CENTRAL TENDENCY

Central tendency is the first aspect of a test to consider. Central tendacy

describes the most typical behavior of a group. Four statistics are used for

estimating central tendency: the mean, the mode, the median, and the

midpoint.

Mean

The mean is probably the single most important indicator of central

tendency. The mean is virtually the same as the arithmetic average that

most teacherscalculate in grading classroom tests. The mean is symbolized

in writing by X said “ex bar”). Another way to define a statistical concept is

to give its formula, so let me also define the mean as:

mere X = mean

X = scores
N = number of scores

C = sum (or add) 2

In order to help clarify the reading of such formulas, I will briefly explain

this one in a stepby-step manner. The formula simply says: To get the mean

(g), sum (or add up) (C) the scores (A') , and divide by the number of scores

(,y). These steps are shown in Table 4.5. To find the mean in the example:

(a) sum, or add up the scores, (b) find the number of scores, and (c)

divide the sum of the scores by the number of scores. So the mean in the

example in Table 4.5 would be 69. As mentioned above, this set of

calculations probably looks very familiar since most teachers use the

arithmetic average in lookinm at the results of a classroom test. What they

are checking in the process IS almost exactly the same as the mean and
therefore is an indicator of the central tendency, or typical performance, of

their class on the test.

As with the formula for the mean, all other formulas in this book are

always explained recipe-book style with plenty of examples. In the case of

this formula, the steps seem very easy because the formula and the concept

of the mean are just another way of expressing something that teachers

already know how to do. However, in general, formulas provide more

mathematical precision for defining and discussing statistical concepts. So3 language testers use such
formulas much as linguists and language teachers

use terms like “syntax” and ‘‘phonology’’ when everyone else calls these

concepts grammar and pronunciation. Such formulas are just part of

learning to “speak” language testing.

Mode

Another indicator of central tendency is the mode. The mode is that

score which occurs most frequently. In Table 4.5, what would the mode be?

It would be 69, the only score received by four students. A memory device

that I use to keep the mode straight in my mind is that “mode” can mean

fashionable (as in ci la mode). Thus, the mode would be that score which is

most fashionable, or the one received by the most students. No statistical

formula is necessary for this straightforward idea. However, note that a set

of scores can have two or more modes. Such distributions of scores are

referred to as being bimodal, tnmodal, and so on.

Median

The median is that point below which 50% of the scores fall and above

which 50% fall. Thus, in the set of scores 100, 95, 83, 71, 61, 57, 30, the

median is 71, because 71 has three scores above it (100, 95, and 83) and
three scores below it (61, 57, and 30). What is the median for the following

set of scores: 11,23,40,50,57,63, 86? Fifty, right?

In real data, cases arise that are not so clear. For example, what is the

median for these scores: 9,12, 15, 16, 17,27? In such a situation, when there

is an even number of scores, the median is taken to be midway between the

two middle scores. In this example, the two middle scores are 15 and 16, so

the median is 15.5. Does that make sense? If so, what is the median for these

scores: 11, 28, 33, 50, 60, 62, 70, 98? Your answer should be 55 because that

is the point halfway between the two middle scores, 50 and 60.

In some cases, there is more than one numerically equal score at the

median-for instance, 40, 45, 49, 50, 50, 50, 57, 64, 77. Here, the midpoint

is clearly 50 because there is an odd number of like scores at the median

separating equal numbers of scores on either side.

Still other situations may arise in determining the median, but the

important thing to remember is that the median is the point that divides

the scores 50/50, much like the median in a highway divides the road into

two equal parts. However, in sets of test scores, the median may have a

fraction because students rarely cooperate to the degree that highways do. 6

Midpoint

The midpoint in a set of scores is that point halfway between the highest

score and the lowest score on the test. The formula for calculating the

in id point is:

High + Low

2 Midpoint =

For example, if the lowest score on a test was 30 and the highest was 100,
the midpoint would be halfway between these two scores. To use the

formula: (a) identifj the high and low scores (100 and 30 here), (b) add

the low score to the high one (100 + 30 = 130), and (c) divide the result by

2 as follows:

100+30 130 = 65 Midpoint - = --

2. 2

To review central tendency briefly, four such measures exist: the mean,

the mode, the median, and the midpoint. These are all measures of central

tendency, and each has its strengths and weaknesses. None is necessarily

better than the others, though the mean is most commonly reported. They

simply serve different purposes and are appropriate in different situations,

as you will see at the end of the chapter.

To further review central tendency, look at Table 4.5. I have explained

that the mean, or arithmetic average, in Table 4.5 is 69. The mode, or most

frequent score, also turned out to be 69. The median, that score which

divided the scores 50/50, was also 69. The midpoint, halfway between the

high score of 77 and the low score of 61, was also 69. In this contrived

example, all four measures of central tendency turned out to be the

same-69. However, as you will see in Table 4.8, these four indices for

actual test data are seldom so universally wellcentered and in agreement

on what constitutes the typical behavior, or central tendency, of a group of

scores. For that reason alone, all four should be used. Furthermore, as I

explain in Chapter 5, the deoree to which these four indices of central

tendency are similar is one indication of the degree to which a set of scores

is normally (as in norm-referenced) distributed. 8