Minitab assignment3

Chapter 12 Minitab Express

The activity includes two questions. Using the data in the textbook dataset, answer the following

questions:

Question 1:

Some of the counties included in the dataset had reformed their welfare systems in the past ten

years. An assumption of experimental design is that random assignment controls for differences

between the groups. In this case, assume that the counties were randomly assigned to reform

their welfare system or to maintain the current system.

Explore the effect of the welfare reform initiative using graphical displays. Compare the

distribution of at least 3 variables that might be affected by welfare reform. You can use

boxplots, histograms, etc. Do you see any effect of the welfare reform initiative on the variables

that you selected?

For this question, you’ll use the variable Experimental Welfare Reform status in 2004 (C38) to identify the two groups. I will investigate the effect of the experimental welfare reform program on infant death rate (C60). I created boxplots (Graph, Boxplot, with groups, select C60, then Group variables for grouping C38).

Here is the output. Notice the summary statistics are shown below the boxplots.

We don’t see much difference in infant death rate between the counties with and without welfare reform. There is an outlier in the counties without welfare reform so I should investigate to identify this county and determine why there is an outlier. I can also create other displays (histograms, dotplots, etc).

Question 2:

Use the t-test (independent samples) to see if a county's welfare system reform status

influenced the unemployment rate.

For this demo, I will investigate the effect of the welfare reform on the infant death rate. Use

Statistocs, Two-samples, t for the means.

Select “both samples are in one column,” select the column with the samples (C60) and the

column that identifies the groups (C38).

In options, select 95 confidence level, Alternative hypothesis: difference ≠ 0.00. Check to see

that “Assume equal variances” is not selected.

Here is my output:

We see that the t-statistic is -0.48 with a p-value of 0.6343. We fail to reject; we do not have

evidence that there is a difference in the infant death rate based on the welfare reform status of

the county. This is supported by the boxplot from question 1.