Table of Contents
Answer 1
The purpose of conducting an experiment is to find out the results for investigation using a cause-and-effect relationship (Jackson, 2012). This is done by manipulating a variable and understanding its effects on depending variables. An experimental design accomplishes its purpose by controlling the environment .i.e. observing the results on the dependent variables by controlling the independent variable designed for the study. In simpler terms, the control of the experiment is in the hands of the researcher who seeks to understand the effects of one variable on another variable. For example, experimental design helps in understanding the relationship between the variables, which helps in accepting or rejecting a hypothesis.
Answer 2
Experimental design is an approach using which researchers can provide statistical evidence to experiments’ outcomes (Paasch, Khalili, & Bonaventure, 2013, p. 393). Experimental design has certain advantages and disadvantages in a business study. Some of the main advantages of experimental design for a business study include opportunity to seek insight into the methods of instruction, more control on variables in understanding relationships, helps in knowing the effects of certain internal or external environmental factors on a business .e.g. effect of currency exchange on international business profitability, high level of control on variables, and clear cut conclusions based on the results of the experiments. On the other hand, some disadvantages may include more potential of human errors in controlling variables, risk of artificial results due to human control, and personal bias towards one particular business field may affect the results of the study.
We can do it today.
Answer 3
An ideal study is one that endorses both types of validity: internal validity and external validity. The reason is that depending on one type of validity cannot be as effective as it should be for an experimental design study. Replication of results is an issue in case of external validity .i.e. even one unsuccessful replication can affect the results that are replicated successfully for more than once. Similarly, internal validity becomes useless in case a study lacks generalized results. Moreover, internal validity tends to make a study more detrimental, whereas external validity decreases with the increase in internal validity. This results in decreasing the extendibility of the study. Therefore, it can be said that both types of validity are essential for an experimental design since it involves control over variables.
Answer 4
To answer this question, it is important to understand that in an experimental study, the researcher seeks to understand the effect of treatment or control of one variable on other variables. The researcher can control the experiment by manipulating one variable by either increasing or decreasing its value or strengthening or weakening its influence and noting its effect on the rest of the variables. The control group is one on which treatment does not make an impact and acts as a baseline from where outcomes can be judges or conclusions can be made (Pithon, 2013). In a single comparison group, which is based on an experimental and a control group, one comparison is used to determine whether a treatment has a different impact on an outcome or not, whereas in a multiple comparison group, more than one comparisons can be used for understanding the effect of treatments with non-treatments.
Answer 5
Confounds are those variables that a researcher is not able to control or ignores to control in an experiment. Such variables can have a hidden impact on the results of the experiment. Ignoring such variables can affect the internal validity level of the experiment.
One-shot experimental case study where there is no control group and no internal or external validity is a research study with three confounds. One way to alter this design would be to use one-group pre-test post-test design in which minimal control would be there in terms of structure and measurement could be done before and after the experiment. This will ensure minimal internal validity. However, the disadvantage would be no external validity.
Researchers can also change the design using within-subjects design. Doing so, the researchers can eliminate confounds based on non-equivalency. However, a disadvantage would be the risk of confounding because of demand characteristics.
Researchers can also change the design using matched-subjects design. Doing so, the researchers would be able to minimize the differences and selecting participants based on some certain matching attributes. However, time consumption would increase and difficulty in understanding the differences would arise if altered to this design.
Answer 6
Cause in experimental research refers to the event or treatment that shapes the outcomes. For example, in understanding the effect of smoking on a person’s health, cause will be the increase or decrease in the volume of smoking that can bring a negative or positive effect on the health of a person. Cause is a very important concept in research because the outcomes are primarily based on it (Cozby & Bates, 2012). In experimental designs that are based on cause and effect relationships, conclusions are drawn based on the effect of a cause. There is a unidirectional relationship between causation and correlation. Causation supports the concept that one event leads to another event, whereas correlation means that change in one variable is not necessarily due to the change in some other variable. Variables are correlated if they have a causal relationship.
Answer 7
Between-participants design:
Two equal groups would be made, an experimental and a control group. The people in the experimental group would be asked to keep smiling, whereas the participants in the control group would be controlled using mood evaluation instrument. An advantage of this design is that the researchers can effectively compare the results of the relationship between an experimental and a control group. A disadvantage of this type of design is time consumption since the researchers need to fully populate the groups to get the results.
Within-participants design:
There would be one group of participants. They would be asked to either smile or not during the first ten minutes or so and then mood evaluation instrument would be used to find the difference. An advantage of this design is that the researchers can make a reliable and effective use of statistical models. Similarly, this model is less time consuming because of lack of need to form a control group. A disadvantage of this type of design is that it is more open to demand characteristics (Jackson, 2015, p. 195).
Matched-participants design:
There would be two groups based on matching of gender, age range, or religion. The rest of the experiment would be same as that for between-participants design. An advantage of this design is that the researchers are able to draw results based on equality between the two groups. Moreover, the advantages of the between-participants design also apply to this design. A disadvantage of this type of design would be the need to have more research participants in each group, thus, time consuming and risk of confounding is there.
Even Question Answers
Answer 1
There can be more than one confounds in the study without a control group. As the study lasted for nine months, it can be possibility that some events might have occurred in the lives of the participants that either made them quit the group or made them happy for some reason. Now, the researcher can be wrong in assuming that improvement occurred due to his/her therapy technique. Secondly, confounding element is the number of participants who left the study due to any reason. The leaving participants could be highly depressed due to which they left the group during therapy. Now, the researcher can be wrong in assuming that his/her therapy made 85 percent people happy. It can be a possibility that the ones who left the group would not have shown improvement. This would have decreased the percentage given by the researcher.
Answer 2
“Internal validity refers to the accuracy of conclusions about cause and effect” (Cozby & Bates, 2012, p. 69). It also refers to the degree to which a study’s results can be based on the independent variable. A good study and measurement design is the key to strong internal validity. On the other hand, external validity, which is based on experiments and generalization of results, explains the level to which a study generalizes the results. Internal validity helps in determining the level of accuracy of an experimental research. Lack of external validity, on the other hand, acts as an obstacle in generalizing the results of a study to other conditions. In short, both internal and external validity are essential in measuring the meaningfulness, correctness, appropriateness, and usefulness levels of a research.
Answer 3
The power of use of statistical models is same for both within-subjects and matched-subjects designs as both of the designs allow the researchers to make an effective use of statistical models to analyze the results. This mainly occurs due to minimized variability between the participants. In within-subjects, there is no separate experimental or control group and all participants are in the same group, whereas in matched subjects design, there is matching of attributes for research participants. As for differences, one of them is the requirement of having one group for within-subjects, whereas a control group and an experimental group for matched-subjects. Another difference is that multiple times evaluation of testing effects is done in case of within-subjects design, whereas for matched-subjects design, it is not the requirement.
- Cozby, P. C., & Bates, S. C. (2012). Methods in Behavioral Research (11th ed.). Boston, MA: McGraw Hill.
- Jackson, S. L. (2015). Research Methods: A Modular Approach (3rd ed.). Stamford, CT: Cengage Learning.
- Jackson, S. L. (2012). Research Methods and Statistics: A Critical Thinking Approach (4th ed.). Belmont, CA: Cengage Learning.
- Paasch, C., Khalili, R., & Bonaventure, O. (2013). On the benefits of applying experimental design to improve multipath TCP. Proceedings of the ninth ACM conference on Emerging.
- Pithon, M. (2013). Importance of the control group in scientific research. Dental Press J Orthod., 18(6), 13-14.