Module 24 – ANOVA (12 of 14 discussion 1) AND Module 25 – Chi-Square Test of Independence (6 of 8 discussion 1)

Module 24 – ANOVA (12 of 14 discussion 1)

Prompt

We will conduct an ANOVA F-test for the variables Color and CFF. The flicker datafile is available in the Data section below. Also, the StatCrunch directions are provided in the list a the bottom of this page.

1. What are the hypotheses for the ANOVA test? Be sure that you define clearly the parameters.
2. Are the conditions that allow us to safely use the ANOVA F-test met? Explain.
   Note: To verify conditions, you will need to examine the distribution of CFF scores for each sample (because the samples are small).
3. Use StatCrunch to create side-by-side dotplots, histograms or boxplots (your choice) to examine the distribution of CFF scores for each sample. You can use either data format; choose one (stacked data in the first two columns; or unstacked data in the the last three columns). To create the side-by-side graphs (for either data format) see the list of StatCrunch directions below. Download the StatCrunch output window (your graph), upload it to your Stats-Class folder, and then embed the .png file (your graph) in your initial post. To recall how to complete these tasks, see the list of StatCrunch directions below.
4. You will also need to compare the sample standard deviations. Use StatCrunch to find the summary statistics, means and standard deviations for the comparison groups (select the the appropriate Descriptive Statistics StatCrunch directions from the list below). Then copy and paste the table into your initial post and explain how the rule of thumb for comparing standard deviations is met.
5. Use StatCrunch to carry out the ANOVA F-test (select the appropriate ANOVA StatCrunch directions from the following two options).

   Anova F-test Stacked Data Format OR
   Anova F-test Unstacked Data Format

   Copy and paste the output table into your initial post.

6. State your conclusion in context of eye color and mean CFF.

Module 25 – Chi-Square Test of Independence (6 of 8 discussion 1)

Prompt

1. A test of independence may be appropriate if we are examining the relationship between two categorical variables in one population. For this situation what is the population? What is the explanatory variable? What is the response variable?
2. What are the hypotheses for the Test of Independence? State hypotheses with reference to the context of the scenario.
3. The spreadsheet of the data looked like this:
   

   Roadside survey data

   Driver	Gender	Alcohol in last two hours?
   Driver 1	M	Yes
   Driver 2	F	No
   Driver 3	F	Yes
   . . .	. . .	. . .
   Driver 619	M	No

   We will not use the raw data. Instead we will use the summarized data shown in the table below.

   Roadside survey summary

   Drank alcohol in last 2 hours?	Yes	No	Totals
   Male	77	404	481
   Female	16	122	138
   Totals	93	526	619

   Use StatCrunch to find expected counts, the Chi-square test statistic and the P-value. (directions)
   Copy and paste your StatCrunch table into your post.

4. How many males in the sample are expected to answer yes to question about alcohol consumption in the last two hours? Show how to calculate this expected count and explain what it means relative to the hypotheses.
5. Explain how we know that this data meets the conditions for use of a chi-square distribution.
6. State a conclusion at a 5% level of significance. Do you think that the data supports the Oklahoma law that forbids sale of 3.2% beer to males and permits it to females?