breakout session #6 numerical statistics
DESCRIPTION
Breakout Session #6 Numerical Statistics. Presented by Dr. Del Ferster. What’s in store for today?. We’re going to wrap up our monthly meetings by considering numerical statistics. Next time, we’ll be working on a review. We’ll look at mean, mode, median, and range. - PowerPoint PPT PresentationTRANSCRIPT
![Page 1: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/1.jpg)
Breakout Session #6Numerical Statistics
Presented byDr. Del Ferster
![Page 2: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/2.jpg)
What’s in store for today?
›We’re going to wrap up our monthly meetings by considering numerical statistics.–Next time, we’ll be working on a review.
›We’ll look at mean, mode, median, and range. –We’ll also look at standard deviation ( a measure of “spread” or dispersion)
![Page 3: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/3.jpg)
What’s in store for today?
›We’ll review how to read a Z CHART (a standard normal chart), to compare scores relative a population’s mean and standard deviation.
›We’ll wrap it up by looking at some practice problems.
![Page 4: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/4.jpg)
First, a picture of Conor!!!Come on now, you knew that I’d do that!
![Page 5: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/5.jpg)
Distributions of DataAn introduction to the Normal Curve(sometimes called the Bell curve)
![Page 6: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/6.jpg)
Different Distributions of Data› Data can be distributed or “spread out” in different ways.
› It can be spread out so that more data falls on the left.
We would say that this distribution is SKEWED RIGHT
(interestingly, the SKEW is NOT where the data clusters, it’s where the tail falls.
![Page 7: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/7.jpg)
Different Distributions of Data› It can be spread out so that more data falls on the right
We would say that this distribution is SKEWED LEFT
![Page 8: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/8.jpg)
Different Distributions of Data› Of course, the data might fall all over the place.
In this case, there doesn’t seem to be any discernible pattern.
![Page 9: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/9.jpg)
Different Distributions of Data› Today, let’s look at the case where data tends to
cluster around a central value, with no tendency to the left or right. It gets closer to a NORMAL DISTRIBUTION, like this:
![Page 10: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/10.jpg)
The Normal DistributionA closer look at key ideas.
![Page 11: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/11.jpg)
The Bell Curve› The Bell curve is a Normal Distribution
› The yellow histogram shows some data that follows the curve closely, but not perfectly. (This is USUAL, as it’s seldom that a data set is perfectly normal.)
![Page 12: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/12.jpg)
What’s normal, Del?› Many things in life closely follow a normal distribution.–The heights of a group of people.–The size of widgets that are produced by a machine.
–Errors in measurements.–Blood pressure.–Grades on a test
![Page 13: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/13.jpg)
A bit more on the normal distribution
› We say that the data is Normally Distributed
› In a normal distriubtion– The mean, mode, and median are the SAME NUMBER
– The graph is symmetric with respect to the center
– 50% of the values fall above the mean, and 50% of the values fall below the mean
![Page 14: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/14.jpg)
A cool online app› This Quincunx or Galton Board app will simulate a
normal curve.
› Let’s look at it.
› Here’s the link
› http://www.mathsisfun.com/data/quincunx.html
![Page 15: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/15.jpg)
Standard Deviation› Standard Deviation is a measure of how SPREAD OUT the data are. Some books refer to Standard Deviation as a measure of dispersion.
› You’re familiar with another measure of dispersion—RANGE.
› The formula for calculating standard deviation is pretty ugly, and luckily, many software packages along with graphing calculators provide a fast, easy way to find it.
![Page 16: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/16.jpg)
Calculating Standard Deviation Rocking Math Old School!!!› Calculate the average (mean) of your set of numbers
› Calculate the difference between each number and the mean
› Square the differences
› Add up the square of all the differences
› Divide this by one less than the number of numbers in your set - this is called the variance
› Take the square root of the variance and you've got the standard deviation
![Page 17: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/17.jpg)
More on Standard Deviation› When you calculate the standard deviation of your
data, you will find, in general:
68% of the data falls within 1 standard deviation of the mean95% of the data falls within 2 standard deviations of the mean99.7% of the data falls within 3 standard deviations of the mean
![Page 18: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/18.jpg)
Statements regarding Data and Standard Deviation› In general, we can say that any value in a data set is:
› “likely” to be within 1 standard deviation of the mean
› 68 out of 100 are
› “very likely” to be within 2 standard deviations of the mean.
› 95 out of 100 are
› “almost certain” to be within 3 standard deviations of the mean.
