RNR 613 lab

1. first approach: use one-sample t-test. Click Analyze, Distribution, put difference into Y, columns. Then click red triangle under difference and click test mean.

2. Second approach: use matched pairs. Click Analyze, Matched Pairs, put two columns (e.g., pretest and posttest) into Y, Paired Response.

· Two-sample t Procedure

(Note: the default assumption here is unequal population variances. For equal variance assumption, after “fit y by x” you need to click on the red triangle and select Means/Anova/Pooled t on the Oneway Analysis output window.)

Exercise 1. Produce side by side box plot for the typing speed of different brands (dataset )

2. Compare the average height between male and female groups( dataset ) – with equal variance assumption.

Lab 3: check http://web.utk.edu/~cwiek/201Tutorials/ for

Transformations of Variables - Examples

· Log Transformation

· Negative Reciprocal Square Root

Graphical Display of Quantitative Data

· Normal Probability Plot and Goodness of Fit Test

Miscellaneous Topics

· Excluding Data from an Analysis

Rank sum test (non-parametric way for two independent samples): dataset (make sure the variable “group” is nominal)

Analyze > Fit Y by X; select red down arrow > Nonparametric > Wilcoxon Test

Signed Rank test (non-parametric way for paired samples):

use dataset: language instruction

approach 1) similar to the first approach in paired t-test, Analyze > Distribution > Test Mean, click box “Wilcoxon Signed Rank test”.

2) similar to the second approach in paired t-test: after the matched pairs , check “Wilcoxon signed Rank”.

Exercise: 1.1) Exclude the outlier(s) in the unemployment data (in the normal prob plot tutorial)

1.2) Obtain the result of the goodness of fit test for the new dataset in 1.1)

2. Use the nonparametric test to compare group means in the paired sample( dataset )

Lab 4: 1) Check the equal variance assumption & Welch test:

Suppose you want to test whether two variances (males and females) are equal, instead of two means.

1.	Open the Big Class.jmp sample data table - dataset
2.	Select Analyze > Fit Y by X.

3.	Select height and click Y, Response.
4.	Select sex and click X, Factor.

5.	Click OK.
6.	From the red triangle menu, select Unequal Variances.

2) Compare several means (doc)

Exercise: 1) determine an appropriate test to compare the two group means in the dataset 1 : i.e, check normality of each group? Equal variance?

2) how about the dataset 2 ?

Lab 5: 1) Extra SS and Linear Combinations and Multiple Comparisons of Means (doc)

2) check http://web.utk.edu/~cwiek/201Tutorials/ for

Simple linear regression

· Least-Squares Regression Line, Residuals Plot and Histogram of Residuals

Exercise: apply various multiple comparison methods to compare the judge groups for the data case0502.jmp

Lab 6: check http://web.utk.edu/~cwiek/201Tutorials/ for

Correlation and Regression

· Inference about Regression Coefficients

· Confidence Intervals and Prediction Intervals for Regression Response

Note: In JMP you can get the lack of fit test result either via the Fit Model platform or via the Fit Y by X platform, as long as there are replicated Y’s at some levels of X’s.

A video for Fit Y by X bivariate platform

A video for “more on simple linear regression”

Exercise: find the relationship between the area and species for the data case0801.jmp, and check the model assumptions.

Lab 7: Multiple linear regression

Exercise: 1) Do data exploration on the variables in dataset ex1014.jmp.

2) Construct an appropriate relationship between the protein (response) and the copper and zinc.

Lab 8: Variable selection in multiple linear regression

Video

Lab 9: Multi-factor ANOVA and repeated measures

Video

Lab 10: check http://web.utk.edu/~cwiek/201Tutorials/ for

Graphical Display of Categorical Data

· Pareto Chart

· Mosaic Plot

Test of Independence for Two Categorical Variables

· Chi Square Test for Two-Way Table (if want to get Fisher exact test result for rXc contingency table, check “contingency analysis” -> “exact test” -> “Fisher exact test”)

A video for “contingency tables”

Practice: ex19.5.jmp

Lab 11: Logistic regression

Lab 12: check http://web.utk.edu/~cwiek/201Tutorials/ for

· Taking a Simple Random Sample

· Forcing Categorical Data to Display in a Specific Order

Graphical Display of Quantitative Data

· Time Plot

Miscellaneous Topics

· Getting Multiple Histograms on the Same Scale

· Using Different Symbols for Different Groups in Graphical Displays