c.

2000 Del Siegle

Suppose we measured the right foot length of 30 teachers and graphed the results. Assume the first person had a 10 inch foot. We could create a bar graph and plot that person on the graph.

9 10 11 12 13 14
Length of Right Foot

c. 2000 Del Siegle

If our second subject had a 9 inch foot, we would add her to the graph. As we continued to plot foot lengths, a 8 pattern would begin to emerge. 7 6 5 4 3 2 1
Number of People with that Shoe Size

9 10 11 12 13 14
Length of Right Foot

c. 2000 Del Siegle

Notice how there are more people (n=6) with a 10 inch right foot than any other length. Notice also how as the length becomes larger or smaller, there are fewer and fewer people with that measurement. This is a characteristic of many variables that we measure. There is a tendency to have most measurements in the middle, and fewer as we approach the high and low extremes. If we were to connect the top of each bar, we 8 would create a frequency polygon. 7 6 5 4 3 2 1
Number of People with that Shoe Size

You will notice that if we smooth the lines, our data almost creates a bell shaped curve.

9 10 11 12 13 14
Length of Right Foot

c. 2000 Del Siegle

8 7 6 5 4 3 2 1

Number of People with that Shoe Size

You will notice that if we smooth the lines, our data almost creates a bell shaped curve.
This bell shaped curve is known as the Bell Curve or the Normal Curve. 8 7 6 5 4 3 2 1

Number of People with that Shoe Size

9 10 11 12 13 14
Length of Right Foot

c. 2000 Del Siegle

Whenever you see a normal curve, you should imagine the bar graph within it.

9 8 7 6 5 4 3 2 1
c. 2000 Del Siegle

Number of Students

12 13 14

15 16 17 18 19 20 21 22 Points on a Quiz

The mean, mode, andscores for 51all fall on the same Now lets look at quiz median will students. value in a normal distribution.
12
12, 13, 13, 14, 14, 14, 14, 15, 15, 15, 15, 15, 15, 16, 16, 16, 16, 16, 16, 16, 16, 17, 17, 17, 17, 17, 17, 17, 17, 17, 18, 18, 18, 18, 18, 18, 18, 18, 19, 19, 19, 19, 19, 19, 20, 20, 20, 20, 21, 21, 22

13 13 12+13+13+14+14+14+14+15+15+15+15+15+15+16+16+16+16+16+16+16+16+ 17+17+17+17+17+17+17+17+17+18+18+18+18+18+18+18+18+19+19+19+19+ 14 14 14 14 19+ 19+20+20+20+20+ 21+21+22 = 867 15 15 15 15 15 15

867 / 51 = 17

16 16 16 16 16 16 16 16 17 17 17 17 17 17 17 17 17 18 18 18 18 18 18 18 18 19 19 19 19 19 19 20 20 20 20
9 8 7 6 5 4 3 2 1 12 13 14

Number of Students

21 21
22

c. 2000 Del Siegle

15 16 17 18 19 20 21 22 Points on a Quiz

Normal distributions (bell shaped) are a family of distributions that have the same general shape. They are symmetric (the left side is an exact mirror of the right side) with scores more concentrated in the middle than in the tails. Examples of normal distributions are shown to the right. Notice that they differ in how spread out they are. The area under each curve is the same.
c. 2000 Del Siegle

If your data fits a normal distribution, approximately 68% of your subjects will fall within one standard deviation of the mean.
Approximately 95% of your subjects will fall within two standard deviations of the mean. Over 99% of your subjects will fall within three standard deviations of the mean.

c. 2000 Del Siegle

The mean and standard deviation are useful ways to describe a set of scores. If the scores are grouped closely together, they will have a smaller standard deviation than if they are spread farther apart.

Small Standard Deviation

Large Standard Deviation

Click the mouse to view a variety of pairs of normal distributions below.

c. 2000 Del Siegle

Same Means Different Standard Deviations

Different Means Same Standard Deviations

Different Means Different Standard Deviations

When you have a subjects raw score, you can use the mean and standard deviation to calculate his or her standardized score if the distribution of scores is normal. Standardized scores are useful when comparing a students performance across different tests, or when comparing students with each other.

z-score
T-score IQ-score SAT-score

-3
20 65 200

-2
30 70 300

-1
40 85 400

0
50 100 500

1
60 115 600

2
70 130 700

3
80 145 800

c. 2000 Del Siegle

Your next mouse click will display a new screen.

The number of points that one standard deviation equals varies from distribution to distribution. On one math test, a standard deviation may be 7 points. If the mean were 45, then we would know that 68% of the students scored from 38 to 52.

24

31

38

45 52 59 Points on Math Test

63

30

35

40 45 50 55 Points on a Different Test

60

c. 2000 Del Siegle

On another test, a standard deviation may equal 5 points. If the mean were 45, then 68% of the students would score from 40 to 50 points.

Data do not always form a normal distribution. When most of the scores are high, the distributions is not normal, but negatively (left) skewed. Skew refers to the tail of the distribution.

Because the tail is on the negative (left) side of the graph, the distribution has a negative (left) skew.
8 7 6 5 4 3 2 1

Number of People

9 10 11 12 13 14
Test Scores

c. 2000 Del Siegle

When most of the scores are low, the distributions is not normal, but positively (right) skewed.
Because the tail is on the positive (right) side of the graph, the distribution has a positive (right) skew.

