Page 168 - Statistics for Dummies
P. 168

152
                                         Part III: Distributions and the Central Limit Theorem
                                                    In Problem 3, you find p(0 < Z < 2.00); this requires Step 5c. First find p(Z <
                                                    2.00), which is 0.9772 from the Z-table. Then find p(Z < 0), which is 0.5000
                                                    from the Z-table. Subtract them to get 0.9772 – 0.5000 = 0.4772. The chance
                                                    of a fish being between 16 and 24 inches is 0.4772.
                                                    The Z-table does not list every possible value of Z; it just carries them out to two
                                                    digits after the decimal point. Use the one closest to the one you need. And just
                                                    like in an airplane where the closest exit may be behind you, the closest z-value
                                                    may be the one that is lower than the one you need.
                                         Finding X When You Know the Percent
                                                    Another popular normal distribution problem involves finding percentiles
                                                    for X (see Chapter 5 for a detailed rundown on percentiles). That is, you are
                                                    given the percentage or probability of being at or below a certain x-value,
                                                    and you have to find the x-value that corresponds to it. For example, if you
                                                    know that the people whose golf scores were in the lowest 10% got to go to
                                                    the tournament, you may wonder what the cutoff score was; that score would
                                                    represent the 10th percentile.
                                                    A percentile isn’t a percent. A percent is a number between 0 and 100; a
                                                    percentile is a value of X (a height, an IQ, a test score, and so on).
                                                    Figuring out a percentile
                                                    for a normal distribution
                                                    Certain percentiles are so popular that they have their own names and their
                                                    own notation. The three “named” percentiles are Q  — the first quartile,
                                                                                                1
                                                    or the 25th percentile; Q  — the 2nd quartile (also known as the median or
                                                                         2
                                                    the 50th percentile); and Q  — the 3rd quartile or the 75th percentile. (See
                                                                           3
                                                    Chapter 5 for more information on quartiles.)
                                                    Here are the steps for finding any percentile for a normal distribution X:
                                                    1a. If you’re given the probability (percent) less than x and you need to
                                                        find x, you translate this as: Find a where p(X < a) = p (and p is the

                                                        given probability). That is, find the pth percentile for X. Go to Step 2.
                                                    1b. If you’re given the probability (percent) greater than x and you need
                                                        to find x, you translate this as: Find b where p(X > b) = p (and p is
                                                        given). Rewrite this as a percentile (less-than) problem: Find b where
                                                        p(X < b) = 1 – p. This means find the (1 – p)th percentile for X.










                                                                                                                           3/25/11   8:16 PM
                             15_9780470911082-ch09.indd   152                                                              3/25/11   8:16 PM
                             15_9780470911082-ch09.indd   152
   163   164   165   166   167   168   169   170   171   172   173