Exploring the Logic of Experimental Design, Research Paper Example

Contents

Jackson (2012) even-numbered Chapter Exercises (p. 244). _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 3

What is the purpose of conducting an experiment? How does an experimental design accomplish its purpose? – 435 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 4

What are the advantages and disadvantages of an experimental design in an educational study? 5

What is more important in an experimental study, designing the study in order to make strong internal validity claims or strong external validity claims? Why?_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 5

In an experiment, what is a control? What is the purpose of a control group? Of single or multiple comparison groups? _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ 6

What are confounds? Give an example of a design that has three confounds. Describe three ways to alter the design to address these confounds and explain the advantages and disadvantages of each. _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 7

What does cause mean and why is it an important concept in research? How are correlation and causation related? _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 8

You are a researcher interested in addressing the question: does smiling cause mood to rise (i.e., become more positive)? Sketch between-participants, within-participants, and matched-participants designs that address this question and discuss the advantages and disadvantages of each to yielding data that help you answer the question. Describe and discuss each design in 4-5 sentences. _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 9

References. _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 11

1. Jackson (2012) even-numbered Chapter Exercises (p. 244).

Q.2. Maturation is the most prominent confound and this occurred due to long time period took for conducting this study. Another possible confound is Hawthorne effect where the individuals are aware of the fact that they are being studied and hence changes their behavior. Selection confound is also likely to exist and the subjects as there is chance for preexisting difference between each chapters categories. Mortality can also be considered as 21 participants left the study during its period and only 85% improved on completing the program(Jackson, 2012).

Q.4. Every study aims to demonstrate the relation between a cause and its effect and when we are able to generalize the findings of a study to a large population it is the one with external validity. An experiment becomes internally valid in the absence of confounds with in the study. Those researches that can be generalized will have more practical value and hence external validity is important. Without internal validity a research will not be able to reach generalization between cause and effect and thus internal validity becomes important to researcher (Jackson, 2012).

Q.6. In within subject design same subjects are used for experimenting each condition. In matched subject design each subject from a particular group is matched with different variables of other group. Both are similar to the extent that all subjects of one group is used to experiment each condition. Only difference is that in with in subject design subjects are matched together and in matched subject design each subject is used separately. Difference arises when all subjects of a group are not experimented with all the conditions. When some are selected and used for particular conditions there arise a difference in this type of experiment (Jackson, 2012).

2. What is the purpose of conducting an experiment? How does an experimental design accomplish its purpose? – 435

An experiment is conducted mainly for two purposes. First is to test the effectiveness of a program and next to provide tests empirical in form for hypothesis that are derived from theories. If an experiment is done for testing the effectiveness of a program or treatment it must be able to provide effective and vital information about the treatment or program in an extensive array of areas. Only an experiment that is well conducted will be able to achieve this. If the experiment is conducted for proving a hypothesis there are two chances. When the result of such an experiment is consistent with the expectations of hypothesis such an explanation proposed will receive support. If the result is different than another hypothesis will have to be developed and experiment will have to be conducted to find whether it can be proved (Ariola, 2006).

If an experiment is capable of meeting random assignment and experimental control, it will be able to accomplish its purpose. Experiments are used for testing hypothesis due to its high degree of control. Control is used for meeting necessary conditions (time order, covariation and plausible alternative cause’s elimination) for an inference that is casual, that is, it is the independent variable that caused changes in dependent variable. Formation of comparable groups before introducing independent variable through placement of sampling objects in experimental condition is called as random assignment. This is helpful for balancing the characteristics of subjects in two groups (Ariola, 2006).

3. What are the advantages and disadvantages of an experimental design in an educational study?

Main advantage of experimental design is that it is capable of demonstrating casualty when conducted properly. Researcher can manipulate independent variables exposure. In this type confounding variables can be eliminated to a great extent. Participants can be randomly assigned to levels or conditions of independent variables. Thus with these two features of assigning of participants randomly to conditions of independent variable and its manipulation makes experimental design different from correlation. Without these two features experimental design cannot be called as true experiment (Jackson, 2012).

Main disadvantage is that it is very difficult to conduct good experiments. Experimental designs need a lot of energy and human resource. To conduct an experimental design successfully and in a good manner one requires lot of cleverness, ingenuity and experience. It is very usual for experiments to take that behavior in which there is interest, even if they are out of context. In such a case there will be artificiality and hence it is said that under such circumstances it is difficult to generalize the findings of an experiment in other contexts. Another issue is with ethics and the ethics of manipulating exposure of people to subject of study is under question (Jackson, 2012).

4. What is more important in an experimental study, designing the study in order to make strong internal validity claims or strong external validity claims? Why?

It is more important to design the study for making strong internal and external validity claims (Campbell & Stanley, 1963). A research is conducted for determining the relation between cause and effect. If we are able to prove that a study has internal validity at high degree then it can be concluded to have strong causality evidence. On the other side if there is low internal validity evidence of causality is to be considered as low or nil. Thus with internal validity we are able to prove that the claim can be substantiated with sufficient evidence. Any claim that can be substantiated can be generalized. Though it is necessary to have strong external validity it is also possible for the external validity to become weak with just one counter example. But if a study with strong internal validity cannot be generalized it is of no use (Cronbach, 1982). For a treatment to become relevant to a broader context it is necessary for casual inference to go beyond specific conditions. Though casual effect can be affirmed by internal and statistical conclusion validity together it is essential to have strong external validity and construct validity for generalization of casual conclusion in a different setting (Briggs, 2008).