9 10 11 12 13 14
Test Scores

c. 2000 Del Siegle

8 7 6 5 4 3 2 1

Number of People

When data are skewed, they do not possess the characteristics of the normal curve (distribution). For example, 68% of the subjects do not fall within one standard deviation above or below the mean. The mean, mode, and median do not fall on the same score. The mode will still be represented by the highest point of the distribution, but the mean will be toward the side with the tail and the median will fall between the mode and mean.

Negative or Left Skew Distribution

Positive or Right Skew Distribution

c. 2000 Del Siegle

Normal Curve Probabilities

The area under the normal curve is 1, with 0.5 to the left of the mean, and 0.5 to the right of the mean. Pg. 993 in your text gives the area under a standard normal curve whose mean is zero and whose standard deviation is 1. The area is given for the area between the mean for the desired number of standard deviations.

c. 2000 Del Siegle

Practice Using the Normal Curve

Use the chart on pg. 993. Sketch the normal curve and shade the appropriate area for each example. Give the probability.
1. z < -2.25 3. z > 1.77 2. z > -2.25 4. -2.25 < z < 1.77

Find z such that P(z) = 0.35 Find z such that P(z) = 0.80
c. 2000 Del Siegle

How big are soldiers heads?

The army reports that the distribution of head circumference among male soldiers is approx. normal with = 22.8 and = 1.1 What % of soldiers have head circumference between 21.7 and 23.9? Put this in terms of : what percentage have heads between - 1 and + 1 ? The 68-95-99 rule says 68% of data falls within 1 SD of , so the percentage of soldiers with head circumference between 21.7 and 23.9 is 68%.
c. 2000 Del Siegle

How big are soldiers heads?

The army reports that the distribution of head circumference among male soldiers is approx. normal with = 22.8 and = 1.1 What % of soldiers have head circumference greater than 23.9? Put this in terms of : what percentage have heads greater than 1 away from ? The 68-95-99 rule says 68% of data falls within 1 SD of , and the curve is symmetric, so the percentage of soldiers with head circumference > 23.9 is 1.0 0.5 0.3413 = .1587, or 15.87%
c. 2000 Del Siegle

Stock Market
Even as bad as the stock market has been lately, the annual rate of return on stock indexes is approximately normal over a long period of time. Since 1945, the S&P 500 has had a mean yearly return of 12%, with a standard deviation of 16.5%. In what range do the middle 95% of all returns lie? Within 2 S.Ds, so between 12-2(16.5) and 12+2(16.5), or -21% to 45%

c. 2000 Del Siegle

More S&P fun

= 12%, = 16.5% (from last slide) The market is down for the year if the return on the index is less than zero. In what proportion of years is the market down? In order to answer this one, we need to know a little more, since 0 is not an even number of standard deviations away from . Remember that our normal curve is for = 0, = 1. We must standardize our data to use the normal area.

c. 2000 Del Siegle

Using z-scores, a.k.a standardizing our data

Z-scores adjust any normally distributed data set back to the standard curve so that we can use our chart. x To standardize, calculate z as: z where x is the value youre looking for, is the mean and is the std. dev. Sketch a picture of the normal curve and the area youre looking for. Then use z in the table to answer your question.

c. 2000 Del Siegle

Back to the problem

= 12%, = 16.5%; want to know the proportion of years with negative returns (x < 0) Calculate z: z = (0 0.12) / .165 -0.73 Sketch a picture. Check the chart: P(z) = 0.2673. But wait, thats from 0 to 0.73 You need LESS THAN -0.73, 0.5 0.2673 = 0.2327, or in about 23% of the years, the S&P 500 has a negative return.
c. 2000 Del Siegle

One more soldier question

The army reports that the distribution of head circumference among male soldiers is approx. normal with = 22.8 and = 1.1. Helmets are mass produced for all except the smallest 5% and the largest 5% of all head sizes. Soldiers with the smallest or largest 5% get custom made helmets. What head sizes get custom-made helmets? The middle 90% get mass produced helmets, which is the average head size + 45% and the average head size 45% . Use the table to find 45%. At z = 1.64, P(z) = 0.4495 and at z = 1.65, P(z) = 0.4505. Split the difference to get P(1.645) = 0.45. Since the table is symmetric, P(-1.645) = 0.45 as well. x Substitute into the formula z to get x > 24.6 and x < 21.0

c. 2000 Del Siegle

Are kids getting smarter?

When the Stanford-Binet IQ test was introduced in 1932, it was adjusted so that the scores for each age group of children roughly followed the normal distribution with = 100 and = 15. The test is readjusted from time to time to keep = 100. If present-day children took the 1932 version of the test, their mean score would be about 120. The reasons for the increase over time are not known, but may include better nutrition and more experience taking standardized tests.

c. 2000 Del Siegle

Are kids getting smarter?

IQ scores above 130 are often called very superior. What percentage of children had very superior scores in 1932? If present-day children took the 1932 test, what percent would have very superior scores? Assume that = 15 does not change. In 1932, 2.28%; present day, 25.14%

c. 2000 Del Siegle