By: Rochelle Cooper, Jon Hale, and Ainsley Hume • It was created in 1963 by the Vice President of the General Mills Company, John Holahan • It was created, at first, by taking orange marshmallow peanuts, cutting them up, and sprinkling them over cheerios • Pink hearts, yellow moons, orange stars, and green clovers were the first in the box. Next came blue diamonds, purple horseshoes, followed by red balloons. After came rainbows, pots of gold and Leprechaun hats • All these combined to make the delicious cereal Reason for picking topic • We decided that we wanted to measure the proportion of items in a type of food • Had to be able to measure it accurately (not extremely hard) • All like Lucky Charms cereal Boxplot for initial weight of Lucky Charms Mini Boxes •Weight in grams •Weight on box= 48.19g •Min.=61.28g •Q1=62.9g •Median=63.89g •Q3=65.89g 62.9g 61.28g 65.89 71.73g 63.89g 60 62 64 66 68 70 •Max=71.73g •IQR=Q3-Q1=(65.89g)-(62.9g)=2.99g 72 74 Outliers for initial weight of Lucky Charms Mini Boxes •IQR=Q3-Q1=(65.89g)-(62.9g)=2.99g •Outlier test Q3+[IQR(1.5)]=High limit:70.375g Q1-[IQR(1.5)]=High limit:70.375g 1 outlier: #25, 71.73g 62.9g 61.28g 65.89 68.04g 71.73g 63.89g 60 62 64 66 68 70 72 74 Boxplot for proportion of marshmallows in Lucky Charms •Min.=.1456 Mini Boxes •Q1=.1898 .146 •Median=.2421 19 .276 .242 •Q3=.2464 •Max=.3777 .1 .2 •IQR=Q3-Q1=(.2464)-(.1898)=.0566 •Outlier test Q3+[IQR(1.5)]=High limit:1.0954 Q1-[IQR(1.5)]=High limit:-.6592 No outliers .377 .3 .4 Histogram for proportion of marshmallows in Lucky Charms Mini Boxes •X-axis:proportion of marshmallows •Y-axis:frequency •Right skewed • x =.24 11 •Range=.24 0 .14 .16 .18 .2 .22 .24 .26 .28 .3 .32 .34 .36 .38 -6 Proportion of Marshmallows Scatterplot for proportion of marshmallows in Lucky Charms Mini Boxes •Slightly positive direction •Moderately weak •Linear 25 20 15 10 5 40 45 50 55 60 65 Mini Boxes total Weight in grams Assumptions for 1-Proportion Z-Test 1.SRS 2.np 10 n(1-p) 10 3.pop 10 n 1.assumed 2.1749(.272) 10 1749(.728) 10 3.pop 10(1749) 1-Proportion Z-Test Ho: p=.272 Ha: p .272 Z= pˆ p = -1.9198 p (1 p ) n 2*P(z<-1.9198)=.054879 We fail to reject the Ho because p>.05= . We have sufficient evidence that the proportion of marshmallows is equal to .272. Assumptions for T-Test of Marshmallow Weight 1.SRS 2.Normal population or n 30 1.assumed 2.34 30 T-Test of Marshmallow Weight Ho : 13.1087 Ha : 13.1087 x 12.6562 n 34 x t 1.1675 s n 2 P (t 1.1675) .25137 We fail to reject the Ho because p>.05= . We have sufficient evidence that the mean marshmallow weight is equal to 13.1087 grams. Assumptions for T-Test of Serving Size Weights 1.SRS 1.assumed 2.Normal population 2.34 30 or n 30 T-Test of Serving Size Weights Ho : 49 Ha : 49 x t 8.041984 s n 2 P (t 8.041984) 2.7987 10^ 27 We reject the Ho because p<.05= . We have sufficient evidence that the mean serving size weight is not equal to 49 grams. Confidence Interval x t * ( s n ) (51.251,52.775) We are 95% confident that the mean serving size weight is between 51.251 and 52.775 grams. Bias • Packaging bias • Lack of mini-cereals in grocery stores – Not many, plus only stocked in Genardi’s • Bias during weighing – Scale might not be exact – Losing pieces of cereal • Calculating population proportion – Had to round up for 1-proportion ztest • Bag added extra weight Conclusions • The marshmallows were close enough to the mean weight, 13.1087 grams. • The cereal was not always the right weight – Generally over the mean weight…good for us! • The proportion of marshmallows to cereal was close enough to the mean proportion, .272. – However, at .01 alpha level, we would reject the Ho. Our Conclusions • The mean serving size weights seemed to be very spread out. This was surprising as I would expect the company to keep it close to or under the mean weight of 49 grams • Visually, thought the weights would be different because the marshmallows in the containers looked not as appetizing as the marshmallows in the box • Surprised at how high the outlier was compared to the mean weight of 49 grams