Scribd
Upload
1K views

Skittles Project 1

- Donna Burrows completed a statistics project for her Math 1040 class where students counted and categorized Skittles candies from individual bags. - She found that the proportions of colors in her bag matched the overall class proportions. However, she noticed there were more purple candies than expected. - Confidence intervals and hypothesis tests were conducted on the data. The results supported claims about the proportions of colors and averages numbers of candies per bag. - Burrows reflected that the project helped develop her critical thinking and calculator skills, and showed her how statistics apply to everyday items like candy.

Uploaded by

api-299868669
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
1K views

Skittles Project 1

- Donna Burrows completed a statistics project for her Math 1040 class where students counted and categorized Skittles candies from individual bags. - She found that the proportions of colors in her bag matched the overall class proportions. However, she noticed there were more purple candies than expected. - Confidence intervals and hypothesis tests were conducted on the data. The results supported claims about the proportions of colors and averages numbers of candies per bag. - Burrows reflected that the project helped develop her critical thinking and calculator skills, and showed her how statistics apply to everyday items like candy.

Uploaded by

api-299868669
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 7

Donna M.

Burrows
Professor: Thomas Sanborn
Math 1040 Skittles Term Project

The Skittles project was to show the class how real life Statistics works. Everyone in the
class was given a 2.17 ounce bag of Skittles. We had to count how many Skittles were in the bag
and to count them by colors. Each bag had red, orange, yellow, green and purple candies in the
bag. Then we had to figure out each proportion that we had of each color. It was amazing to see
that the proportions of my bag added up to 1.
I used a TI-84 plus Silver Edition calculator to obtain values by using the stat button on
the calculator and then finding #1Edit on the calculator and pressed enter button to have L1 and
L2 graphs come up and entered the data into the graphs then pressed stat button again and used
arrow keys to calculate and scrolled down to #2 Variable Stats to obtain my answers for the
entire class sample. The data presented here was the 5 number summary; such as the mean, min,
Q1, med, Q3, and the max. I was also able to find the standard deviation as well.
The goal of this assignment was to see how many candies were in a bag of skittles and
how much the candies deviated from the norm. The data was also used to find the average
amount of candy in each bag. It was also a way to see what categorical data we had, such as the
colors of the candies.

Table showing class total proportions and my bag of skittles and its proportions.
Color
red
orange
yellow
green
purple

Class Total
0.18
0.189
0.203
0.17
0.258

My Bag
0.194
0.129
0.242
0.145
0.29

This did show what I expected to see because the class total and my bag totaled up to 1.
The only thing I noticed that I didnt expect to see was there are more purple candies according
to the Pareto chart. I didnt really notice this with the other graphs and the Pareto chart was
clearer about the amount of purple candies within the sample.

Its not quite like a usual bell shaped curve and there is an outlier that skews the graph to
the right. The boxplot data also has an outlier noted at 66. Both graphs are somewhat comparable
to each other. My bag of candy falls in range with the class data on numbers of candies per bag.
Summary statistics:
Column
Total of candies

n
15

Mean
61.

Std. dev.
1.4

Median
62

Min Max Q1
60

66 61

Q3
62

The number of candies in my Skittles bag was 62 and the


Total sample of candies was 926.
Categorical data has more to do with the description of something like colors, where
quantitative data has more to do with numerical figures.
The histogram and the boxplot graphs wouldnt make sense for the categorical data. The
pie chart works better for the categorical data because its easy to see the divisions between
colors.
The boxplot and histogram works better for the quantitative data. These two graphs focus
on the key numerical data. They also allow you to actually visualize the content of the graphs.
The type of calculations that make sense for categorical data are sample totals and
proportions because they appear to translate better to the visualization of the categories.
For the quantitative data it makes more sense to use calculations such as: The mean,
standard deviation and the standard number summary. This gives us information about the data
and how much the data actually deviates from what the average is supposed to be in a given
product.

Confidence Interval Estimates:


A Confidence Interval is the true value of a population parameter. This is done by using a range
of values to estimate the true value of the population parameter.

