Skittles Project Math 1040

Zachery Gallman Pd. 7

Report Intro:

In this project we took a bag of skittles just the plain, old, regular, one from the store. We

counted the number and amount of different colored skittles in the bag, quite an arduous process.

Being able to use my personal data that I collected as well as the other class data I will be

able to estimate the true number of red skittles per bag. Part of the reason is that we are using the

class numbers for the confidence interval in order for us to more fully interpret accurately the

total number of different skittles per bag. One of the main reasons for this is because of factory

error not every single bag is going to have the correct number of different colors of skittles per

bag. Having the whole class take their data is important because it allows for us to see a mean

interpretation of what possibly is a correct estimate .

Data Collection:

Number of Number of Number of Number of Number of

Red Candies Orange Yellow Green Purple

Candies Candies Candies Candies

Class Data 369/30= 339/30= 410/30= 358/30= 385/30=

Average : 12.3 11.3 13.67 11.93 12.83

Confidence Interval Estimates:

A confidence interval is an interval estimate combined with a probability statement. This

basically means that if we used the same sampling method to select different samples and

compute an interval estimate for each sample, we would expect the true population parameter to

fall within the interval estimates 99% of the time if that was our given confidence interval.

1. We are trying to estimate the true proportion of given colors of skittles. Our best guess is

p hat=.25 because of sampling variability, we are unlikely to be correct. So, we will

calculate a 99% z-interval for p.

2. Conditions.

a. Random sample (yes)

b. Independence condition (yes)

Assume population greater than 376

c. Normality condition? No but we will hopefully assume the population is normal

and will proceed with caution.

.1404 .25 .3596

4. Thus, I am 99% confident that the interval from .1404 to .3596 captures the true

proportion of red candies.

Construct a 95% confidence interval estimate for the true mean number of candies per bag

1. We are trying to estimate the true mean number of candies per bag. Our best guess is

mean= 60.548 but because of sampling variability, we are unlikely to be correct. So,

we will calculate a 95% t-interval for the average.

2. Conditions

a. Random Sample- Yes

b. Sample <5% of population? Yes

c. Normality-large sample size? Yes the population is greater than 30.

3. 95% CI = x− +− t s/sqrt(n) = ( 59.197 , 61.8997 ) df=32

51.1971 60.548 61.8997

the true average of the true mean number of candies per bag.

Because in the first confidence test that we did there was a 99% confidence that was

between (.1404, .3596) as the confidence interval and proportion of yellow skittles to the total

skittles. Because there were 5 different colors you would expect a 25% proportion which clearly

falls in the range.

Also the next interval represents the amount of total skittles that you would expect to see

in the full package. Based on the information provided in the sample we took there was 95%

confidence that the number would fall between (59.1971, 61.8997). The samples that have been

collected show this as the average of 60.548 which is in it.

Hypothesis Tests:

The purpose of hypothesis testing is to determine whether there is enough statistical

evidence in favor of a certain belief, or hypothesis, about a parameter. These

questions/hypotheses are similar in spirit to the discrimination example studied earlier.

We are using a= 0.05 significance level to test the claim that 20% of all Skittles candies

are red.

1. At first glance, it appears that the true proportion of red skittles is greater than .2

since p^ = .25. However, it is also possible that the true proportion is p=.2 and
 we will conduct a sample proportion this high/low because of sampling

variability. To decide, we will conduct a ​1 sample z test for p​ ( a=.05).

2. Ho: p= .20 Ha p blank not equal to .20

3. Conditions:

a. Random sample of skittles

b. Independence n<.05N? Yes

c. Normality? Yes

2 P (> or < _.5186____ ) = 2P nppP porz)1(ˆ = 2P (z > or < __ − .6455___) ___ −

z=-.6455 p=.5186

4. Since the p-value is < a, we reject the null hypothesis.

Use a 0.01significance level to test the claim that the mean number of candies in a bag of Skittles

is 55

Step 1: The original claim is that 20% of all Skittles candies are green. p = 0.20

Step 2: The opposite of the original claim is p = 0.20

Step 3: The null hypothesis is p = 0.20 and the alternative hypothesis is p not equal to 0.20

Step 4: The significance level = 0.01

which makes this a normal distribution.

Step 6: The test statistic t = -3.29 is calculated as: p-value = 0.001 < 0.01

Step 7: Because the p-value is less than the significance level of = 0.01 the null hypothesis is

rejected.

Step 8: From this hypothesis test, because the null hypothesis was rejected there is sufficient

evidence to warrant rejection of the claim that 20% of all Skittles candies are green.

For the first hypothesis test that we looked at a problem that wanted to test the claim that

20% of all skittles candies are red. We used a .05 significance level to find this. The probability

that this was possible was .5186 this being able to show that it was about 50% possible that 20%

were red.

The second test and test to the claim was about the mean number of candies or skittles

per bag is 55. Not only did I see that using the TTest and the t curve would help me get the

degrees of freedom necessary in order for me to finish the things I need in order to solve this and

it allowed me the necessary skills to solve these problems.

Reflection Time:
Some of the things that I have learned as a result from this paper are that statistics is hard,

but that you can use it in real life situations. Although I’m not sure that I would ever have to

i know how many candies are in a Skittles bag but it made the learning fun and different. I was

surprised to learn that not all candy is created equal. Some bags of candy had more than others,

which makes me wonder what else in life may be unequal.

My initial conclusion to this project’s data seemed to be that since the colors red, orange,

and purple were the highest number of candies per each bag that these colors all rely on red as

the primary pigment, suggesting that red is the most used color additive in making this candy.

This seems to be the case throughout this study. This proves that you really can taste the rainbow

or at least Red #5.

Other things that I learned from this project are how to use the Statcrunch program and

how to graph, which may come in handy in the future when I prepare graphs for presentations

and papers. I am surprised at how helpful the graphs were by seeing a visual depiction of the

information, which helped me to understand the information given and that it’s not all just

numbers and stats, but what they represent.

I’ve also been able to learn critical skills that I’ll be able to take to the next level and especially

in my career choice in the United States Army. Being able to have problems and use critical

thinking and data analysis to solve them is important in any situation but especially when dealing

with issues that are in normal life. Also, relying on case studies and statistics, I’ve learned is not

always correct; that they may have used voluntary responses to collect their data or may have

distorted the data somehow by misrepresenting or inaccurately showing all of the information.
This will help me in being more careful in where and how I choose to get my statistical

information in the future.

It will also help me in being better at consuming news media and being critical at where that

information is coming from, and not buying into every little statistic as fact, but being more

judicious in where that information is coming from.

Most everything in this class was all new information to me. I felt I struggled and

stretched the limits of my brain many times, but I feel like this project helped me understand it a

little bit better and will help me in the future as a journalist, that I may rely on true and accurate

information and not information that is biased or misrepresenting in anyway, but also to test

those statistics with my newly gained skills from this class.