› 997 our of 1000 are
![Page 19: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/19.jpg)
Z scores› A z score tells us the number of STANDARD DEVIATIONS that a given data point falls above or below the mean.
› Some books refer to these as STANDARD SCORES
› Our formula for finding z scores
› This process is sometimes called STANDARDIZING
xz
![Page 20: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/20.jpg)
Example: Determining a z score› Beth scored a 92 percent on her last statistics exam. The class average was 80 percent and the standard deviation was 5.
› Calculate her z score.
xz
92 80
5z
2.4z
Beth’s score exceeded the mean by 2.4 standard deviations.
![Page 21: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/21.jpg)
Why standardize?› Most data sets have very different means and standard deviations.
› The process of standardizing (or calculating z scores) allows us to convert all normal distributions to one (the STANDARD NORMAL DISTRIBUTION)
![Page 22: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/22.jpg)
Let’s do a problem together› Dr. Ferster has just given a quiz to his Math 204
class. Out of 60 points, his students recorded the following grades:
20, 15, 26, 32, 18, 28, 35, 14, 26, 22, and 17
›Dr. F. is bummed , these students must not have studied very much.
›Perhaps they were watching the Packers.
![Page 23: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/23.jpg)
Continuing our problem
20, 15, 26, 32, 18, 28, 35, 14, 26, 22, and 17
› If Dr. F grades normally, it will be pretty ugly.
›Suppose he decides to grade on a normal curve, so that the only people who fail will be those who score more than 1 standard deviation below the mean.
![Page 24: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/24.jpg)
Continuing our problem
20, 15, 26, 32, 18, 28, 35, 14, 26, 22, and 17
›Calculate the mean› I’ll tell you that the standard deviation is 6.6
›Calcualate the z score for –SUZY who scored 20.–SPARKY who scored 15.–STELLA who scored 26.
xz
![Page 25: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/25.jpg)
SUZY—scored 20
› Suzy scores .45 standard deviations below the mean, however she still passes (she isn’t more than 1 standard deviation below the mean—that would be z = -1)
xz
20 23
6.6z
0.45z
23
. . 6.6
mean
st dev
![Page 26: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/26.jpg)
SPARKY—scored 15
›Sparky scores 1.21 standard deviations below the mean, So he fails!
xz
15 23
6.6z
1.21z
23
. . 6.6
mean
st dev
![Page 27: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/27.jpg)
STELLA—scored 26
› Suzy scores .45 standard deviations above the mean, so she passes. Her score falls above the mean.
xz
26 23
6.6z
0.45z
23
. . 6.6
mean
st dev
![Page 28: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/28.jpg)
A picture for that last problem
Sparky
Suzy Stella
![Page 29: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/29.jpg)
A look at the standard normal curve and areas› We can determine the percentage of our data that
falls – Below a given z score– Above a given z score– Between a given z score and the mean– Between 2 different z scores.
› In the old day, we used tables or charts.
› I’m going to show you how to read one later on…I know that you’re psyched for that!
› We can also look at a visual, which we’ll do on the next slide.
![Page 30: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/30.jpg)
A look at the standard normal curve and areas› The picture below illustrates how the percentages of area
under a normal curve are related to different z scores.
![Page 31: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/31.jpg)
A look at the standard normal curve and areas› Of course, in this age of technology, you shouldn’t
be too surprised that we can find an online app that determines areas, too.
› Let’s look at one. Here’s the link
› http://www.mathsisfun.com/data/standard-normal-distribution-table.html
![Page 32: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/32.jpg)
Old School Stats with Del—Reading a z chart› I’m partial to the charts that read areas ALL THE WAY TO THE LEFT...or below the z score that we’re reading.
› I’m going to give you 2 charts—1 is useful for positive z scores, the other is used for negative z scores.–We’ll read both charts the same way.–Suppose we want to find the area to the LEFT of2.16z
![Page 33: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/33.jpg)
Reading a z chart› We seek the area below
› First, grab chart that reads positive z scores.–Now, locate the 2.1 on the left-most column.
–Read across the top of the page till you find the column that is headed 0.06.
–Read the intersection and you find our area
–Did you get 0.9846?
2.16z
![Page 34: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/34.jpg)
Let’s look at a problem.› Skippy’s Widget factory manufactures widgets that
have a length of life that is normally distributed with a mean of 12.3 years, and a standard deviation of 2.5 years.
› What percentage of the widgets that Skippy’s Widget factory produces will last:
› 1. Less than 9.425 years
› 2. More than 16.925 years
› 3. Between 7.75 years and 17.9 years.
