NURS 8200 Week 7: Quantitative Methods: Linear Regression



NURS 8200 Week 7: Quantitative Methods: Linear Regression
NURS 8200 Week 7: Quantitative Methods: Linear Regression
Quantitative Methods: Linear Regression

Suppose you were involved in a research study examining the effect of drinking soda on a child’s weight. You performed a study over the course of several months on a sample of fifth graders, allowing one group to drink two cans of soda per day, another group one can per day, and the control group no soda at all. After you gathered your data, you would need to analyze the results for each of the three groups to determine whether to accept either the null or alternative hypothesis in your study. A useful method of analysis for this particular study is known as linear regression.

As you examine linear regression, you may notice some limitations or shortcomings of this method of statistical analysis. Linear regression assumes that the relationships between variables are linear and that the variables themselves are continuous in nature. Linear regression is therefore not useful to examine variables that are binary or dichotomous (i.e., variables that only have two possible outcomes, such as gender).

This week continues your exploration of correlation and relationships between variables in quantitative research studies, focusing on the concepts of linear regression. This week also provides an overview of the concepts and applications of logistic regression, especially as it pertains to the health care field and evidence-based practice. Last week you examined the uses and methods of simple linear regression as a basis for this type of analysis. This week, you expand on those basic concepts and explore multiple regression, which can be used to show relationships between more than two variables.

ORDER NOW FOR ORIGINAL PAPER
Learning Objectives

Students will:

Analyze, interpret, and report results of a linear regression analysis
Analyze, interpret, and report results of a logistic regression analysis
Assess the application of logistic regression in nursing research and practice
Photo Credit: [blackred]/[iStock / Getty Images Plus]/Getty Images

Learning Resources

Note: To access this week’s required library resources, please click on the link to the Course Readings List, found in the Course Materials section of your Syllabus.

Quantitative Methods: Linear Regression Required Media
“Multiple Regression”

Used by permission from SPSSVideoTutor.com A division of ConsumerRaters LLC., 1121 S Military Trail, 314, Deerfield Beach, FL 33442, USA

Note: The approximate length of this media piece is 5 minutes.

Accessible player –Downloads– Download Video w/CC Download Transcript

“Logistic Regression”

Used by permission from SPSSVideoTutor.com A division of ConsumerRaters LLC., 1121 S Military Trail, 314, Deerfield Beach, FL 33442, USA

Note: The approximate length of this media piece is 15 minutes.

Accessible player –Downloads– Download Video w/CC Download Transcript

Quantitative Methods: Linear Regression Required Readings
Gray, J.R., Grove, S.K., & Sutherland, S. (2017). Burns and Grove’s the practice of nursing research: Appraisal, synthesis, and generation of evidence (8th ed.). St. Louis, MO: Saunders Elsevier.

Chapter 24, “Using Statistics to Predict”. This chapter asserts that predictive analyses are based on probability theory instead of decision theory. It also analyzes how variation plays a critical role in simple linear regression and multiple regression.
Statistics and Data Analysis for Nursing Research

Chapter 9, “Correlation and Simple Regression” (pp. 208–222). This section of Chapter 9 discusses the simple regression equation and outlines major components of regression, including errors of prediction, residuals, OLS regression, and ordinary least-square regression.
Chapter 10, “Multiple Regression”. Chapter 10 focuses on multiple regression as a statistical procedure and explains multivariate statistics and their relationship to multiple regression concepts, equations, and tests.
Chapter 12, “Logistic Regression”. This chapter provides an overview of logistic regression, which is a form of statistical analysis frequently used in nursing research.
Hoerster, K. D., Mayer, J. A., Gabbard, S., Kronick, R. G., Roesch, S. C., Malcarne, V. L., & Zuniga, M. L. (2011). Impact of individual-, environmental-, and policy-level factors on health care utilization among US farmworkers. American Journal of Public Health, 101(4), 685–692. doi:10.2105/AJPH.2009.190892

Note: You will access this article from the Walden Library databases.. This article discusses the results of a study of how many U.S. farmworkers accessed U.S. health care. The study considered this question on several levels—individual, environmental, and policy—and used logistic regression to analyze the multivariate data gathered.

Tritica-Majnaric, L., Zekic-Susac, M., Sarlija, N., & Vitale, B. (2010). Prediction of influenza vaccination outcome by neural networks and logistic regression. Journal of Biomedical Informatics, 43(5), 774–781. doi:10.1016/j.jbi.2010.04.011

Note: You will access this article from the Walden Library databases. This article describes the methods and results of a neural network study on the effectiveness of the influenza vaccine using historical data in three neural network algorithms. The article also provides a discussion of logistic regression in comparison to the neural network algorithms used.

