Part 1 Case Study

Part 1 Case Study




Part 1: Case Study


Potential problems in the manufacturing process

Both the small, medium and large engines would be skewed to having poor performances as listed.

There are also mechanisms that have led to compact, large, mid-size, small, sporty and vans. Passengers are observed to have issues with the vehicles owing to how they have been made.

There are also various vehicles having issues when it comes to the MPG.

The fuel tank is also observed to be having issues with the developments. This is in relation to large, medium and small tanks.

Investigate these high and low measurements

Manufacturer Origin Model Type Passengers Price

Honda non-US Civic Small 4 12.1

Geo non-US Metro Small 4 8.4

Suzuki non-US Swift Small 4 8.6

The price of Honda more so, the small one is affordable.

Highest 10% and the lowest 10% of these measurement

Q1 1.2175

Q4 2.285

Better ways to select the outlier values

When the interquartile range is given a multiplier effect of 1.5. Then we are able to ascertain the outlier values that are present.

Subtraction of 1.5 interquartile range from the 1st quarter would result in less values deemed as outliers. Hence, the numbers would be sorted from high to low. This can be visualized through a box plot.

Fences can be used to introduce interquartile ranges.

Statistical procedures can be used to come up with the skewed values.

Reasoning and answers for distributions that are normal or skewed

Their ought to be outliers in a system of which would result in the framework having a normal curve. The values cannot be the same at any given point. Thus, proving that they are correct and have a statistical element.

Part 2: Box Plot

Population size: 92

Median: 17.6

Minimum: 7.4

Maximum: 47.9

First quartile: 12.125

Third quartile: 23.15

Interquartile Range: 11.025

Outliers: 47.9 40.1

Weight data from the Cars data set located in the Group Project assignment

Values for the 5-number summary

Upper and lower fences

Discuss the shape of the data outliers in your comments.