99% confidence level estimate for the true proportion of yellow skittles candies.
.169<p<.237
95% confidence interval estimate for the true mean number of candies per bag.
60.925<mu<62.475
98% confidence interval estimate for the standard deviation of the number of candies per bag.
.970<s<2.427
We are 99% confident that the interval (.169, .237) contains the true population proportion of
yellow skittles. We are 95% confident that the interval (60.925, 62.475) contains the true
population mean for the number of candies per bag. We are also 98% confident that the interval
(.970, 2.427) contains the population standard deviation of the number of candies per bag.

The purpose of a hypothesis is to test a claim someone has about something in a population. This
means it is to determine if something is actually true, or not.

Testing the claim that 20% of all skittle candies ae orange.


Ho: p=0.2, p=.402>0.05 so fail to reject null hypothesis, There is not sufficient evidence to reject
the claim that 20% of all skittles are orange.
Testing claim of mean candies per bag is 57.
Ho: mu =57, mu =3.319*10^-9 which is < 0.01 so reject null hypothesis. Which means there is
sufficient evidence to reject the claim that the mean is 57 skittles per bag.
The conditions for doing confidence interval estimates are a simple random sample. Conditions
for binomial distribution is there are a fixed number of trials. The trials are independent and there
are two outcomes. The probabilities remain constant. There are at least 5 successes and at least 5
failures and are both satisfied.
Our confidence intervals met the conditions. It was a simple random sample collected by the
class. There was a fixed number of 15 trials. The trials were independent of each other. There
were two outcomes for example there was either yellow candy or no yellow candy. The
probability of getting a yellow candy remained constant. There was 188 yellow candies and 738
that were not yellow. So this makes at least 5 successes and at least 5 failures.
The conditions for doing a hypothesis test is a sample observation of a simple random sample.
The condition for a binomial distribution are satisfied. np_> 5and nq _>5 are both satisfied.
Our hypothesis testing met the requirements. It was an observation of a simple random sample
collected by the class. It was a fixed number of 15 trials. The trials were independent of each
other. There are two outcomes for example it was an orange candy or not an orange candy.
926*.2 = 185.2 _>5 and 926*.8=740.8_>5. So this means np_>5 and nq_>5 are both satisfied.
Mistakes could happen by miscounting and partial skittles found in any of the skittles bags. Too
small of a sample could also lead to mistakes. Another mistake could be by a type 1 error which
is the mistake of rejecting the null hypothesis when it is actually true. A type 2 error could have
been made which is when you fail to reject the null hypothesis when it is actually false.
Sampling could be improved by increasing the sample size, or using mechanical ways of
counting.
In conclusion with the skittles test, even though some of the colors had higher proportions. It is
likely that all colors are equally likely to be found in a skittles bag. This is based off the testing
used by interval estimates and hypothesis we used in this experiment.

Reflective Writing /e-portfolio

Donna M. Burrows
Professor: Thomas Sanborn
Statistics
Skittles project reflection
This project had to do with real word statistics and I found that I really had to use a lot of
critical thinking skills. These skills might have an impact on me if I was going into engineering
or another heavy related science field. I am already in the field of nursing and I do use critical
thinking skills on a daily basis. However, I dont use this kind of math for what I do every day.
The project helped me develop some new skills such as: using a graphing calculator and
being able to construct a scatter plot and understand what it means. I also learned you have to
pay attention to detail in this class. Its important to use the formulas and to be careful about
what formulas you use for the correct problems. I needed to understand the use of my calculator
much more than I did. Its important to have other computer skills as well.
The project helped me follow directions a little closer, since I had to resubmit the
assignment twice because of mistakes I had made. The project helped me see the use of statistics
in things I never thought of such as: The average amount of candy in a bag of skittles, or M&Ms
or Starbursts for that matter. It would be interesting to see how many red Starburst are in a
regular bag of Starburst candies as well.
My view on real world statistics would be on someones studies and whether their
hypothesis are true or whether we should reject them. These are some of the things that I need to
improve on myself.

You might also like