STAT 13

(Sec. 1a-1c)

Introduction to Statistical Methods for the Life and Health Sciences

Instructor: Ivo Dinov, Asst. Prof.

Departments of Statistics & Neurology
    http://www.stat.ucla.edu/~dinov/


Lab 10

Also see this for an additional tutorial.

Thursday, Nov. 20, 2001

Objective:In this lab, you will perform and interpret 1- and 2-sample t-tests (or z-tests).

For today’s lab we will be using the dataset at the following location:

http://www.stat.ucla.edu/~rgould/datasets/m12s00.dta

This is the dataset collected from Prof Gould’s course last spring.Students were asked seven questions:

1)gender (m/f)

2)height (inches)

3)weight (pounds)

4)Do you smoke? (0=no; 1=yes)

5)Who do you want for President? Bush, Gore, other

6)Rate your math ability: (1,2,3,4,5) 1 is much below average, 3 is average, 5 is much above average

7)Rate your math anxiety: (1,2,3,4,5) 1 is much below average, 3 is average, 5 is much above average

Let’s ask the following research questions:

1.Are females shorter than males, on average?

2.Do males and females differ in their weights, on average?

3.Does the proportion of students who currently smoke differ from the previous finding that 15% of students were smokers?

To perform a 2-sample t-test of means with independent samples and unequal variances, type

ttest varname, by(groupvar) unequal

To perform a 1-sample z-test of proportions, type

prtest varname=#[this performs a z-test with p=#]

For each of the three research questions:

Specify the null and alternative hypotheses;

Perform the appropriate statistical test;

Verify that the SE used in the statistical test is computed correctly;

Draw a sketch of t or z distribution, identify the mean of the distribution, the sample statistic, and the region covering the p-value.

How many degrees of freedom are there according to what you learned in class?

Using a = .05, what do you conclude about the null hypothesis?

What do you conclude about your research question?


\Ivo D. Dinov, Ph.D., Departments of Statistics and Neurology, UCLA School of Medicine/