Corrected for Spring 2015 on April 7, 2015

Math 1040 Skittles Project Worksheet

For your own single 2.17-ounce bag of Skittles, record the numbers in the table below.
Number of
red candies

Number of
orange candies

Number of
yellow candies

Number of
green candies

Number of
purple candies

13

13

16

11

Using the data compiled from the entire class, record the following information:
The total number of candies in the sample = _____920_____
Number of

Proportion

red candies

Number of
orange
candies

Number of
yellow
candies

Number of
green candies

Number of
purple candies

197

155

208

179

181

.214

.168

.226

.195

.197

Throughout this entire project, use decimals rounded to three places

for all of your proportions. Do not use percents.

The total number of candies in your own single 2.17-ounce bag of Skittles = ___62___
The total number of bags in the sample collected by the entire class = ____15____
The total number of candies in the sample collected by the entire class = ____920_____
For the entire sample:

x = ___61.3____ (the mean number of candies per bag rounded to 1 decimal place)
s = ____3.67___ (the std. deviation of the number of candies per bag rounded to two decimal places)

5- number summary: (round to one decimal place where necessary)

Min: 6. Max: 25. Q1: 9. Q3: 15. Median: 12.
Fill in the appropriate values on this page and keep it handy as you do your calculations.

Quick Reference for Confidence Intervals

For the interval estimate of the proportion of purple candies:
n = __920___

^p = ___.197__

x = __181___

= __.1___

For the interval estimate of the mean number of candies in a bag:

x = __61.333___

n = __920___

= __.1___

For the interval estimate of the standard deviation of the number of candies in a bag:
n = __920___

s = __3.67___

= __.1___

X 2R = 7.34

X 2L =

7.34

Quick Reference for Hypothesis Tests

For testing the claim that 20% of Skittles are green:
n = __920___

^p = __.195___

x = __179___

H 0 : x = .20

= __.1___

H 1 : x =/= .20

FAIL TO REJECT--the sample size is small compared to all Skittles, so the numbers are well within the
borders of not rejecting.

For testing the claim that the mean number of Skittles in a 2.17-oz. bag is 56:
n = ___920__

H0 :

x = 56

x = __61.333___

H1 :

x =/= 56

= __.1___

REJECT--Despite the small size, the standard deviation is large enough to cause much uncertainty. Most
bags seem to be well above 56 as well.

INTRODUCTION
This project is about determining statistical properties of Skittles within bags bought by each
student in the classroom. Students counted their own bags, submitted the data to Canvas, and accessed a
Google Sheet to get data from each student. Data was used comparing the individuals Skittles to those of
the whole class. For most of the project, we determined such things as average and standard deviation of
the entire classs sample. We used the data to determine the likelihood of a certain percentage of Skittles
being a particular color. All this math was done with the aid of a calculator for adding up and finding
statistical properties.
DISPLAY SKITTLES DATA
(Class)

These graphs are about what is expected: slightly off twenty percent for each color of candy.
Several of the amounts are similar. There is an understandable difference between the minimum and
maximum, but nothing very big. This doesnt quite agree with my proportions within my own bag. That
data is shown below.
(Self)

A graph of the distribution for each of these data sets would be bell shaped and roughly symmetrical,
assuming the order of candies was correct. However, since Skittles are not inherently ranked in any way
affecting their production (it doesnt appear they are made in significantly different proportions, and that
any combination landing in a bag is chance, yet likely to be somewhat evenly distributed), that
distribution would need to be determined while organizing data sets. For these sets, the data is all above,
on the first two pages.
REFLECTION
This assignment taught me how to apply statistics in my life. While it is not likely I will be
finding data sets for candy all that often, I know how I can. This can be useful in ordering party supplies
or other things for groups of people. It is pretty tedious, though, so I would rather just take surveys and
determine roughly how many things to order. It would be nice if the universe followed mathematical
laws closely all the time, but so often other things happen, so the people using statistics are mostly those
polling people for political functions and such. This was a good insight into their procedure in processing
raw data.