Design Experiments

Part 1
1. Explain the difference between the terms factors and treatments. Give an example
Within an experimental setting, the term factor which is also referred to as an independent variable is a variable that is explanatory and can usually be manipulated by the one conducting the experiment. Every factor typically has different values meaning that there are two or more levels for the factor. On the other hand, treatment can be defined as a combination or amalgamation of these factor levels. That is, it is a mixture of the different values of the independent variable (Heath, 2002).
2. What are some potential problems of a poor experimental design?
One of the potential problems of a poor experimental design is biasedness. This is largely for the reason that such a design fails overlooks and fails to take into consideration the rules of randomization and facilitates augmented experimental mistakes. In addition, this implies that there is the potential problem of validity and reliability of the research study or experiment.
3. Draw an outline of a randomized comparative experiment with five different treatment groups.
An example of a randomized comparative experiment can encompass the provision of two different kinds of treatment, including Vitamin C and Vitamin K. The subjects of the study are assigned in a random manner.
The following is an outline of a randomized comparative experiment with five different treatment groups:
Vitamin C
0 mg
40 mg
60 mg
80 mg
100 mg
Vitamin K
0 mg
20 mg
50 mg
70 mg
100 mg
4. Explain why it is important to use the following:
a) Randomization
Randomization takes into account the use of chance in the assigning process of experimental groups or subjects to treatments. The advantage of using randomization is that it avoids any bias in the practice of selection and guarantees that there is no inadvertent bias. In addition, it generates groups that are comparable and eradicates the source of bias within the assignment done to treatments (Laurel, 2003).
b) Replication
It is important to make use of replication in a study because it facilitates the use of enough or sufficient experimental units in every group in order that any differences in the treatment effects can be distinguished from the variances that exist between the groups (Laurel, 2003).
c) control groups
A control group is delineated as the group within a study that does undergo the treatment being examined by the researchers or experimenters and is thereafter employed as a standard to measure the manner in which other subjects being tested perform or respond.
5. Explain the meaning of the term block design. When is such a design appropriate?
In a block design randomized study, the person conducting the experiment categorized the subjects into smaller subgroups referred to as block, in the sense that the variability existing within these blocks is lesser as compared to the variability existing between these blocks. Subsequently, the subjects within every within every block or subgroup are randomly assigned to the different treatment conditions in the study (Laurel, 2003).
6. Explain what is meant by a matched pairs design. Give an example.
A matched pairs design is a distinctive case of a randomized block design. This design can solely be used at an instance where the experiment has solely two treatment conditions. Therefore, the participating subjects in the studies are categorized into pairs, on the basis of a kind of blocking variable. Subsequently, in every pair categorized, randomization is used to assign subjects to dissimilar treatments (Mitchell and Jolley, 2012).
7. Explain the meaning of the following terms:
a) Placebo
Within an experiment, the participants involve usually react in dissimilar manners subsequent to receiving treatments. A placebo is a treatment that is neutral that does not have any actual effect on the subject, despite the fact that the subject shows a positive response to such a treatment (Heath, 2002).
b) double-blind
In an experiment, double-blind takes into account a testing practice whose design is purposed to eradicate biased outcomes, in which the individuality of those obtaining a test treatment is hidden not just on the subjects but also the experiment administrators up to the point when the study is finished (Mitchell and Jolley, 2012).
c) Confounding
An experiment design that is confounding is one in which a number of treatment effects are appraised by the similar linear mixture or grouping of the experimental observations akin to a number of blocking effects. In this regard, there is confounding in both the blocking effect and the treatment effect (Mitchell and Jolley, 2012).
d) response
A designed experiment is deemed to be a controlled research study whereby one or more treatments are employed to subjects. The one conducting the experiment thereafter makes an observation of the effect of changing and fluctuating these treatments on a response variable. In other words, a response variable is considered to be the variable of interest within an experiment (Mitchell and Jolley, 2012).