Xiao, Y., Griffin, M. P., Lake, D. E., & Moorman, J. R. (2010). Nearest-neighbor and logistic regression analyses of clinical and heart rate characteristics in the early diagnosis of neonatal sepsis. Medical Decision Making, 30(2), 258–266. doi:10.1177/0272989X09337791

Note: You will access this article from the Walden Library databases. This article outlines the procedures and findings of a study on the use of two methods of neonatal sepsis diagnosis: nearest-neighbor analysis and logistic regression analysis. The results indicated that each method generates unique information useful to diagnosis, and therefore both methods should be used simultaneously for improved accuracy of diagnoses.

Quantitative Methods: Linear Regression Optional Resources
Walden University. (n.d.). Linear regression. Retrieved August 1, 2011, from http://streaming.waldenu.edu/hdp/researchtutorials/educ8106_player/educ8106_linear_regression.html

NURS 8200 Week 7 Discussion 1: Peer Support: Linear Regression
Using this Discussion, post questions you have about using SPSS and collaborate with your colleagues to complete Assignment 5, assigned and due this week. The questions you post will not only give you the opportunity to address any problems you encounter and assist your colleagues, but will also give your Instructor an idea of the challenges and successes you and your classmates are experiencing. This will allow your Instructor to identify overall areas in which there is a lack of comprehension and areas of mastery and complete understanding, which will be useful in better explaining SPSS in future online courses.

Note: You do not earn any points for participating in this Discussion. It is not required that you participate; however, it is an opportunity for you to connect with your colleagues to discuss the statistical exercises.

Post your responses to the Discussion based on the course requirements.

Your Discussion postings should be written in standard edited English and follow APA guidelines as closely as possible given the constraints of the online platform. Be sure to support your work with specific citations from this week’s Learning Resources and additional scholarly sources as appropriate. Refer to the Essential Guide to APA Style for Walden Students to ensure your in-text citations and reference list are correct. Initial postings must be 250–350 words (not including references).

Assignment 1: Article Critique [Major Assessment 4]
Continue to work on your article critique, assigned in Week 2 and due in Week 9. Continue evaluating your selected research article and developing the required sections of your paper. Your evaluation should be nearing its final stages.

By Day 7 of Week 9

You are not required to submit this assignment this week. Your article critique is due by Day 7 of Week 9.

Assignment 1: Article Critique [Major Assessment 4] SAMPLE
Tritica-Majnaric, L., Zekic-Susac, M., Sarlija, N., & Vitale, B. (2010). Prediction of influenza vaccination outcome by neural networks and logistic regression. Journal of Biomedical Informatics, 43(5), 774–781. doi:10.1016/j.jbi.2010.04.011

What are the goals and purposes of the research study the article describes?

The goals of the study is to design model to enable successful prediction of the outcome of influenza vaccination based on real historical medical data. To compare a non-linear neural network approach and a logistics regression one.

How is logistic regression used in the study? What are the results of its use?

Logistic regression was used as a comparison model to a neural network model in estimating the risk of reaction to influenza vaccine and to extract variables which are found to be important in risk prediction. The use of logistics regression for this study was due to the multivariate data involving a dichotomous response.

“Artificial neural networks are a computer-based method which can incorporate non-linear effects and interactions between multiple variables in a valid probability mode” (Tritica-Majnaric, 2010, pg.776). Three neural network algorithms were tested to include: multilayer perceptron (MLP), radial-basis function network (RBFN) and probabilistic network (PNN).

What other quantitative and statistical methods could be used to address the research issue discussed in the article?

Well, I think that this study needs to be repeated in multiple populations in multiple geographical areas. I think correlation statistics could aid in identifying what the best preventive model is best across the world and allow for variances in different areas. We don’t have to all use the same model, however we should all compare our probability evaluations to ensure best practices are being evaluated.

What are the strengths and weaknesses of the study?

The strength of the study was the implementation of a 10-fold cross-validation which was due to the weakness of sample size and recognition of potential bias. The weakness of the study was the sample size, age variance, and subjective 26 input variables which could be bias by the voice of the preprocessing method.

How could the weaknesses of the study be remedied?

The weakness of the study could be remedied with future research focused on other preprocessing methods in the modeling and the use of more datasets. Also, the sample size could be larger and more geographical data besides just one area. In this study it was Croatia

How could findings from this study contribute to evidence-based practice, the nursing profession, or society?

The findings of this study could initiate more solidified evidence of predictive models for the influenza vaccine. Correlation studies could be done with other populations and age groups to identify the best predictive model and algorithm for preventative influenza vaccines and measures. Preventing influenza is extremely important to vulnerable populations in this country due to the increase in mortality rate if in infected. It is important to the nursing profession not only of better patient outcomes but also for protection of our fellow nurses and medical professionals. Better preventative vaccines and measures would aide in workplace efficiency and complications with the medical professionals being infected with influenza.


Scroll to Top