1073 words - 5 pages

Weight versus Rate:

The Relation of Vehicle Weight to Fuel Consumption

Darren Jay Chapman

Davenport University

The Relation of Vehicle Weight to Fuel Consumption

The purpose of this brief analysis was to determine the possible correlation of vehicle weight to fuel-consumption (MPG). The vehicles selected for the analysis are all classified as high performance-sports cars and are models manufactured in 2006-2008. Although the performance packages may differ, the basic vehicle designs are two door coupe models with manual transmissions. The study does not account for aftermarket modifications or individual driving habits. The vehicles were randomly selected without replacement ...view middle of the document...

Operation Vehicle Weight - Weight OL MPG - MPG OL

Mean 3,301 3,268 25.53 26.36

Median 3,221 3,221 26 26

Mode 3,000 3,000 25 25

Midrange 875 875 9.5 9.5

Standard Deviation 420.8 384.8 4.257 2.972

Figure 1 - Collected Data Summary

Development of a scatterplot does reveal a line of regression; see table 4 in appendix A. Before the strength of the regression can be determined, we need to look for outlier (OL) data values that may have an impact on the data. First, the data was inspected for data entry error and no data entry errors were noted. Secondly, a box plot was constructed to reveal any outlier data. There are two pieces of data that lie outside the standard deviation of 4.257 miles as seen in tables 5 and 6 in appendix A. These two pieces of data do have a significant impact on the on the standard deviation of the data as seen in figure 1. The histograms correlating to the new data collection can be viewed in tables 7 and 8 in appendix A. Another scatterplot was constructed with the two outlier data pieces removed and the line of regression is much more obvious as seen in table 9 in appendix A. With the two outlier pieces of data removed, we will also be working with a sample size of twenty eight (n = 28). For our purpose, we will figure the correlation strength for both sets of data.

Now that all data has been examined, we will look at the actual strength of the linear correlation. Figure 2 shows the Pearson Correlation Coefficients of the data (Uconn, 2008).

n α=.05 α=.01 P-Value Pearson Value

30 .349 .449 .023 -0.414

28 .361 .463 0.121 -0.300

Figure 2 - Computed Pearson Correlation Critical Values

Table 10 in appendix A indicates critical values of .349 and .449. These values state there is a 95% chance that a sample size of 30, with no linear correlation, the coefficient (r) will exceed the value of .349 and a 1% chance that the absolute value of r will exceed the critical value of -0.449 (Triola, 2007). Since the sample size of 30 has a P-value less than α=.05...

Tap into the world’s largest open writing community