Exploring Sampling Distributions with Fathom using Random Rectangles, Lab Reports of Mathematics

A step-by-step guide on how to use fathom software to explore sampling distributions using a collection of 100 random rectangles. It covers sampling, computing summary statistics, and comparing sampling distributions with population distributions. The activity involves drawing samples of size 5 from the collection and computing measures such as mean, median, and maximum. It also includes predictions, sketches, and comparisons of sampling distributions.

Typology: Lab Reports

Pre 2010

Uploaded on 08/18/2009

koofers-user-lwa-1
koofers-user-lwa-1 🇺🇸

10 documents

1 / 3

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
!
Exploring Statistics with Fathom 4: Sampling Distributions 127
© 2007 Key Curriculum Press
In this activity you’ll use Fathom to draw samples of size 5 from a collection of
rectangles many times to explore the sampling distribution of various summary
You will need
Sampling
Rectangles.ftm
statistics.
EXAMINE DATA
1. Open SamplingRectangles.ftm. The collection named Rectangles contains
100 random rectangles.
Q1 Make a plot of Area and describe the distribution of Area in terms of shape,
center, and spread.
INVESTIGATE
Taking a Sample
2. With the collection selected, choose CollectionSample Cases. By default,
Fathom takes a sample of ten cases with replacement and places them in a new
collection named Sample of Rectangles. Youll change this to fi ve cases without
replacement.
Notice that animation
is on by default. You
may want to change
this later.
3. Double-click the sample collection to
show its inspector. On the Sample
panel, change the settings to match
these. Click Sample More Cases.
Remember that you
can drag an attribute
from the Cases panel
of the inspector.
4. You now have a sample of 5 areas from
the collection of all 100 rectangles. Make
a dot plot of the attribute Area for your sample.
Choose GraphPlot
Value and type
!"#$%&. Then do
the same for the other
values.
5. You need to compute some summary statistics
for your sample. Plot the values for the mean,
median, and maximum of your sample.
Q2 In a moment, you are going to repeat this
sampling, say, 200 times, to create sampling
distributions of these summary statistics (or
measures). But fi rst, sketch distributions to
predict what you will get for the set of 200 mean, median, and maximum areas.
Introduction to Sampling Distributions—Random Rectangles
pf3

Partial preview of the text

Download Exploring Sampling Distributions with Fathom using Random Rectangles and more Lab Reports Mathematics in PDF only on Docsity!

Exploring Statistics with Fathom 4: Sampling Distributions 127

In this activity you’ll use Fathom to draw samples of size 5 from a collection of

rectangles many times to explore the sampling distribution of various summary

You will need

  • Sampling

Rectangles.ftm

statistics.

EXAMINE DATA

  1. Open SamplingRectangles.ftm. The collection named Rectangles contains

100 random rectangles.

Q1 Make a plot of Area and describe the distribution of Area in terms of shape,

center, and spread.

INVESTIGATE

Taking a Sample

  1. With the collection selected, choose Collection Sample Cases. By default,

Fathom takes a sample of ten cases with replacement and places them in a new

collection named Sample of Rectangles. You’ll change this to five cases without

replacement.

Notice that animation

is on by default. You

may want to change

this later.

  1. Double-click the sample collection to

show its inspector. On the Sample

panel, change the settings to match

these. Click Sample More Cases.

Remember that you

can drag an attribute

from the Cases panel

of the inspector.

  1. You now have a sample of 5 areas from

the collection of all 100 rectangles. Make

a dot plot of the attribute Area for your sample.

Choose Graph  Plot

Value and type

!"#$%&. Then do

the same for the other

values.

  1. You need to compute some summary statistics

for your sample. Plot the values for the mean,

median, and maximum of your sample.

Q2 In a moment, you are going to repeat this

sampling, say, 200 times, to create sampling

distributions of these summary statistics (or

measures). But first, sketch distributions to

predict what you will get for the set of 200 mean, median, and maximum areas.

128 4: Sampling Distributions Exploring Statistics with Fathom

Collecting Measures

Now you need to define the measures and collect them for several samples.

  1. Double-click the sample collection to show its

inspector. On the Measures panel, define the

three measures shown. The values in the

inspector should be the same as those plotted

on your graph.

You can see the samples being taken by watching the dot plot.

  1. Select the Sample of Rectangles collection and choose Collection

Collect

Measures. You should see Fathom take five samples from the Rectangles

collection. Each time, Fathom places the measures in a new collection named

Measures from Sample of Rectangles.

  1. Double-click the Measures from Sample of Rectangles collection to show its

inspector. Go to the Cases panel. Confirm that the attributes are the measures

you defined in step 6.

  1. Make three histograms, one for each of the attributes in Measures from Sample

of Rectangles.

Notice that Replace

existing cases is off

by default, so you

need only 195 more

measures to make a

total of 200 measures.

  1. Five samples don’t make a very good

distribution. Show the inspector for the

Measures from Sample of Rectangles

collection. Go to the Collect Measures

panel and change the settings to match

these.

You can also select

the measures

collection and choose

Collection  Collect

More Measures.

  1. Click Collect More Measures. Even with the animation off, it will take some

time to collect another 195 samples of size 5. You can see Fathom taking

these samples from the Rectangles collection and calculating the measures by

watching the histograms you created in step 9.

Q3 Sketch your histogram of the sampling distribution for each measure.

continued