State the test statistic and corresponding p-value |Operation Research
Part A: Analysis of traffic data
In this section you will use the data you collected in Assignment 1.
Car Size (m^3) | Body |
Year |
Year |
9.941 |
Small Cars |
2003 |
Less than 2005 |
10.251 |
Small Cars |
1996 |
Less than 2005 |
10.699 |
Small Cars |
2012 |
Greater than 2005 |
9.139 |
Medium Cars |
1972 |
Less than 2005 |
10.426 |
Medium Cars |
1986 |
Less than 2005 |
11.304 |
Medium Cars |
2001 |
Less than 2005 |
12.255 |
Medium Cars |
2007 |
Greater than 2005 |
12.332 |
Medium Cars |
2007 |
Greater than 2005 |
13.305 |
Large Cars |
2004 |
Less than 2005 |
13.387 |
Large Cars |
2004 |
Less than 2005 |
13.955 |
Large Cars |
2004 |
Less than 2005 |
13.402 |
Large Cars |
2011 |
Greater than 2005 |
13.470 |
Large Cars |
2011 |
Greater than 2005 |
13.484 |
Large Cars |
2010 |
Greater than 2005 |
14.618 |
Luxury Cars |
2010 |
Greater than 2005 |
14.987 |
Luxury Cars |
2010 |
Greater than 2005 |
13.814 |
Luxury Cars |
2002 |
Less than 2005 |
14.278 |
Luxury Cars |
2002 |
Less than 2005 |
1. Comparing two means:
a) Confidence interval for difference between means
(i) Use Minitab to construct a 95% Confidence Interval for the difference between the means of your numeric variable for the different levels of your 2-level categorical variable.
(ii) Interpret your confidence interval in context (remember units).
(iii) Use your confidence interval to draw a conclusion about the difference (if any) between the two levels of your categorical variable.
b) Two-tailed hypothesis test for the difference between two means
(i) State the null and alternative hypotheses (in words and symbols) for testing if there is a significant difference between the means of your numeric variable for the different levels of your 2-level categorical variable.
(ii) Use Minitab to carry out the test. State the test statistic and corresponding p-value for these hypotheses.
(iii) Explain whether you have evidence for or against the null hypothesis.
(iv) State your conclusion in a form that a non-statistician would understand.