![Randomised block design
• Subject population is grouped into relatively
homogenous subgroups [ blocks] within which
the experiment is replicated
• It is used when there is inherent difference
between subjects and possible differences in
experimental conditions.
• If there are a large number of experimental group
the randomised block design is used because this
design make the group homogenous.](https://image.slidesharecdn.com/experimentaldesigntechniques-210522073615/85/Experimental-design-techniques-28-320.jpg)
Experimental design techniques
- 2. • The design of fundamental relationship under controlled situation is called experimental design. • Controlling is nothing but holding one variable constant while other variables are free to vary within the experiment. • Independent variables are manipulated in order to measure their effect on dependent variable.
- 3. • An experiment is an observation under controlled conditions. • It is a design in which some of the variables being studied are manipulated to control the conditions.
- 4. Types • Pre –experimental designs • Quasi-experimental design • Observational or true experimental design
- 5. Pre- experimental designs • It is the design in which the basic experimental steps are followed without assigning any control group. • The researcher studies a single group with no any comparison within this group. • It is a preparatory to true experimental design. • It is the simplest form of research design.
- 9. Quasi experimental design • The study subject are not randomly assigned to the groups. • Manipulation of an independent variable. • It differs from experimental research because there is no control, random selection and no active manipulation. • It is performed when a control group cannot be created or random selection cannot be performed. • It is used in certain medical and paramedical studies.
- 13. True experimental design • It employs statistical analysis to support or reject a hypothesis. • It has control group. • It is the most accurate type of experimental research it supports or disproves the hypothesis using statistical analysis • It is the only design where it cause and effect the relationship between the variables.
- 14. True experimental design satisfy the following • There must be a control group which wont change • An experimental group which will experience the changed variables. • There must be a variables that can be manipulated by researcher. • There must be a random distribution • It can be further categorized as
- 15. Post test only control group design
- 18. After-only design • Experimental and control group are similar. • Uncontrolled factors affect both the groups casually. • The experimental group is exposed to assumed variable but the control group remains unexposed. • After experimentation both the groups are compared there is some effect produced in the experimental group
- 19. • The control group assumed to e exactly same but it is not true.
- 20. Before after desi • Dependent variables is measured both before and after the exposure of groups to experiment. • It use several group. • It has one or more control group. • Major weakness is a group is influenced by the external factors . • It is more reliable than after- only d
- 21. Ex-post design • Researcher depend on historical background t, because sometimes it is not possible to divide the population into two clear and similar group. • It is necessary to study the entire historical background. • The past is studied through the present.
- 22. Panel study designs • Method of study of a particular subject over time by using different kinds of data. • The researcher secure direct evidences of time dependent relationship among variables. • It involves repeated observations on the same subject at different period of time is known as time-series study. • It is continuous , deep and reliable design • It has limitation such as loss of panel member absenteeism and rigid attitude of members.
- 23. Post -test only design • It has two randomly assigned groups. • Experimental group and control group. • The treatment is applied to experimental group. • The post-test is conducted on both the group to measure the effects of the treatment. • When pretesting is not possible or not required this design is commonly used.
- 24. Pre test post test only design • In this design experimental and control group are formed • They are pretested for independent variable. • During experiment only experimental group receives the treatment. • Finally both the groups are post tested to investigate the effects of the independent variable on the dependent one.
- 25. Solomon four group design • Four groups randomly formed. • Two experimental and two control group. • One pre tested and one un pretested group receives the treatment. • All four group receives the post test. • Post test result demonstrate the effect of the dependent variable comparing to the effect of the independent variable on the dependent variable. • It is used to eliminate potential source of error.
- 26. Factorial design • In this design two or more independent variable are simultaneously manipulated to observe their effects on the dependent variable. • It is useful to the researcher to test two or more hypothesis in a single experiment. • The independent variables in this design are called as factors.
- 27. • Subdivision of factors is called a levels. • It helps the researcher to investigate the individual effect of each treatment on the dependent variables as well as their joint effect.
- 28. Randomised block design • Subject population is grouped into relatively homogenous subgroups [ blocks] within which the experiment is replicated • It is used when there is inherent difference between subjects and possible differences in experimental conditions. • If there are a large number of experimental group the randomised block design is used because this design make the group homogenous.
- 29. • It is used when need to reduce the noise or variance in data. • The noise may be attributed to the differences between the blocks . • This design helps to detect and examine actual effects of interest much more accurately.
- 30. Cross over design • Different orders of treatment is randomly assigned to the subjects. • It is also known as repeat measures design. • The groups for comparison need to have an equal distribution as well as there should be high level of similarity among the subjects. • The subject in this design serve as their own control
- 31. • Limitation is subjects exposed to first treatment may effect their responses to second treatment. • Experimental research is aimed to draw conclusions by controlling the external factors in order to provide definite conclusions about the effect of the independent variables on the dependent variables. • The purpose of experimental research design is to test the hypotheses of a fundamental relationship between variables.