Part 2
1. People who provide an answer to telephone surveys typically listen to an interviewer’s introductory remarks, and then decide if they want to continue or hang up. Suppose that a study made telephone calls to randomly selected households and asked about opinions for an upcoming election. To investigate responses, the interviewer sometimes gave her name, in others she identified the university she enrolled in, and in still others she identified both herself and her university. The study recorded the percentage of each group of interviews that was fully completed.
a) Is this telephone survey an example of an observational study or an experiment?
The key difference between an experiment and an observational study is that in the latter, the experimenter does not have any control over the variables whereas an experiment is an approach encompassing the application of treatments to a group and having records of such effects. In this particular case, the telephone survey is an example of an observational study. This is because in an experiment there is a control group and a treatment, which are lacking in this particular survey.
b) What are the explanatory and response variables for this survey?
The response variable in a study is the emphasis of a question whereas the explanatory variable is one that elucidates changes in the response variable. In this particular survey, the explanatory variable is the number of households as this can be manipulated by the researcher whereas the response variable is the opinions on the upcoming elections.
2. A clinical trial was created to study sinus infections. The experimental design is a completely randomized experiment that assigned 250 patients at random among four treatments as given below:
Antibiotic Pill Placebo Pill
Steroid Spray 53 64
Placebo spray 65 68
a) Provide an outline for this experimental design
I. Subjects
The study includes 250 subjects included to study sinus inflections
II. Assignment of subjects
Subjects are assigned randomly using four treatments with the combination of the antibiotic pill and steroid spray, placebo pill and the placebo spray, antibiotic pill and the placebo spray, and placebo pill and steroid spray.
III. Variables
The independent variable is the treatment design and the dependent variable is the number of subjects
IV. Operational definition
There is random assignment to the treatment groups
V. Techniques
There is the use of a placebo to control the placebo effect
b) Describe how you would label the 250 subjects
Randomized-controlled subjects
c) Explain briefly how you would perform the random assignment of patients to treatments. Assign the first five patients who will receive the first treatment.
I would form a block in order to assign sample numbers in an equal manner to every group and assign the block. Since the block size is two, then it is possible to make the first five possible sequences to be Antibiotic pill+ steroid spray, antibiotic pill + placebo spray, placebo pill + steroid spray, placebo pill + placebo spray, and then again Antibiotic pill+ steroid spray.
3. A business PhD student who lived on a naval base, looked at prices of items at the Naval Commissary and prices of the same items at Kmart. Assume that these are a random sample of items 10 items. The student wanted to test the hypothesis that the mean price at the naval base was less than the mean price at Kmart.
Items Kmart Navy
Tide 19.99 18.13
Clorox 4.29 2.57
Gain 5.39 5.13
Swiffer 7.99 5.99
Pledge 2.99 2.48
Secret 3.99 2.94
Spam 2.49 1.30
Pepsi 1.49 1.38
Chunky 1.79 1.29
Water 1.50 0.99
a) Provide an outline for the design of this experiment
I. Subjects
The study includes 10 items
II. Assignment of subjects
Items are randomly selected from K-Mart and Navy
III. Variables
The independent variable is items selected
IV. Operational definition
There is random assignment to the treatment groups
b) Identify the response variable
The response variable is the prices
c) Use statistics learned from earlier activities to investigate the results
The main statistics used to investigate the results in correlation. In accordance to the analysis, the two data sets have a correlation of 0.9945. However, the mean price of K-mart is $5.19 whereas that of Navy is $4.22
d) Determine if you can reach a definitive conclusion.
Therefore, the definitive conclusion is that K-Mart’s average price is greater compared to that of Navy’s.
4. Independent random samples from four different brands of recently produced batteries were placed on a life test. The following lifetimes (in minutes) were recorded.
Brand A Brand B Brand C Brand D
110 118 108 117
113 116 107 112
108 112 112 115
115 117 108 119
112 119 109 118
a) Provide an outline for the design of this experiment
I. Subjects
The study includes 4 different brands of produced batteries
II. Assignment of subjects
Subjects are assigned randomly assigned to the four different brands
III. Variables
The independent variable is the lifetime in minutes and the dependent variable is the number of subjects
IV. Operational definition
There is random assignment to the treatment groups
b) Identify the response variable
The response variable is lifetime in minutes
c) Create a graph for this data. Can you see a difference in the average lifetime of these four brands?



