Week 4 Essay

.docx

School

Liberty University *

*We aren’t endorsed by this school

Course

EDCO 735

Subject

Medicine

Date

Apr 29, 2024

Type

docx

Pages

5

Uploaded by Wjdn2002 on coursehero.com

MODULE 4: DATA SCREENING BASICS 1 Module 4: Data Screening Basics Tanya Miller School of Doctor of Education: Community Care and Counseling, Liberty University EDCO 735: Teaching and Learning Dr. Frederick Volk 13 April 2024 Author Note Tanya Miller I have no known conflicts to disclose. Correspondence concerning this article should be addressed to Tanya Miller. Email: tanya.miller74@yahoo.com
MODULE 4: DATA SCREENING BASICS 2 Module 4: Data Screening Basics Prompt 1: Given a highly non-normal empirical frequency distribution, does it make sense for a researcher to find the percentile rank using the z score and the standard typical distribution table? Why or why not? Describe two distributions (other than average) that a researcher might encounter in data and when. When considering whether to find the percentile rank using the z score and the standard typical distribution table, it stands to reason that a researcher would not do so because a non- normal empirical frequency distribution may generate outliers. Outliers generate inadequate statistics, resulting in sporadic dissemination, distorted graphs, and biased findings (Warner, 2021). Not every distribution will be that perfect bell curve, and some will have extreme outliers – for example, the drinking of tea or coffee. There may be those who do not enjoy it and thus do not drink it, and then there will be those who drink only tea, coffee, or both. There could be many outliers when determining the drinking habits of these caffeinated drinks (Sainani, 2012. The best way to employ non-normal numerical methods is by realizing several extreme values. While normal distribution is a regularly employed and recognized approach for determining statistical distribution, others like lognormal, exponential, binomial, and multinomial exist. The ones that make the most sense for the type of work this researcher does are binomial and multinomial distribution. The binomial distribution is the comparison of two discrete outcomes. The probability distribution is set to interludes of (0,1), illustrated by two definite shape factors assigned as alpha and beta. It appears to be an exponent of random variables manipulating the distribution shape. This type of distribution is utilized when there are two possible outcomes, and researchers need to determine the results from two individual test samples. If a researcher is attempting to
MODULE 4: DATA SCREENING BASICS 3 determine if sexual assault prevention training in the military will reduce sexual assault, they may utilize this distribution method as there are two outcomes they are attempting to obtain. The multinomial distribution is aware that research could yield more than two outcomes. This sampling looks at multiple combinations of items, recognizing that possible outcomes can occur (National institute of standards and technology, n.d.). Using this method and addressing sexual assault in the military may determine that prevention is helping to reduce sexual assault but not sexual harassment. It may also recognize that while prevention has reduced sexual assault, it has seen an increase in reports made regarding sexual assaults and not harassment. Prompt 2: How might the absence of data screening affect a researcher's data quality, their interpretations of the data, and their interpretation of their study's findings? What quantitative rule may be used to determine univariate outliers, and are there situations in which deleting a case/participant may be justified? Explain. One of the most critical actions in validating research data is data screening. According to Warner (2021), data screening is crucial for several different reasons: 1) to correct mistakes, 2) to familiarize researchers with data, 3) to govern where hypotheses and expectations are fulfilled, 4) to right supposition violations, and 5) identification and correction of skews, values, and outliers. When a researcher fails to show data screening, this can damage the value of the research. Warner argued that outliers could bias the expected confidence interval, standard errors, samplings, parameters, and investigations, leaving statistical data not as healthy as to stand against outliers. Univariate data are outliers with data points that have set variables far away, either to the right or left of the mean; these data points are not the norm. There are five reasons that univariate data can occur: input of data wrong, confusing questioning, thus giving skewed answers,
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help