![Page 35: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/35.jpg)
Solution to our problem› mean of 12.3 years, standard deviation of 2.5
years.
› What percentage of the widgets that Skippy’s Widget factory produces will last:
› 1. Less than 9.425 years
12.3 2.5
xz
9.425 12.3
2.5z
1.15z
![Page 36: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/36.jpg)
Solution to our problem (continued)› Now, we pick up our z chart and locate
› We’re looking to find the area below this z (remember we want to find the percentage of widgets that last less than the given number of years)
› so, we read our z chart and find
›.1251›So, 12.51% of the widgets last
less than 9.425 years
1.15z
![Page 37: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/37.jpg)
Solution to our problem› mean of 12.3 years, standard deviation of 2.5
years.
› What percentage of the widgets that Skippy’s Widget factory produces will last:
› 2. More than 16.925 years
12.3 2.5
xz
16.925 12.3
2.5z
1.85z
![Page 38: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/38.jpg)
Solution to our problem (continued)› Now, we pick up our z chart and locate
› We’re looking to find the area above this z (remember we want to find the percentage of widgets that last more than the given number of years)
› so, we read our z chart and find
› .9678
› So, 96.78% of the widgets last less than 16.925 years
› This means that 100%-96.78% or
› 3.22% of the widgets will last more than 16.925 years
1.85z
![Page 39: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/39.jpg)
Solution to our problem› mean of 12.3 years, standard deviation of 2.5
years.
› What percentage of the widgets that Skippy’s Widget factory produces will last:
› 3. Between 7.75 years and 17.9 years.
12.3 2.5
xz
7.75 12.3
2.5z
1.82z
![Page 40: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/40.jpg)
Solution to our problem› mean of 12.3 years, standard deviation of 2.5
years.
› What percentage of the widgets that Skippy’s Widget factory produces will last:
› 3. Between 7.75 years and 17.9 years.
12.3 2.5
xz
17.9 12.3
2.5z
2.24z
![Page 41: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/41.jpg)
Solution to our problem (continued)› Now, we pick up our z chart and locate
› We want to find the AREA between these 2 z-scores
› Let’s read our z charts—Fun, eh??
1.82z 2.24z
2.24z 1.82z
![Page 42: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/42.jpg)
Now, let’s work on that solution
We seek this shaded areaSo, we’ll read our 2 z-scores, and subtract the areas that we get from the z chart.
![Page 43: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/43.jpg)
Now, let’s work on that solution
Reading z= 2.24, we get 0.9875Reading z= -1.82, we get 0.0344So, the area shaded in red can be found by subtracting these values (I get 0.9531)
So, 95.31% of the widgets will last between 7.75 years and 17.9 years
![Page 44: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/44.jpg)
“Test-Like Problems”Dealing with Numerical Statistics
![Page 45: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/45.jpg)
Practice Problems› Let’s take some time and work on some problems
that deal with numerical statistics.
› The last page of the handout contains a “partial” z chart.– Of course, you can use the full z chart that we
used earlier.
› On a happy note!– I spoke with Dr. Morgan about a formula sheet for
the exam that you’ll take in May– Let’s say that I’m cautiously optomistic!
![Page 46: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/46.jpg)
Solutions to Practice Problems
› 1. Mean time spent with dad =
› 2. Mean time spent with mom=
› 3. Range of time spent with mom=
› 4. Range of time spent with dad=
› 5. Median time spent with mom=
› 6. I actually calculated the standard deviations, the greater spread is with the time spent with MOM
38.3 min .
242.5 min.
25 min.
20min.
242.5 min.
6.96
8.54Dad
Mom
![Page 47: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/47.jpg)
Solutions to practice problems (continued)
› 7.
› 8. raw score (x) =49
› 9A. English: Science:
History:9B. English (biggest z score)9C. .6915 or 69.15%9D. 1-.9452 = .0548 0r 5.48%
2.50z
1.60z 0.50z 1.25z
![Page 48: Breakout Session #6 Numerical Statistics](https://reader036.vdocument.in/reader036/viewer/2022070403/568139b8550346895da15753/html5/thumbnails/48.jpg)
Wrapping Up›Thanks for your attention and participation.
› It’s been a tough winter—not just the snow, but all of the adjusted schedules.
› I have the utmost respect for what you do as a professional!
›Here’s hoping that Spring brings us sun and smiles.
› If I can help in any way, don’t hesitate to shoot me an email or give me a call.