Levels of Measurement: Categorical vs. Continuous Data; Descriptive Statistics and Probability Theory Basics
What is the incidence of blood clots from COVID-19 in females over the age of 35?
The above question is an example of a research question. A research question consists of three parts and guides the methods and approaches in which you will study the question to find answers. The research question includes the question, the topic, and the population or variables. In the example provided above, the question examines the prevalence of blood clots from severe COVID-19 in a selected population. From this question, the variables can be assessed, considerations can be analyzed, and populations can be sampled in order to guide the research.
For this Discussion, you will analyze a selected work to identify and analyze the variables, comparisons, and sample sizes. You will explore the potential levels of measurement for your variables and the rationale for the labels, as well as consider the advantages and challenges that you might experience in the statistical analysis.
Reference: Gray, J. R., & Grove, S. K. (2020).
Burns and Grove’s the practice of nursing research: Appraisal, synthesis, and generation of evidence (9th ed.). Elsevier.
Resources
Be sure to review the Learning Resources before completing this activity.
Click the weekly resources link to access the resources.
Learning Resources
· Bullen, P. (n.d.).
How to choose a sample size (for the statistically challenged)Links to an external site.
. tools4dev.
· Centers for Disease Control and Prevention. (2024, March).
The NHSN standard infection ratio (SIR)Links to an external site.
.
· “Overview of the Standard Infection Ratio (SIR)” (pp. 4–5)
· Dang, D., Dearholt, S. L., Bissett, K., Ascenzi, J., & Whalen, M. (2021).
Johns Hopkins evidence-based practice for nurses and healthcare professionals: Model & guidelines (4th ed.). Sigma Theta Tau International Honor Society of Nursing.
· Chapter 6, “Evidence of Appraisal: Research” (pp. 147–157)
· Salkind, N., & Frey, B. (2019).
Statistics for people who (think they) hate statistics (7th ed.). SAGE Publications.
· Chapter 3, “Computing and Understanding Averages: Means to an End” (pp. 65–68)
· Chapter 5, “Creating Graphs: A Picture Really Is Worth a Thousand Words” (pp. 88–118)
· Chapter 8, “Hypotheticals and You: Testing Your Questions” (pp. 167–180)
· Chapter 9, “Probability and Why It Counts: Fun With a Bell-Shaped Curve” (pp. 181–200)
· Niedz, B. (2024).
Descriptive statistics [Video]. Walden University Canvas.
· Document:
Descriptive Statistics (PowerPoint presentation)
Download Descriptive Statistics (PowerPoint presentation)
Required Resources for Topic: Infections
· Beydoun, A. S., Koss, K., Nielsen, T., Holcomb, A. J., Pichardo, P., Purdy, N., Zebolsky, A. L., Heaton, C. M., McMullen, C. P., Yesensky, J. A., Moore, M. G., Goyal, N., Kohan, J., Sajisevi, M., Tan, K., Petrisor, D., Wax, M. K., Kejner, A. E., Hassan, Z., … Zenga, J. (2022).
Perioperative topical antisepsis and surgical site infection in patients undergoing upper aerodigestive tract reconstructionLinks to an external site..
JAMA Otolaryngology-Head & Neck Surgery, 148(6), 547–554.
· Sood, N., Lee, R. E., To, J. K., Cervellione, K. L., Smilios, M. D., Chun, H., & Ngai, I. M. (2022).
Decreased incidence of cesarean surgical site infection rate with hospital‐wide perioperative bundleLinks to an external site.. B
irth: Issues in Perinatal Care, 49(1), 141–146.
· Sauer, K. (2023).
Testing for the treatment of urinary tract infections in symptomatic adult patients residing in long-term care facility: An evidence-based quality improvement projectLinks to an external site.
(Publication No. 30569808) [Doctoral dissertation, Phoenix University]. ProQuest Dissertations and Theses Global.
To prepare:
· View the required media.
· It is recommended you complete the quiz prior to constructing your initial response.
By Day 3 of Week 3
Post a response including the following:
· Choose a research study, QI article, or EBP DNP project and interpret at least one continuous demographic variable and one categorical variable.
· Differentiate between comparisons made using descriptive statistics (e.g., the mean and standard deviation) and comparisons based on inferential statistics (e.g., a
t test).
· Compare and contrast the sample sizes used in the research study, the QI project, and the DNP project in terms of type 1 and type 2 errors.
· Explain the SIR rate, how it is developed, and how organizations use it.
