401077 Introduction to Biostatistics, Autumn 2016
Assignment 2
Due Tuesday April 24, 2016
Please answer each question in the template document provided and submit via Turnitin by the due date. The marks allocated to each question are shown in the assignment. A total of 30 marks are available and this assignment is worth 30% of your overall grade.
Question 1. (5marks)Note: Students will get different answers as the data sets differ.
Using the assignment data set assigned to you and R Commander
Tabulate the relationship between blood group (variable name: bldgrp) and gender (variable name: sex). Type and label the table yourself rather than just present a R Commander screenshot. (1 mark)
Present the row or column percentages for the table in a).(1 mark)
Using the percentages in b) only is there any evidence of relationship between blood group and gender? Explain why or why not. (1 mark)
Would it be appropriate to run a Chi-square test of independence to test for relationship between blood group and gender? Explain why or why not. (2 marks)

Question 2 (5 marks) Note: Students will get different answers as the data sets differ.
Using the assignment data set assigned to you and R Commander address the following questions about the mean depression score for people in this population. The variable for analysis is depress. Assume your data set represents a random sample from the population of interest.
Provide a point estimate of the population mean. (1 mark)
Calculate the 95% confidence interval for the population mean depression score in this population. (1 mark)
Carefully write in words, what this confidence interval is telling us (2 marks).
Does this confidence interval suggest that the mean depression score in this population is statistically significantly greater than 28? Explain why or why not. (1 mark)

Questions 3 and 4 both use the following information:
Femoral neck stress fractures occur in young people engaged in sports, combat training, and other similar activities. Such fractures have a relatively high rate of poor healing, which, in many cases, leaves the patient somewhat impaired. Many such events could be avoided if risk factors (i.e. those characteristics that predict a greater probability of fracture) could be identified. One potential risk factor is bone density. A large database (several thousand) of bone density measures of healthy people has been compiled by the University of California-San Francisco (UCSF). Norms were established for age and sex.
The data set bonedensity.RData includes the age (years), sex, bone density (mg/ml) and the UCSF norm value for that age and sex (mg/ml) for 18 young people with femoral neck stress fractures.
(source: data base 7, Riffenburgh, 2012)
Question 3 (9 marks)
Suppose we wish to test the hypothesis that the average bone density differs between males and females with femoral neck stress fractures.
Explain why an independent samples t-test is not recommended in this instance. (2 marks)
Conduct a non-parametric test to address the above research question. Present your answer using the 5 step process. (7 marks)
Question 4 (3 marks)
Estimate the minimum sample size required to detect a difference of 10 mg/ml of bone density between the means for men and women with a=0.05 and power=0.80 (ß=0.20). Assume the population standard deviation is s=24 mg/ml. Present your answer asa sentence which summarises the required sample size to achieve what power subject to what conditions.

Question 5 (8 marks) Note: Students will get different answers as the data sets differ.
Using the assignment data set assigned to you and R Commander address the research question:
Does the mean systolic blood pressure in the population differ between those whose blood group is O and those whose blood group is not O?
The variables for analysis are bldgrp2 and sysbp. Assume your data set represents a random sample from the population of interest. Fully document your hypotheses, your choice of test, yourworking and your answer to the question.