Module 8: Logistic Regression

profileKCplul76
DiscussionMODULE8.docx

Instructions

For this assignment, you will submit a document (PDF, .DOCX or .DOC formats are acceptable) containing your work in JASP, as outlined below. Prior to completing this assignment, watch the following video for some additional guidance and insights:

JASP Statistics. (2018, February 11). How to perform a logistic regression analysis in jasp [Video file]. Retrieved from https://www.youtube.com/watch?time_continue=2&v=bUgpJeeReBY (7:05mm)

Instructions:

· Instruction Open JASP. https://jasp-stats.org

· Click on the File tab at the top, then “Data Library” and “Regression.”

· Click on the “Titanic” JASP file.

· Navigate back to “Titanic” in the Data Library and open the dataset in a second window. 

· Read the “Titanic” JASP File and work through the examples in the dataset window. This working through the examples does not need to be included in your submitted assignment.

Load the “SalariesRC” dataset into JASP. SalariesRC.csv

· Run a multiple regression with salary as your outcome and publish, yrs.service, and Sex as predictor variables. Report and interpret your results, including investigating the difference each variable made to the fit of your model and reporting your findings.

· Run a logistic regression to see whether faculty tenure status can be predicted by the number of hours spent teaching, gender, and their interaction. First, carry out a hierarchical regression, with the first model including only HRSTEACH as a predictor, and then in model 2, add in SEX.

· Assess the overall fit of each model, interpret and report on your findings. Include a comparison of models 1 and 2. Report any difference that exists and whether or not it is statistically significant.

· Continue on to the next steps with the model that fits best.

· Consider the model’s coefficients. Report and interpret each coefficient’s b-value and test statistic.

· For each coefficient, calculate, report, and interpret the odds ratios and their confidence intervals.

· Perform case wise diagnostics for the model (predicted probabilities and residuals) and report your results.

· Test assumptions model assumptions (multicollinearity and linearity of the logit) and report your results.

Instructions:

· Title pages, citations, and the like are not necessary for this assignment given the nature of the tasks at hand.

· A title page and reference list are not required. Please identify the title of the assignment and your name on the top-left of the first page.

· References are not required, but if used, please use the most current APA format.

· An abstract is not required.

· Please refer to the rubric associated with this assignment for detailed guidance about expectations and grading.