Study Guide for STAT 305 Final Exam – Fall 2018

Make sure you review all homework assignments, as this material will comprise most of the exam.

Introductory Material:

Given a description of a study/experiment be able to discuss any potential problems or limitations it might have.

Given a specific research question be able to explain how you might design a study to help answer it.

Descriptive Methods:

Numerical

Be able interpret and discuss the following numerical summary statistics. The statistics marked with an asterisk you should be able to calculate by hand if given a very small data set.

  • Mean *
  • Median *
  • Mode
  • Standard Deviation*
  • Standard Error of the Mean*
  • Percentiles/Quantiles
  • InterquartileRange (IQR)
  • Coefficient of Variation (CV)*

Graphical

Be able interpret and discuss the following graphical displays.

  • Histogram
  • Boxplot (Outlier and Comparative)
  • Normal Quantile Plot
  • Bar graph for a single categorical variable
  • 2-D Mosaic Plots
  • Correspondence Analysis
  • Scatterplot

Probability and Probability Distributions

Given data be able to calculate

  • Conditional Probability *
  • Relative Risk(RR)*
  • Odds Ratio (OR)*
  • Sensitivity, Specificity, False Positive Rate, False Negative Rate
  • Positive Predictive and Negative Predictive Values using Baye’s Rule.

Given a Binomial Table (from JMP) be able to calculate and use probabilities associated with a binomial random variable. Also you should be able to calculate the mean and standard deviation of binomial random variable. Understand how to use the binomial distribution in decision making.

Understand the role probability plays in decision making.

Given information about a normally distributed random variable be able to calculate probabilities and find quantiles associated with the random variable.

Statistical Inference

Population Mean ()

Be able to construct and interpret a CI for the population mean.

Be able to carry out a hypothesis test.

Population Proportion ()

Be able to construct and interpret a CI for the population proportion.

Be able to carry out a hypothesis test using the Binomial Exact Test.

Comparing Two Population Proportions ()

Be able interpret a CI for the difference in the population proportions.

Be able to interpret the results from Fisher’s Exact Test.

Be able to interpret RR/OR and CI’s for both RR/OR.

Goodness of Fit and Independence Tests (Chi-square)

Be able to carry out a goodness of fit test and interpret the results.

Be able to carry out a test of independence and interpret the results.

Be able to interpret and discuss the following from JMP output in regards to

Goodness of Fit tests:

  • Bar graph
  • Goodness of Fit test results
  • CI’s for the category

Be able to interpret and discuss the following from JMP output in regards to Tests of Independence:

  • Mosaic plot
  • Correspondence analysis
  • Contingency table
  • Test results

Comparing Two Population Means Using Independent Samples ()

Be able to interpret and discuss the following from JMP output:

  • CI’s for the difference in the population means.
  • P-value from the test for comparing the population means.
  • P-values from the tests for comparing the population variances.
  • Comparative boxplots, histograms, and normal quantile plots

Comparing Two Population Means Using Dependent Samples

Be able to interpret and discuss the following from JMP output:

  • CI’s for the population mean of the paired differences.
  • from a hypothesis test of the population mean paired difference.
  • Histograms and normal quantile plots of the paired differences.

Nonparametric Tests – know when and why we use them

Be able to interpret results from the Wilcoxon Rank Sum test.

Be able to interpret results from the Wilcoxon-Signed Rank Test.

Be able to interpret the results from a Kruskal-Wallis Test.

One-way ANOVA for Comparing Several Population Means

Understand the underlying required assumptions.

Be able to interpret the following from JMP output:

  • Comparative boxplots, histograms, and normal quantile plots.
  • The overall ANOVA test and associated p-value.
  • p-values from the tests for comparing the population variances.
  • Results of Tukey’s HSD for making pairwise comparisons.

Randomized Complete Block Designs

Make sure you understand the reason for blocking and what the benefits of

blocking are.

Be able to interpret the following from JMP output:

  • The overall ANOVA test and associated p-value for comparing treatment means.
  • P-values from the tests for comparing the population variances.
  • Results of Tukey’s HSD for making pairwise comparisons.

Two-Way ANOVA

Make sure you understand what an interaction is and can discuss what an

interaction plot shows.

Be able to interpret the following from JMP output:

  • P-values for testing the interaction and main effects.
  • Plots of the interaction and the main effects.
  • Residual plot
  • Histogram and normal quantile plot of the residuals.
  • Pair-wise comparisons of the treatment means.

Simple Linear Regression and Correlation

Be able to interpret and discuss a scatterplot/scatterplot matrix.

Be able to interpret and discuss the sample correlation ().

Be able to interpret and discuss the following from JMP output:

  • R-square
  • RMSE
  • Estimated regression parameters.
  • P-values from the tests we discussed.
  • Residual plots(residuals vs. fitted values, normal quantile plot, etc.)
  • Confidence intervals for
  • Prediction intervals for an individual

Understand the assumptions required for simple linear regression.

Use the estimated regression model to make predictions.

That should do it!