5. In an experiment, what is a control? What is the purpose of a control group? Of single or multiple comparison groups?

Control can be simply said as a portion on which experiment is not performed. Control is the factors that are controlled by the person conducting experiment. For example while conducting an experiment about after effects of drugs, control is those who do not consume those drugs. This will help the experimentalist to know whether a change took place in those persons who took drugs is actually due to consumption of drugs or is due to other factors. Control group is the one that is separated from experiment and this group cannot influence test results. They are independent variables and are not influencing the experiment. But it is necessary to have control for it will help the experimentalist to draw conclusions or find alternate explanations with regard to experiment results. Control groups are not necessary for all experiments. But they are really useful for those experiments which are difficult to segregate or are complex in nature. Single group refers to a group on which study is conducted and no comparison is made with another group on which study is not conducted. Multiple comparison group generally refers to comparison of two groups and at times more, and will include control group and treatment group (Jackson, 2012).

6. What are confounds? Give an example of a design that has three confounds. Describe three ways to alter the design to address these confounds and explain the advantages and disadvantages of each.

Confounds are independent variables that effect experimental groups in varying manner. Those variables treated for the purpose of research are not considered as confounds (Bernard, 1999). For example a study was conducted to find whether children can be taught alphabet easily using singing method or pronouncing method. For this two groups were selected. It was found that children studies alphabet easily using singing method. But there are chances for several confounders in this case. Age, sex and IQ are three confounders. There is chance that singing group were all female with high IQ and higher age and those in pronouncing group were all male of lower age and lower IQ. In order to avoid such issues what is to be considered is to select members of similar properties to each group. In this case male and female students of same age and same IQ can be selected in equal numbers to both groups.

If children of both sexes are selected in both groups it is possible to know whether both male and female students learn alphabet easily using singing method. To reach to this conclusion it is also necessary that children thus selected are of same IQ level and are of same age. If there is age or IQ difference between children of both groups we will not be able to know whether it is because of this that they are learning through singing easily.

7. What does cause mean and why is it an important concept in research? How are correlation and causation related?

Cause is a factor that is capable of bringing change in the state of a phenomenon or its process. A cause may be sufficient or necessary. Necessary cause is the one condition that is unavoidable for an effect to occur that is no effect is possible without that cause. Sufficient cause is the one that can guarantee a particular effect. Sufficient cause is necessary for an effect to take place and necessary cause alone cannot make sure that an effect will take place for sure. But it cannot be said that sufficient cause is the only possible method to get a particular result or effect.

Correlation is the relation between two variables and causation is the fact that causes an effect. In correlation there is a mutual or complementary relation between two or more variables and this means that one thing is caused by other. Causation can be said as a relation between independent and dependent variable while result is not caused by independent variable. Study of correlation will help to understand the relationship between variables. Thus an experiment can be conducted more precisely and understand causation better. But mere correlation cannot lead to an assumption of causation (Babbie, 2010).

8. You are a researcher interested in addressing the question: does smiling cause mood to rise (i.e., become more positive)? Sketch between-participants, within-participants, and matched-participants designs that address this question and discuss the advantages and disadvantages of each to yielding data that help you answer the question. Describe and discuss each design in 4-5 sentences.

Researcher wants to know whether smile is associated with positive mood of a person and hence conducts research in this regard. Different groups are selected for the purpose and effect of smile at varying situations is observed.

Between participants – It is noticed that when a participant smiles to another in usual case other one also smiles. This is due to causation and even then there are several confounds that are likely to effect this behavior. Participant might do so to please experimentalist, experimentalist might be biased, participants might be of same mental stage or participant might be trying to make a fake smile on his face to please other participant. All such confounds are possible between participants. All such disadvantages will influence the result. At the same time if a participant smiles to other participant in order to please him or experiment and is feeling happy in mind it is a positive factor and conclusion can be drawn that smile enhances mood.

Within participants – Participants might smile because all the members of the group are smiling and are seemingly in a pleasant mood. Here also there are chances of confounds like experimentalist bias, where experimentalist selects people consciously with an intention to keep them smiling. Another confound is participant trying to please experimentalist or others in the group even if he is not happy in mind. As all the participants of groups are smiling there comes a feeling that everyone is supposed to smile leading to field bias. On the other hand if the participants are all feeling happy being in a company where everyone is smiling and is feeling happy at mind it is a positive factor and conclusion can be drawn that smile can increase your positive mood.

Matched participants – There is higher chance for drawing the conclusion that smile can raise your mood positively in such a situation. This is because participants are matching in several ways like age, gender and social background, educational and emotional background. Even then there is chance for experimenter effect, selection, and participant effect and field wide bias. But if all the participants of a matched group react in a similar manner it can be assessed that people from matching situations feels an increase of positive mood with smile.

References

Ariola, M. M. (2006). Princples and methods of research’ 2006 Ed. Manila: Rex Bookstore, Inc.

Babbie, E. R. (2010). The Practice of Social Research. California: Cengage Learning.

Bernard, H. R. (1999). Social Research Methods:Qualitative and Quantitative Approaches. California: Sage.

Briggs, D. C. (2008). Comments on Slavin: Synthesizing causal inferences. Educational Researcher , 15-22.

Campbell, D., & Stanley, J. (1963). Experimental and quasi-experimental designs for research. Chicago, IL: Rand-McNally.

Cronbach, L. (1982). Designing evaluations of educational and social programs. San Francisco: Jossey-Bass.

Jackson, S. L. (2012). Research Methods and Statistics:A Critical Thinking Approach. California: Cengage Learning.