Sunteți pe pagina 1din 5

Jenna Becker

Skittles Project
Introduction: This project is to determine the data collected by the class regarding
skittles candies. It will show how many of each candy was in each bag overall, the
frequency of each color of candy, the frequency of how many candies where in each
persons bag, distribution of the amount of candy and so forth. It is to help the
students put what we have learned into a real life scenario to use our statistic skills.

The data in the graphs came out to


be about what I had expected
because each color came up in
about the same amount. I expected
each color to have about the same
number as the next. The data from
the rest of the class was similar to
my own data because each person,
including myself has just about 60
candies in our bags. The most
variation in the data was how many
of each specific color each
individual bag had inside it.
My Data

Red: 8

Yellow: 10

Purple: 15 Green: 13

Entire
Sample

Red: 282

Yellow:
273

Purple:
294

Green:
305

Orange:
13
Orange:
339

Total: 59
Total:
1496

Mean: 59.8
Standard Deviation: 2.89
Q1: 59
Median: 60
Q3: 61.5
Minimum: 53
Maximum: 68

The graphs have a general bell shape which I


expected to see from looking at the data and
seeing that most bags has around the same
amount of candies. I am however surprised
to see that there are some outliers in the
data, I was not expecting that because we all
got the same size bags of candy and
therefore they should all have generally the
same amount of candies. Overall, the data
collected from the class agrees with the data
from my own bag of candy.

My Bag: 59 Candies
Total # of bags in the sample: 25

Categorical data is data that can be found with summary and cannot really be
ordered in any specific way, like the colors of candies in the bag. Quantitative data
is numerical data that can be organized like the number of each color of candy in
that bag. Graphs that would not make sense for categorical data are histograms and

box blots because they rely on quantitative values to create them. The types of
graphs that would not make sense for quantitative data include pie and pareto
charts because they are made b organizing categories of data rather than
mathematical data. Calculations for categorical data would be stuff like addition and
subtraction and some that wouldnt make sense would be finding the standard
deviation and 5 number summary. For quantitative what makes sense is finding the
5 number summary and standard dev. Because it can give us more information
about the data and its outliers. Something that would not make sense would be to
had them or subtract them because it would give you not useable information.

A confidence interval is a range of values that described the uncertainty


surrounding an estimate. It is there to tell people looking at your sample how
confident you are that your sample shows correct values and data.

95% confidence interval results: Purple Candies


Count: 294
Total: 1496
Sample Prop: 0.19652406
Lower Limit: 0.17638791
Upper Limit: 0.21666022

99% confidence interval results: True Mean # of Candies per Bag


Mean: 59.8
Standard Dev: 2.89
N= 1496
Lower Limit: 59.608
Upper Limit: 59.992

98% Confidence Interval Results: Standard Deviation


Sample Variance: 8.39
DF: 24
Lower Limit: 4.6849894
Upper Limit: 18.547651

These confidence intervals show a lot to us. The first one show us that the
percentage of purple candies in a bag varies from about 18 percent to about 22
percent, 95 percent of the time. The second shows us that the mean is 59 to about
60 candies per bag 99 percent of the time. The last one shows us that the standard
deviation should be around 5 to 19, 98 percent of the time.

Hypothesis testing is an inferential procedure that uses sample data to evaluate the
credibility of a hypothesis about a population. They are uses to create the possible
strategic outcomes that could occur.

Hypothesis test results:


p : Proportion of successes
H0 : p = 0.2
HA : p 0.2
Proportion Count Total Sample Prop. Std. Err.

Z-Stat

P-value

0.37488858

0.7077

305

1496 0.20387701

0.010341754

Reject the Claim that 20 percent of skittles in a bag are green.

Hypothesis test results:


: Mean of variable
H0 : = 56
HA : 56
Variable
Sample Mean
# of candies
59.84

Std. Err.
DF
0.57930993 24

T-Stat
P-value
6.6285761
<0.0001

Reject the claim that the mean number pf candies in a bag is 56.

Reflection:
This project really challenged me and helped me a lot in statistics. I had to work
very hard to understand some of the equations and use them in real life situations
like this one. I had to work with stat crunch and my TI-84 to get the desired outcome
of my equations and tables. The most important part of this project to me is that is
applied so much to the real world. It shows you how to use programs (like excel and
stat crunch) that you may have to use in your future career. It shows how to sort
and categorize data to stay organized and on top of your work.
Overall I would say I enjoyed this project. At first I found it tedious and annoying. I
couldnt understand how counting a bag of skittles was going to help me learn

statistics but it really did! The way it made me apply myself and my math skills
really helped me a lot! I cant wait to use these new skills in my future classes and
jobs.

S-ar putea să vă placă și