Liberty University Ethical Perspective on Deceptive Statistics Paper Discussion Board Forum: Even in secular areas of science, unethical behavior is frown

Liberty University Ethical Perspective on Deceptive Statistics Paper Discussion Board Forum:

Even in secular areas of science, unethical behavior is frowned upon and viewed in disgust. How much more then must Christians strive to not just be “ethical” but also to present findings or information in a way that will not mislead the reader? As 1 Peter 2:12 says “having your conduct honorable among the Gentiles, that when they speak against you as evildoers, they may, by your good works which they observe, glorify God in the day of visitation”(NKJV). Although the meaning of this verse and those that precede it were used in a somewhat different context, its words are still applicable here. When performing research and statistical analysis, you cannot let it be thought that you are unethical as that perception will harm not only your reputation but also the reputation of Christianity. Below are the instructions for the thread. Read this article in order to properly discuss this prompt. Special Note: the article is attached below

Thread – Discuss the following:

How could you make the ideas mentioned in the article more in line with the Christian worldview? Tie the article’s message to at least 1 Scripture in addition to other ideas you may have.
What do you think it says about a person who will let the pressure to publish cause him or her to be deceptive.
What safeguards will you implement to prevent yourself from being deceptive in statistical analysis and research?

Forum Post Instructions:

You are required to provide a thread in response to the provided prompt for each forum. Each thread must be at least 500 words, demonstrate course-related knowledge, integrate 1 biblical principle, and provide a total of 2 citations from any of the following sources: peer-reviewed journal articles, published textbooks, or publications directly associated with the content being discussed (requires prior approval from the instructor).

References:

Haff, G.G. & Triplett, N. T. (2016). Essentials of Strength Training and Conditioning. Champaign, IL: Human Kinetics

Powers, S., & Howley, E. (2018). Exercise physiology (10th ed.). New York: McGraw Hill.