It is possible to perceive a difference in the average lifetime of these four brands. Brand B appears to have the longest average lifetime followed by Brand D. This is followed by Brand A and lastly by Brand C.
d) Is there any evidence that the sample data contradicts the claim that there are no differences in the average lifetime of these four brands?
There is proof because the Brand B has a considerably different average lifetime compared to Brand C.
Part 3
a) Write an essay that describes the article’s main points
Smoking continues to be the prevalent avoidable cause of untimely death in the US. In particular, 70 percent of smokers in the nation proclaim that they aspire to quit smoking. However, solely 2 percent to 3 percent of the smokers accomplish this aspiration. Preceding research studies have indicated that offering smokers with financial incentives to cease smoking has a positive impact by increasing enrollment in smoking-cessation programs and short-term rates of cessation. However, these studies not indicated significant rises in longstanding rates of cessation. The research paper, “A Randomized, Controlled Trial of Financial Incentives for Smoking Cessation” by Volpp et al. (2009) conducted a study to test the efficacy of a financial incentive of a maximum amount of $750 in enhancing longstanding rates of smoking cessation.
The methodology of the study encompassed randomized control study of 878 employees in a global company in the US. The control group includes of 442 employees who simply received information regarding smoking-cessation programs whereas 436 employees were included in the experimental group and included the provision of information regarding smoking-cessation plus financial incentives. The outcomes of the study demonstrated that the incentive group has a considerably greater rates of smoking cessation in comparison to the group that solely received information in a period of 9 months or a year subsequent to being enrolled. The inference of the study indicated that financial incentives for smoking cessation has a significantly increasing impact on the rates of smoking cessation. This shows that financial incentives cause the employees of the large company to be inclined to cease smoking (Volpp et al., 2009).
b) Describe the experimental design used. Explain the response, the treatments, where randomization is used, the structure of the experiment; and then provide the results of the experiment
The experimental design of a study alludes to the manner in which participants are apportioned to the various conditions in an experiment. In this case, there were two conditions in the experiment. One condition was the provision of information related with smoking-cessation only. The other condition was the provision of information related with smoking-cessation together with financial incentives of up to $750. As aforementioned, the response variable is deemed to be the variable of interest in the study. In this case, the response variable is smoking-cessation amongst the subjects. Treatment in this case study is the mixture of different values of the independent variable. In this case, treatment is the provision of information on smoking cessation and the amalgamation of such information together with financial incentives. The financial incentives included $100 for finishing a smoking-cessation program. $250 was granted for cessation of smoking with a period of 6 months subsequent to study enrollment and lastly $400 was for abstaining from smoking for an extra 6 months period after the initial cessation of smoking (Volpp et al., 2009). Randomization was used in the experiment to apportion subjects into the incentive group and the only receiving-information group. Randomization was performed in permuted blocks of four and was stratified based on income, site of work and also heavy and non-heavy smoking.
The results of the study indicated that subjects within the incentive group had considerably higher rates of smoking cessation in comparison to the subjects within the information-only group after a period of 9 to 12 months after being enrolled into the study and also 15 months or 18 months after being enrolled to the study. In addition, subjects within the incentive group had considerably higher rates of enrolling into a smoking-cessation program, completing such a program, and ceasing to smoke within the first six month-period after being enrolled into the study (Volpp et al., 2009).




References
Heath, D. (2002). An introduction to experimental design and statistics for biology. CRC Press.
Laurel, B. (2003). Design research: methods and perspectives. Massachusetts: MIT press.
Mitchell, M. L., & Jolley, J. M. (2012). Research design explained. New York: Cengage Learning.
Volpp, K. G., Troxel, A. B., Pauly, M. V., Glick, H. A., Puig, A., Asch, D. A., ... & Corbett, E. (2009). A randomized, controlled trial of financial incentives for smoking cessation. New England Journal of Medicine, 360(7), 699-709.