This preview shows page 1. Sign up to view the full content.
Unformatted text preview: which we can see from the histograms. Mean is affected from this outlier but the median is not. Therefore, we can conclude that median is more insensitive location measure to the outliers than mean. 02 March 2011 (d) Change the data 558 with 122. Construct a scatter plot for the new data set. • In Excel follow the route Insert ‐> Scatter ‐> Scatter with only Markers • A new tab “Design” will automatically open • To identify the input data click “Select Data” • Choose the new data as “Chart data range” and click “OK” You will obtain the following scatter diagram: 128
126
124
122
120
118
116
114
112
110
108 Series1 0 10 20 30 40 50 (e) Comment on whether there is any clear indication or not the sample came from a population that depicts a bell shaped distribution. The data is scattered quite randomly, therefore there is no clear indication for a bell shaped population. Note: Remember our discussion on percentile calculation in part (a). The percentiles were calculated by assuming that the underlying population comes from a normal distribution. Normal distribution is a bell shaped distribution and before starting your statistical analysis, it is always recommended to plot a scatter diagram of your sample. If there is a bell shape pattern in your scatter plot, than you can feel free to assume that the population is normally distributed. If there is no such clear pattern, than the population may or may not normally distributed hence you should be careful to use the methods that assumes normally distributed population. (f) Consider the following smal...
View
Full
Document
This document was uploaded on 02/22/2014.
 Spring '14

Click to edit the document details