· Using the same articles, pick one and differentiate between one descriptive and one inferential statistic used in any one of the three studies/projects.
By Day 6 of Week 3
Read a selection of your colleagues’ posts and
respond to
at least two of your colleagues on
two different days by expanding upon their reflections, making connections to your perceptions, and offering additional insights.
RESPOND TO THIS DISCUSSION
Patience
Statistical Analysis in Nursing Research
Statistical analysis forms the backbone of evidence-based nursing practice, transforming raw clinical data into meaningful insights that guide patient care improvements. Understanding the statistical frameworks becomes crucial when examining fall prevention strategies, quality improvement initiatives, and doctoral projects. This analysis explores key statistical elements across three research papers.
Continuous and Categorical Variables in Fall Prevention
In Tzeng and Yin’s (2017) multihospital study on fall prevention, we can identify key variable types that shape research outcomes. Length of stay stands out as a continuous demographic variable. Patients who fell with serious injuries stayed 6.3 days longer than non-fallers. This time measurement exists on a ratio scale, allowing precise day counting with actual zero points and equal intervals. Hospital unit type works as a categorical variable in this research. The study notes, “Each hospital inpatient specialty area had its own top 10 effective interventions” (Tzeng & Yin, 2017). This variable sorts patients into distinct groups without numeric relationships between categories.
Descriptive vs. Inferential Statistics
Descriptive statistics summarize data without drawing broader conclusions. In Bangura’s (2024) project, reporting “ten falls per month” at the Veterans Affairs facility compared to the national rate of “five falls per month” describes reality
without testing hypotheses or making predictions. Inferential statistics differ significantly because they allow researchers to conclude the sample data. When Tzeng and Yin (2017) determined that “operational costs for patients who fell with serious injury were $13,316 higher,” they likely used t-tests to confirm that this difference was not random chance. This statistical approach aims to validate patterns that can apply to broader populations.
Sample Sizes and Error Risk
The three studies show different sample approaches affecting error risks. Tzeng and Yin’s (2017) research included multiple hospitals, creating a larger sample that reduces both error types. With more data points, false positive findings (Type I errors) become less likely since random variations have less impact. The more significant sample also increases statistical power, reducing Type II error risk. Khoja and Moosa’s (2023) quality improvement article targeted specific hospital units for six months. This mid-sized sample balances practical constraints with statistical needs. The clinical project Bangura (2024) conducted monitored a single unit with 25 beds for six weeks. A sample of this size introduces two kinds of calculation errors. Inadequate data collection could demonstrate deceptive results (Type I) or overlook beneficial actions (Type II).
Understanding SIR Rates
The Standardized Infection Ratio evaluates observed infections against expected national rates using the mathematical formula SIR = Observed infections ÷ Predicted infections. Healthcare organizations create this measurement by implementing standardized infection surveillance methods and statistical risk adjustment models. Healthcare organizations use SIR to monitor infection prevention advances and determine improvement zones while fulfilling regulatory needs and measuring performance relative to national benchmarks. A SIR value below 1.0 reflects success in exceeding performance predictions, whereas an index above 1.0 signals that infection rates need organizational attention.
Statistics in Bangura’s DNP Project
The fall prevention project led by Bangura (2024) includes a statistical description and inference. At baseline, there were ten recorded patient falls monthly, which serves as pure descriptive data without making inferences. The conclusion about fall reduction through intentional rounding using the Morse fall tool depends on inferential statistical analysis. The study mandatorily needed pre-intervention data comparison to post-intervention data to check if needed pre-intervention data comparison to post-intervention data to check if changes exceeded statistical significance.
Conclusion
Nurses who understand statistical concepts can evaluate research findings while designing evidence-based interventions with assurance. The examined studies show that statistical methods provide solutions for different healthcare inquiries.
References
References
Bangura, F. (2024). Development and Evaluation of a Nurse Practitioner Intentional Rounding Strategy and Its Impact on Decreasing Falls in a Veterans Long-term Care Facility. Wilmington University (Delaware)
Khoja, A., & Moosa, L. (2023). Impact of tailored interventions for patient safety (TIPS) to reduce fall rates.
MEDSURG Nursing,
32(2), 89.
to an external site.
Tzeng, H. M., & Yin, C. Y. (2017). A multihospital survey on effective interventions to prevent hospital falls in adults.
Nursing Economics,
35(6), 304–313.
to an external site.
Reply
