments has not been applied the equivalence of the groups is not assured.

The Pretest – Post test Non- Equivalent Group Design

                          This design is often used in classroom equipments when experimental and control group are such naturally assembled groups as intact classes , which may be similar. As in the pretest – post test equivalent group design , analysis of covariance may be used with the pretest as the covariate. Because this design may be the feasible one.

The time – Series Design

               In this design a measured change or gain from observation to observation would indicate that the treatment had an effect. This design particularly                 to the failure to control the extraneous variable for it is possible that distracting simultaneous event at the time of the intervention would provide a rival hypothesis for the change.

The Equivalent Time Samples Design

                 In this the design the experimental condition X is present between some observations and not (C) between significance usually ‘t’ test or an analysis of variance. If the difference between the means  too great to attribute to sampling error  the difference may be attribute to the treatment variable effect.

The Pretest – Post test  Equivalent Group Design

               This design is similar to the previously described design except that pretest are administerd before the application of the experimental and treatment period. This strong design but there may be a possibility of the influence of the interaction effect of tutering with the experimental variable.

The solamon Four Group Design

            In this design ;

1) Subjects are randomly assigned to four groups.

2)  Two groups receive the experimental treatment.

3)One experimental group receives a pretest.

4) Two groups do not receive treatment.

5) One control group receives a pretest.

6) All four groups receives post test.

                       The design is really a combination of the two-two group designs previously described the post test only and the pre test – post test. The Solaman Four Grop Design permits the evaluation the effects of testing , history and maturation. A major difficulty is finding enough subjects to assign randomly to four equivalent groups.

Merits of Experimental Method

            Experimental method recemples scientific method in any respects . Therefore efforts are being made to make more and more use of this method .

1) It is rarely possible to solve scientific problems ascertain appropriateness of techniques or determine accuracy of the data determined by reasoning alone. These factors can best be learned by experimentation .

2) Experimental method helps in the establishment of cause and effect relationship more clearly than others methods .

3) It is more precise and accurate since the variable under study is manipulated learning others untouched. We can also locate with precision the effect of varying degree of stimulus.

4) It is the best method for testing a hypothesis casual observation may help in the formulation of a hypothesis but it only through experimental method that hypothesis can be tested and verified.

5)  The conclusions drawn from an experimental method are subject to verification at any time .

6) Laws framed by experimental method are universal in their application. Through complete universality is only a myth.

7) Prediction can be done with sufficient accuracy.

Limitations of Experimental Method.

1) Extraneous variables influence the results of the experiment in ways that have difficult to evaluate .

2) Causation implies an in variant one – to – one relationship between certain consequents .In keeping with the emphasis on the law of the single variable, the investigator attempts to control all relevant factors except the experimental factor and thus it promotes certain outcomes which are then measured and attributed to the operation of the variable under investigation.

3] Experimental control imposing control on a situation tends to make the situation artificial. It would only serve to violate the basic principle of maintaining the situations in its natural form. A

4] Between initial and subsequent observations subjects may become tried, bored, wiser  or influence by the incidental learning and experiences that day encounter through normal maturation.

5] Pretesting may produce a practice effect that may make subjects more proficient in subsequent test performance

6] In some experiments the effect of treatment may carry over to subsequent treatments. This is one of the major limitations of the single group design.

7] In pretest-retest situations there is a normal regression effect may lead researches enormously to attribute unwarranted again to the lowest group and loss or small gains to the highest group.

8] Morality or loss of subjects likely in a long term experiment, in introduces a confounding elements.

 9] Contamination is a type of bias introduced when a researcher has some previous knowledge about the subjects involved in an experiment. This knowledge about of subject status may effect the objectivity of his judgment.

Conclusion

Though use of experimental method in research is limited as our field of inquiry is human beings who poses many problems but use of scientific, laboratories, model building, sample techniques, measurement of variables has made it possible to use this experimental method in special cases.