<PRE><B>Question 715509</B>
19) A bimodal distribution is going to have some kind of underlying binary variable that will result in a separate mean for each value of this variable. One example can be human weight - biological sex is (mostly) binary and is a statistically significant indicator of how heavy a person is. A rectangular distribution will have an equal value under all parts of the curve; that is, it will be close to uniform. One possible example is the decimal part of a human&#39;s weight; that is, if a person weighs exactly 158.938 pounds, the variable of interest is .938. There&#39;s no reason for one value of this to be significantly more likely than any other. A positively skewed distribution has the bulk of its values below the mean, so that the median is less than the mean. Income is a good example of a positively skewed variable -- there will be a few people with extremely high incomes, but most people will have incomes bunched together below the mean.&lt;br&gt;

That said, I don&#39;t know what examples are in your book. If you can&#39;t use those, follow the advice for each distribution and find something else that roughly fits that shape.&lt;br&gt;

20) FOX News is a very good source for misleading graphs *nod*. Try, in particular, to find either the one with the unemployment rate during Obama&#39;s first term, or the one showing &quot;job loss by quarter&quot; starting in 2007. Those are both pretty &quot;good&quot; and should be easy to find via Google search. If this has to from a newspaper or magazine, I can&#39;t give you a specific graph, but you can use the ones I showed you as a comparison of what to look for.&lt;br&gt;

21a) A frequency table is the list of all of the possible results that can occur during a random event, and each result&#39;s respective chance of happening.
21b) This is not possible without Table 1-10. I&#39;m a statistician, not a psychic :P&lt;br&gt;

22a) I don&#39;t know exactly how the definition changes given a different set of data, but this data set is broken into groups because the researchers expected there was an underlying variable that would make the two groups different.
22b) Again, this is difficult without the table. Would also make it possible for a better idea of what makes 21a different from 22a.</PRE>