Caffeine & Calories in Soda. Statistics. Anthony W Dick

Similar documents
Feasibility Study of Store-Brand Sodas

STAT 201 Chapter 3. Association and Regression

AP Statistics Practice Test Ch. 3 and Previous

Chapter 3: Describing Relationships

Chapter 3 CORRELATION AND REGRESSION

Section 3.2 Least-Squares Regression

Rhino Soft Serve. Island Oasis Smoothie. Nutrition Facts Strawberry. Nutrition Facts Wildberry

Unit 1 Exploring and Understanding Data

STATISTICS INFORMED DECISIONS USING DATA

Beverage Density Lab

Results & Statistics: Description and Correlation. I. Scales of Measurement A Review

3.2A Least-Squares Regression

Simple Linear Regression the model, estimation and testing

10. LINEAR REGRESSION AND CORRELATION

3.2 Least- Squares Regression

Chapter 4: More about Relationships between Two-Variables Review Sheet

REVIEW PROBLEMS FOR FIRST EXAM

Evaluating Energy Drinks: A Feasibility Report. Paula Bendet Austin Greenameyer Scott Hale Peter Khaya

Which could be made worse by the over-consumption of sugar or calories?

Business Statistics Probability

Chapter 4: Scatterplots and Correlation

Coke Floats (Or Does It?)

To Drink or Not to Drink: The Importance of Nutritional Beverages

Problem #1 Neurological signs and symptoms of ciguatera poisoning as the start of treatment and 2.5 hours after treatment with mannitol.

Further Mathematics 2018 CORE: Data analysis Chapter 3 Investigating associations between two variables

Chapter 3: Examining Relationships

ice cream pint halo top pint vanilla bean vanilla bean Chocolate Halo Chocolate Ice Cream Pint Top Pint

Major Food Allergens listed in Red. Not all items are served in all stores

1.4 - Linear Regression and MS Excel

Math 075 Activities and Worksheets Book 2:

INTERPRET SCATTERPLOTS

SCATTER PLOTS AND TREND LINES

Section 3 Correlation and Regression - Teachers Notes

STATISTICS & PROBABILITY

A response variable is a variable that. An explanatory variable is a variable that.

Power House: Youth Experience Session 11

Diurnal Pattern of Reaction Time: Statistical analysis

Whose Choice is it Anyway? Montague's Experimental Results

Lecture 6B: more Chapter 5, Section 3 Relationships between Two Quantitative Variables; Regression

14.1: Inference about the Model

MULTIPLE LINEAR REGRESSION 24.1 INTRODUCTION AND OBJECTIVES OBJECTIVES

Unit 8 Bivariate Data/ Scatterplots

EXECUTIVE SUMMARY DATA AND PROBLEM

Beware of Confounding Variables

Unit 8 Day 1 Correlation Coefficients.notebook January 02, 2018

By: Denise Lord, Abbey Moore, Jeffery Coffman, Lauren Lofgren, Giovanni Ellzey

5 To Invest or not to Invest? That is the Question.

Choosing a Significance Test. Student Resource Sheet

Chapter 1: Exploring Data

Nutrition Facts. Calories from Fat. Total Fat (g) Calories

NORTH SOUTH UNIVERSITY TUTORIAL 2

Caffeine Consumption and Anxiety Levels: A Convenience Sample. Operation Righteous Cowboy Lightning: Ian Sande, Johnny Rider,

M 140 Test 1 A Name SHOW YOUR WORK FOR FULL CREDIT! Problem Max. Points Your Points Total 60

CP Statistics Sem 1 Final Exam Review

AP STATISTICS 2010 SCORING GUIDELINES

UNIVERSITY OF TORONTO SCARBOROUGH Department of Computer and Mathematical Sciences Midterm Test February 2016

Examining Relationships Least-squares regression. Sections 2.3

WDHS Curriculum Map Probability and Statistics. What is Statistics and how does it relate to you?

How Long Does It Take a Person to Sober Up? Some Mathematics and Science of DUI

STP 231 Example FINAL

M 140 Test 1 A Name (1 point) SHOW YOUR WORK FOR FULL CREDIT! Problem Max. Points Your Points Total 75

Eligibility The NCSF online quizzes are open to any currently certified fitness professional, 18 years or older.

IAPT: Regression. Regression analyses

bivariate analysis: The statistical analysis of the relationship between two variables.

UF#Stats#Club#STA#2023#Exam#1#Review#Packet# #Fall#2013#

Commercial soft drinks: ph and in vitro dissolution of enamel

Preliminary Report on Simple Statistical Tests (t-tests and bivariate correlations)

Understandable Statistics

Chapter 14: More Powerful Statistical Methods

MULTIPLE CHOICE. Choose the one alternative that best completes the statement or answers the question. 1) 1) A) B) C) D)

Statistical questions for statistical methods

Class 7 Everything is Related

Part 1. For each of the following questions fill-in the blanks. Each question is worth 2 points.

Jonathan McIntyre. SVP R&D Global Beverages

Statistics and Probability

Homework #3. SHORT ANSWER. Write the word or phrase that best completes each statement or answers the question.

Correlational Method. Does ice cream cause murder, or murder cause people to eat ice cream? As more ice cream is eaten, more people are murdered.

