STUDY DESIGN 1
GROUP DISCUSSION BOARD FORUMS 2–7 GRADING
RUBRIC
Criteria Points Possible Points
Earned
Forum
0 to 50
•All questions associated with Part 1 are divided
amongst the group members and all questions are
answered fully.
•Each student responds to at least 1 question for
Part 1.
•Each student submits at least 2 substantive
replies for the Part 2 collaborative effort and
demonstrates engagement with the material. Each
reply includes at least 1 strength and 1 weakness
of the group member’s thread.
•Each thread is 500–750 words, and the replies (at
least 2) are 450–600 words each.
•Posts are submitted by the stated submission
deadlines.
•Each thread and reply references at least 2 peer-
reviewed sources and 1 biblical integration.
•All sources are cited in current APA format.
•Proper spelling and grammar are used.
Sentences are complete, clear, and
concise.
Total 50
Instructor’s Comments: Excellent work, I recall your email and no penalty was assessed on
the late posting.
Compensation and Response Rate (9.9)
STUDY DESIGN 2
John W Bayse
Liberty University
BUSI600 – D14
STUDY DESIGN 3
Compensation and Response Rate
I have been asked to develop and experiment for a study of the effect that compensation
has on the response rates secured from personal interview subjects. This study involves 300
people who will be assigned to one of the following conditions: (1) no compensation, (2) $1
compensation, and (3) $3 compensation. A number of sensitive issues will be explored
concerning various social problems, and the 300 people will be drawn from the adult population
(Cooper and Schindler, 2014, p. 212). This paper describes the design.
According to Lewsey (2004) “randomized designs are popular in cluster randomized
trials … because they increase the chance of the intervention groups being well balanced” (p.
897). There are several types of randomized designs. The simplest approach for a randomized
design is Completely Randomized where the 300 people for the study are randomly split into
three groups of 100, represented by R. The design (Figure 1) can be shown as:
R X1O1
R X2O2
R X3O3
Figure 1: Completely Randomized design. X1, X2,, and X3 represent the three
compensation models and O1, O2,, and O3 represents the response rate based on the
corresponding compensation.
A more insightful approach would be the Randomize Block design where an extraneous
factor influences the outcome. There are also options where multiple extraneous factors have an
influence. Although more complex, it would be possible to compare how compensation along
STUDY DESIGN 4
with education level affects the response rates. R is the randomly selected members of each
group based on education. The design (Figure 2) can be shown as:
Active Factor
Compensation
Blocking Factor (Education)
Grade School High School College
$0 R X1X1X1
$1 R X2X2X2
$3 R X3X3X3
Figure 2: Randomize Block Design. X1, X2,, and X3 represent the three compensation
models and O
1
, O
2,
, and O
3
are assumed and represent the response rate based on the
corresponding compensation.
Addelman (1969) emphasizes [i]t is important that the experimental design be presented along
with the mathematical model so that one can see whether or not the model reflects the actual
procedure for performing the experiment” (p. 35). Because it is based on mathematical models,
it is easier to show the statistical relevance of the outcome. That makes this option one of the
better choices for a design.
Often taking on tasks that have not been done before can be daunting. Attempting to
pick he correct parameters for a study or the best method can be difficult. The wrong model can
skew the results and lead to useless data. So how to boldly approach any study with the belief
that the correct framework in place? Philippians 4:13 (New King James Version) says “I can do
all things through Christ who strengthens me.” As Christians, Christ never leaves us to walk
alone, but provides strength to handle the challenges that lay ahead. So we should not fear new
challenges. We should embrace them knowing that we take the journey with Jesus by our side.
STUDY DESIGN 5
References
Addelman, S. (1969). The generalized randomized block design. The American Statistician,
23(4), 35-36.
Cooper, D. R., & Schindler, P. S. (2014). Business research methods (12th ed.). New York,
NY: McGraw-Hill/Irwin.
Lewsey, J. D. (2004). Comparing completely and stratified randomized designs in cluster
randomized trials when the stratifying factor is cluster size: A simulation study. Statistics
in Medicine, 23(6), 897-905. doi:10.1002/sim.1665
DB5 Responses
John W Bayse
Liberty University
BUSI600 – D14
STUDY DESIGN 6
DB5 Response – Brandon Blankenship
Brandon defined how a study should be done to determine the causal relationship between four different
independent variables - temperature, humidity, artisan expertise level, and production supervisors - and a dependent
variable, percentage of defective glass shells being manufactured. The first three potential independent variables
have data available every day for the past year while the final one only had data for 242 of the 365 days. The paper
was well written and thought with sufficient examples to help the reader understand his points.
Strength
The temptation for a research project where there is already theoretical data available for cause is to jump
right into comparing the potential independent variables and their role in the dependent variable. But the author
took a step back to understand the bigger picture instead of making assumption just because there are theoretical
inputs to explain the issue. By using steps learned earlier in the course he outlined how to frame the problem and
determine what the appropriate outcome needed to be therefore setting himself up to pick the correct method for
gathering and comparing data to reach the appropriate conclusion. The significance of understanding the causal
relationship cannot be understated. As Johnston, Maguire, and McGinnity (2008) point out, “[i]f causal
relationships are unknown then it is difficult to differentiate whether the observed process effects coincide with
planned project changes or are simply a natural process variation” (p. 6251-6252).
