Scribd
Upload
186 views

Chi Lab

This document summarizes a chi-square analysis performed on the observed and expected color distributions of M&Ms in bags. The null hypothesis stated that any differences between observed and expected color ratios were due to chance. Observed color counts from a single bag and aggregated class data both showed significant differences from the expected ratios. The chi-square calculations for both datasets exceeded the critical value, so the null hypothesis was rejected. Possible explanations for the non-random differences included uneven distribution by packaging machines or changes in machine settings over time. Further suggested studies included direct observation of packaging machines and testing other randomly distributed consumer products.

Uploaded by

api-319923127
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
186 views

Chi Lab

This document summarizes a chi-square analysis performed on the observed and expected color distributions of M&Ms in bags. The null hypothesis stated that any differences between observed and expected color ratios were due to chance. Observed color counts from a single bag and aggregated class data both showed significant differences from the expected ratios. The chi-square calculations for both datasets exceeded the critical value, so the null hypothesis was rejected. Possible explanations for the non-random differences included uneven distribution by packaging machines or changes in machine settings over time. Further suggested studies included direct observation of packaging machines and testing other randomly distributed consumer products.

Uploaded by

api-319923127
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Hallie Ryan

Lab Partners: Katie Alpert, Conor Wright, Quinn Meistrich


The Chi-Square Analysis Used on Expected and Observed M&M Statistics based on
Distributed Colors per bag of M&M

Introduction:
A Chi Square Analysis has several important components to it, a major one being
the null hypothesis. A null hypothesis is a prediction that something is not present, or in
other words, an observed pattern and an expected pattern of data are the same, and are
only different by chance, not because they just are different. For Chi Square analysis the
null hypothesis is always constant. In this analysis, the goal is to confirm or revoke this
null hypothesis.
In this Chi Square Analysis, M & M statistics will be used to confirm or revoke
the null hypothesis. To accept the null hypothesis, the data must be consistent with what
would be expected and any slight variance is due to chance; however, the hypothesis
would be rejected if the data is too different from what was expected, and is not due to
chance. To find this, the Chi-Square value will be calculated, the degrees of freedom (the
number of classes/color categories minus one) will be found and then using the
probability chart, it will be determined whether the null hypothesis is confirmed or
rejected.
Experimental Question:
How does M&M color ratio in a bag of M&Ms vary from the given statistics
versus what is observed?
(Null) Hypothesis: If there is any change from the given data (the M&M statistics), then
it is due to chance.
Material List: 1 bag of M&Ms, paper towels, clean hands
Procedure:
1. Wash your hands with soap and water.
2. Open the bag of M&Ms and pour then carefully onto the paper towels
3. Separate the M&Ms into color categories
4. Count the amount of M&Ms in each color pile made
5. Record the numbers counted onto a data table
6. Calculate the expected number for each color M&M using the statistics given,
record them
7. Determine the Chi Square value

Results

Discussion and Conclusion


If there is any change from the given data (the M&M statistics), then it is due to
chance. In the experiment there was a change from the given data; however, it was not
due to chance. Looking at the degrees of freedom in this Chi Square analysis, there were
5 degrees. Then, using that degree of freedom and the probability of .05, in order to
accept the null hypothesis, the Chi Square analysis would have to be below 11.07. In the
1 package of M&Ms and the entire classes data of M&Ms, both were above the value of
11.07, being 24.001 and 13.87. This means that the change from the given data was not
due to chance. So, the null hypothesis should be rejected because the P-value was less
than the chi-square value).
Firstly in the individual experiment, the observed amounts of different colored
M&Ms changed a lot from the given/expected data. The Brown M&Ms varied the least,
the expected was 48.62, and observed was 48. However, with all the other colors, there
was a much more apparent variation. The Blue expected was 89.76 and observed was 109,
Orange expected was74.8 and observed was 67, Green expected was 59.8 and observed
was 82, Red expected was 48.6 and observed was 34 and Yellow expected was 52.4 and
observed was 34. This shows that the data observed was very different from the given
data. The total of both the expected and observed however were the same, both equally
374. Then, when using the Chi Square equation to find the value 24.001, it is clear the
null hypothesis must be rejected due to the fact that it is above the P-value of 11.07.
Then in the entire class experiment, the observed amounts of different colored
M&Ms still changed from the given/expected data, giving the same results. In this case,
the least varying color was Red, which had an expected value of 293.54 and an observed
value of 281. With the other colors, there was a greater fluctuation in the expected vs.
observed colored M&Ms. The Blue expected was 548.4 and observed was 532, Orange
expected was 457 and observed was 474, Green expected was 361.28 and observed was
408, Brown expected was 297.05 and observed was 315 and Yellow expected was 316.12
and observed was 276. The totals were also different, expected being 2273.39 and
observed being 2285. Then, when using the Chi Square equation to find the value 13.87,
it is clear the null hypothesis must be rejected due to the fact that it is above the P-value
of 11.07 again.
In both the individual data and the entire class data, the Chi Square analysis
proved the null hypothesis to be incorrect and not due to chance but due to other factors.
Some explanations for the for the outcomes could include that the machines distributing
the M&Ms did so unequally or there was a change in the machine settings from when the
given data was taken to what was observed. The classes data for the expected vs.
observed totals which were 2273.39 and 2285 shows that what was actually in the bags
was more than what was said by the Mars Company. Although that is only a 12 M&M
difference, it shows that some of the bags had a distribution problem and not only a
coloration problem. The total of the individual bag observed had the same amount in
observed and expected which means there was not a number of total M&Ms distributed
issue.

Further Inquiry
Some possible research that came out of this experiment could be seeing how
machines distribute the M&MS and how the bags are being measured for the amount of
M&Ms in each one and the different colors in each one. Also on a larger scale, testing
other products to see how accurate their percentages (expected) to what is actually in
their bags, like Starbursts for example. For future study, more bags can be tested and new
given percentages can be given to the bags of M&Ms to make it a more accurate
depiction of what is really in each bag.

You might also like