118 PART 3 Getting Down and Dirty with Data

approach. The difference is that each type of mean adds a slightly different twist

to the basic mathematical process.

INNER MEAN

The inner mean (also called the trimmed mean) of N numbers is calculated by

removing the lowest value (the minimum) and the highest value (the maximum),

and calculating the arithmetic mean of the remaining N – 2 inner values. For the

sample of seven values of DBP from study participants from the example used

earlier in this chapter (which were 84, 84, 89, 91, 110, 114, and 116 mmHg), you

would drop the minimum and the maximum to compute the inner

mean: 84

89

91

110

114

5

488 5

97 6

/

/

. .

An inner mean that is even more inner can be calculated by making an even

stricter rule. The rule could be to drop the two (or more) of the highest and two (or

more) of the lowest values from the data, and then calculate the arithmetic mean

of the remaining values. In the interest of fairness, you should always chop the

same number of values from the low end as from the high end. Like the median

(discussed earlier in this chapter), the inner mean is more resistant to extreme

values called outliers than the arithmetic mean.

GEOMETRIC MEAN

The geometric mean (often abbreviated GM) can be defined by two different-

looking formulas that produce exactly the same value. The basic definition has

this formula:

Geometric Mean

II

GM

X

N

We describe the product symbol Π (the Greek capital pi) in Chapter 2. This formula

is telling you to multiply the values of the N observations together, and then take

the Nth root of the product. Using the numbers from the earlier example (where

you had DBP data on seven participants, with the values 84, 84, 89, 91, 110, 114,

and 116 mmHg), the equation looks like this:

GM

84

84

89

91 110 114 116

83 127 648 746 160

93 4

7

7

,

,

,

,

,

.

Even with technology, this formula is computationally challenging. By using log-

arithms (which turn multiplications into additions and roots into divisions), you

can develop a numerically stable alternative formula, which is:

log(

)

log(

) ,or

antilog

log(

)

GM

X

N

GM

X

N

This formula may look complicated, but it really just says, “The geometric mean

is the antilog of the mean of the logs of the values in the sample.” In other words,