Weakness
The author of the original post took the approach of comparing data between the four potential
independent variables only on the days it was available from all four variables. This is a wise decision since
picking dates where the supervisor information is not available could hide a potential relationships if there is not a
clear delineation of the problem based upon just the three theoretical causes. More detail could have been provided
how the study would be conducted relative to the four potential causes. Would they be compared together, in pairs,
or in larger groups? How would a variable be determined to be in or out of the relationship and is there a way to
STUDY DESIGN 7
determine what percentage of a factor an included variable is? The approach and definitions are great, but leave a
gap in determining the actual causal relationship. The material presented was an excellent approach to solving the
problem, but by expanding the explanation of the approach to include how the determination will be made in the
end, it will leave no doubt to the reader how the experiment will be carried out. In the end, ambiguity in research
can affect the ability to make the appropriate decision. Borgonovo and Marinacci (2015) point out “decision
makers are often uncertain about one or more probabilities of interest … this uncertainty ambiguity might affect
decision makers’ preferences” (p. 823).
Biblical Integration
Luke 14:28-30 (New King James Version) says “[f]or which of you, intending to build a tower, does not
sit down first and count the cost, whether he has enough to finish it – lest, after he has laid the foundation, and is
not able to finish, all who see it begin to mock him, saying, ‘This man began to build and was not able to finish’.”
Brandon did a great job of laying out the plan from the beginning, but the lack of detail at the end of the plan
leaves some ambiguity that could prevent the experiment from being finished as intended.
STUDY DESIGN 8
References
Borgonovo, E., & Marinacci, M. (2015). Decision analysis under ambiguity. European Journal of Operational
Research, 244(3), 823. doi:10.1016/j.ejor.2015.02.001
Johnston, A. B., Maguire, L. P., & McGinnity, T. M. (2008). Disentangling causal relationships of a manufacturing
process using genetic algorithms and six-sigma techniques. International Journal of Production Research,
46(22), 6251-6268. doi:10.1080/00207540701427029
DB5 Responses
John W Bayse
Liberty University
BUSI600 – D14
STUDY DESIGN 9
DB5 Response – Kwabena Antwi Nimarko
Kwabena addressed the question of differences in essential characteristics of a true experiment compared
to other research designs. The paper was well written and had detailed definitions of each of the types of true
experiments and research designs.
Strength
An area of experimentation that is rarely discussed is the ethical nature of recording data from something
that is not true. People’s emotions can be fragile and making them believe an untruth, even if it is for a greater
good, will always rile those who believe data gathered from a falsehood is unethical. But how can researcher’s
gather data if they do not create some controlled environment to study the effects of a theoretical situation. And if
the researcher tells the subject what they are doing, or even tells them some people will be involved in theoretical
situations, then how can the results be considered to be valid when the subjects reactions will be based on the
knowledge that what they are dealing with may be false. True responses are only gathered when the subject
believes the scenario is real. I’m not proposing this type of research should not be done, just acknowledging it is a
very valid concern for any experiment that looks to determine an outcome from something where real time data
may not be available. Spicker (2011) points out [c]overt research “is often muddled with deception, and
condemned as intrinsically unethical. The basis of that condemnation is a legitimate concern with the rights of
research subjects” (p. 118). The question will always remain whether those rights outweigh academic research.
STUDY DESIGN 10
Weakness
While the definitions were spot on, listing the definitions of the types of true experiments versus research
design did not take the next step of addressing the difference between the two types of research. Cooper and
Schindler (2014) put it simply “[t]he major deficiency of the preexperimental designs is that they fail to provide
comparison groups that are truly equivalent” (p. 206). Research and surveys that just observe can provide valuable
data, but without an understanding of the data pool used or the evaluation, it is difficult to determine the ability to
extrapolate the data to a larger segment. According to Ranstam (2008), a “statistically important finding is simply
reliable, and cannot be explained by random sampling” (p. 220). That is why true experiments work randomly
assigned groups from appropriate segments to ensure a statistically significant result.
Biblical Integration
Joshua 1:8 (New Living Translation) says “[s]tudy this Book of Instruction continually. Meditate on it
day and night so you will be sure to obey everything written in it. Only then will you prosper and succeed in all
you do.” Especially in regard to the question of ethics in research, when the Christian spends time studying God’s
word, it becomes easier for them to determine how God would have them act when presented with a potentially
ethically questionable issue. And for most Christians, it then isn’t an issue of knowing what to do; it is just having
the courage to do it.
STUDY DESIGN 11
References
Cooper, D. R., & Schindler, P. S. (2014). Business research methods (12th ed.). New York, NY: McGraw-
Hill/Irwin.
Ranstam, J. (2008). Statistical significance. Acta Radiologica, 49(2), 220-221. doi:10.1080/02841850801924256
Spicker, P. (2011). Ethical covert research. Sociology, 45(1), 118-133. doi:10.1177/0038038510387195
