Dq Week 4 Res/341

2461 Words10 Pages
1. Why is the population shape a concern when estimating a mean? * If the population shape is symmetrical, it will be a concern when estimating the mean. The distribution would be close to the center. It all depends on how sensitive the mean is, for example when it is very sensitive in the extreme values and the distribution is not symmetrical, and the mean will be away from the center and more near the extreme values. In statistics normality is important so the underlying population is normally distributed. (Doane & Seward, 2007) * What effect does sample size, n, have on the estimate of the mean? Is it possible to normalize the data when the population shape has a known skew? How would you demonstrate the central limit theorem to your classmates? * When the sample size is larger the smaller the standard deviation or error, then you will have a more dependable estimate. Also remember to analyze the mean with bigger samples. Based on the normality, the central limit theorem is relied on all statistics and tests. It is possible to normalize the data when the population shape has a known skew. There are many ways for normalizing skewed distribution, for instance using the square root transformation and logarithmic transformation just to name a few. (Sekaran, 2003) Starting with a set of data which is not standard, any non-normal or probably the uniform distribution would work in explaining it better to other students. * Example 1; Here is an example of how a not so normal histogram analysis (“Histograms”, 2012) * * The choice of choosing samples from the set of data of size 6, calculates the mean, then repeat several times more, then change to a larger sample size. When your sample sizes increases you will see the histogram of the mean look like a normal distribution. When you have added target, upper and lower limit lines, you can
Open Document