Vincent, W. J., & Weir, J. P. (2012). Statistics in kinesiology. Champaign: Human Kinetics. 9. American Chemical Society. ACS ethical guidelines to publication of
chemical research. In: The ACS Style Guide. Washington, D.C; American Chemical Society; 1986:217-22.
10. Editorial consensus on authorship and other matters. Lancet.
1985;2:595.
11. INTERNATIONAL COMMITTEE OF MEDICAL JOURNAL EDITORS. Guide-
lines on authorship. BrMedJ. 1985;291:722.
12. HUTH EJ. Standards on authors* responsibilities [Editorial]. Ann Intem Med. 1985;103:797.
Science, Statistics, and Deception
JOHN C. BAILAR III, M.D., Ph.D.; Washington, D.C.
Some common scientific practices cannot quite be called
lying, though they are potentially, and sometimes
deliberately, deceptive. Some examples are the failure to
explain to readers all the weaknesses in data, statistical
testing oi post hoc hypotheses, fragmentary or selective
reporting of findings, and reporting as “negative” a study
that had insufficient chance of detecting an effect. The
first step toward controlling potential problems is a
redefinition of ethical standards to bar readily avoidable
as well as deliberate deception. Other remedies include
substantially greater restraint in the use of questionable
practices, full disclosure and justification each time these
practices are used, and greater skepticism by readers.
Pressures to publish tend to promote deception. A
broadened concept of ethical standards in science should
be reflected in training programs and in the structure of
scientific rewards, including a sharply reduced emphasis
on publication as an end in itself.
IN SCIENCE, lying is condemned, even by some of its few
practitioners. Deliberate or careless deception short of
lying, however, seems to be universally accepted and
sometimes even promoted as a part of the culture of science. I do not suggest that scientists as a group are careless, venal, or otherwise depraved: they may even be
above the human average in developing and adhering to
detailed, albeit tacit, standards of professional conduct.
Those who are clearly violators are drummed out of our
ranks, loudly and publicly. But what about less clearcut
deception?
My thesis is that our professional norms are incomplete and that several kinds of widely accepted practices
(Table 1) should also be widely recognized as potentially deceptive and harmful. Some of these practices also
have much value, but at times they are inappropriate and
improper and, to the extent that they are deceptive, unethical.
The scientific method is fundamentally concerned with
the processes of inference, generally from data that are
necessarily inaccurate to some degree, incomplete, drawn
from small samples, or not quite appropriate for a specific
task. Inference—that is, drawing conclusions or making
deductions from imperfect data—provides most of the
excitement and intellectual ferment of science. Scientific
rewards are probably more closely related to valid inferences established than to such related activities as imaginative hypotheses formulated, elegant experiments de-
• From the Office of Disease Prevention and Health Promotion, Department of
Health and Human Services; Washington, D.C.
Annals of Internal Medicine. 1986;104:259-260.
signed and conducted, or new methods developed. The
rewards for publishing a first-class inference can include
income, position and power, professional status, and the
respect of colleagues. Such rewards may sometimes count
for more than self-respect and the joy of discovery. We
must therefore be attentive to scientific norms and activities that may distort the processes of inference.
An example of a deceptive practice is the statistical
testing (such as the calculating ofp values) oi post hoc
hypotheses. It is widely recognized that f-tests, chi-square
tests, and other statistical tests provide a basis for probability statements only when the hypothesis is fully developed before the data are examined in any way. If even the
briefest glance at a study’s results moves the investigator
to consider a hypothesis not formulated before the study
was started, that glance destroys the probability value of
the evidence at hand. Certainly, careful and unstructured
review of data for unexpected clues is a critical part of
science. Such review can be an immensely fruitful source
of ideas for new, before-the-fact hypotheses that can be
tested in the correct way with other new or existing data,
and sometimes findings may be so striking that independent confirmation by a proper statistical test is superfluous. Statistical “tests” are also used sometimes in nonprobabihty ways as rough measures of the size of an
efiFect, rather than to test hypotheses. (An example is the
column ofp values that sometimes accompanies a table
comparing the pretreatment characteristics of patient
groups in a randomized clinical trial.) When either the
test itself or the reporting of the test is motivated by the
data, a probability statement such as “p < 0.05" is deceptive and hence damaging to inference. Other potential problems are the selective reporting of findings and the reporting of a single study in multiple fragments. These practices can obscure critical aspects of an investigation, so that readers will misjudge the evidential value of the data presented. Such reporting may be deceptive, whether deliberately or accidentally. On the other hand, these practices sometimes have positive value that should be preserved. For example, they can facilitate the tasks of both the investigator and the user when a demand for a monolithic analysis might seriously delay or frustrate the progress of both. "Negative" conclusions of low statistical power—that is, reporting that no eflFect was found when there was Httle chance of detecting the effect—can also distort inference, especially when investigators do not report on 259 Table 1. Some Practices That Distort Scientific inferences Failure to deal honestly with readers about non-random error (bias) Post hoc hypotheses Multiple comparisons and data dredging Inappropriate statistical tests and other statistical procedures Fragmentation of reports Low statistical power Suppressing, trimming, or "adjusting" data; or undisclosed repetition of "unsatisfactory" experiments Selective reporting of findings Statistical power. The concept of power is formally defined in terms of the random variability of results that is inherent in a specific combination of data structure, sample size, statistical models, and analytic method; but I believe that the concept should be substantially broadened to include the likelihood that a particular effect would be detected and reported if it were present to some specified degree. Such an analysis rarely accompanies "negative" findings, and readers may be left with an unjustified sense that an effect not demonstrated is an effect not present. Again, however, there are counterarguments: a report with low power may be better than no report (and no power), or meta-analysis (1) of several low-power reports may come to stronger conclusions than any one of them alone. Reporting negative studies of low power can create ethical problems, but those problems may be largely mitigated if the low power is accurately and clearly reported as weU. Too many scientists resist the objective reporting of this kind of weakness in their work, and pressure for "strong" results may be greatest during the formative years of graduate training and career entry. Thus we may be training new scientists in unethical methods. Despite the occasionally useful roles of these and other practices listed in Table 1, each can seriously distort the processes of inference and should therefore be an object of concern. Where the practices have legitimate applications, they should, of course, be used; but even then they should be fully and explicitly disclosed by the investigator, justified in some detail, and accepted with caution by readers. A combination of restraint in their overall use, limiting their use to clearly appropriate situations, providing full disclosure and justification, and maintaining the readers' skepticism will help to diminish the frequency and severity of ethical problems. Full disclosure here means more than a few words buried in the fine print of a Methods section; it means not just that the author send a message, but that the author also work to ensure that the message is received and correctly interpreted by readers. There are parallels here to the evolving requirements for informed consent by experimental subjects. Pressures to publish can be great and may account for many of the abuses suggested by Table 1. I fear that even the constrained use of these potentially damaging practices will leave attractive loopholes for an army of ambitious practitioners of science, each feeling great pressure to publish, who will rush in to explain why his or her situation is different, why full disclosure is inappropriate, and so forth. I am convinced that science, scientists, and society as a whole would benefit from substantially broader concepts, ultimately based on the need to protect the processes of inference, about ethical standards and violations in science. • Requests for reprints should be addressed to John C. Bailar III M D Ph.D.; 468 N Street, S.W.; Washington, DC 20024. Reference 1. LOUIS T A , FINEBERG H V , MOSTELLER F . Findings for public health from meta-analysis. Annu Rev Public Health. 1985;6:l-20. 260 February 1986 • Annals of Internal Medicine • Volume 104 • Number 2 Purchase answer to see full attachment