Making Inferences from Experiments

CRITERIA FOR USE. A GRAPHICAL EXPLANATION OF BI-VARIATE (2 VARIABLE) REGRESSION ANALYSISSys

Multiple Linear Regression Analysis

HW 3.2: page 193 #35-51 odd, 55, odd, 69, 71-78

Lecture 12: more Chapter 5, Section 3 Relationships between Two Quantitative Variables; Regression

This means that the explanatory variable accounts for or predicts changes in the response variable.

Relationships. Between Measurements Variables. Chapter 10. Copyright 2005 Brooks/Cole, a division of Thomson Learning, Inc.

Chapter 3 Review. Name: Class: Date: Multiple Choice Identify the choice that best completes the statement or answers the question.

Introduction to regression

Table of Contents. Plots. Essential Statistics for Nursing Research 1/12/2017

Write your identification number on each paper and cover sheet (the number stated in the upper right hand corner on your exam cover).

Sugar Concentration in Sodas. Chemistry Period 3 Crater School of Business Innovation and Science Mallory Heard April 15th, 2016

STATISTICS 201. Survey: Provide this Info. How familiar are you with these? Survey, continued IMPORTANT NOTE. Regression and ANOVA 9/29/2013

12.1 Inference for Linear Regression. Introduction

Gooseberry Sawfly Laying Eggs

The Overconsumption of Sugar

Assessing Agreement Between Methods Of Clinical Measurement

Lesson materials for printing and projecting. School Curriculum. Helping Youth. Eat Real. Classroom Lessons to Transform Youth and Their Communities

AP Statistics. Semester One Review Part 1 Chapters 1-5

7. Bivariate Graphing

Learning Zone Express Learning Zone Express

Transcription:

1 Caffeine & Calories in Soda Statistics Anthony W Dick

2 Caffeine & Calories in Soda Description of Experiment Does the caffeine content in soda have anything to do with the calories? This is the question I wish to investigate in this project. I know that caffeine has no calories, so I do not believe that this is an explanatory relationship, but just an association. Because of this, neither variable is explanatory or response, but I will use caffeine as the x variable throughout the experiment. However, because the sodas with caffeine are intended to give energy, I believe that these sodas will also have higher sugar content, but I do not believe this will be a strong association. Data Collection To gather the data, I went to a local Wal-Mart and looked at the labels of caffeinated sodas. There were only 34 to choose from, so I assigned 1 34 to each of them and randomly chose 4 to exclude (with a random number generator). This should eliminate error from sampling method, but will only pertain to the sodas available at Wal-Mart, since this is the only location I went to. Data Presentation I organized the data from high to low values of caffeine, as displayed in figure 1 on page 3. The diet sodas form another group which is distinct from the rest of the sodas, and was a trait that was not considered as I chose the initial sample. Also, similar brands have essentially the same caloric content, so the scatter plot appears almost like a categorical plot. The scatterplot (figure 3, page 5)shows a positive direction, although the slope appears to be close to zero. If the diet sodas are excluded, there seems to be only one possible outlier at (69mg of caffeine, 69 calories); this is identified as Coke Blak. It appears to linear, although almost a horizontal line. It does not appear to be very strong, and with a r-squared value of 0.0136, this is conformed. I decided to take out all the diet sodas and do the same analysis, hoping to establish a stronger relationship; this data table (figure 2, page 4) and graph (figure 4, page 5) showed siginificant change from the initial r-squared value,and also affected the regression line. In the regression line with diet sodas omitted, the r-squared value increased to 0.0244, an increases of almost 79%, and the slope of the regression equation changed from a positive relationship to a negative relationship. The y-intercept also changed. From the scatterplot, the slope is meaningful (or rather the change of slope). Because the caloric content changes only slightly, but the caffeine content was a much wider range, the slope is relatively flat. This adds strength to the argument of a very weak relationship.

3 Figure 1 SODA Caffeine content per 12 oz. (mg) Calories per 12 oz. (mg) Vault 71 174 Jolt Cola 72 150 Mountain Dew MDX, regular 71 180 Coke Blak 69 69 Code Red 54 165 Mountain Dew 54 165 Pepsi One 54 1.5 Mello Yellow 53 177 Diet Coke 47 0 Diet Coke Lime 47 0 TAB 46.5 0 Dr. Pepper 42 150 Dr. Pepper diet 44 0 Pepsi 38 150 Pepsi Lime, regular or diet 38 150 Pepsi Vanilla 37 165 Pepsi Twist 38 165 Pepsi Wild Cherry, regular or diet 38 165 Diet Pepsi 36 0 Pepsi Twist diet 36 0 Coca-Cola Classic 35 145.5 Coke Black Cherry Vanilla, regular or diet 35 150 Pibb XTA 35 156 Coke Cherry, regular 35 156 Coke Lime 35 156 Coke Vanilla 35 157.5 Coke Zero 35 0 Barq's Diet Root Beer 23 0 Barq's Root Beer 23 166.5 Table of specific sodas with their corresponding data (includes diet sodas)

4 Figure 2 SODA Caffeine content per 12 oz. (mg) Calories per 12 oz. (mg) Vault 71 174 Jolt Cola 72 150 Mountain Dew MDX, regular 71 180 Coke Blak 69 69 Code Red 54 165 Mountain Dew 54 165 Mello Yellow 53 177 Dr. Pepper 42 150 Pepsi 38 150 Pepsi Lime, regular or diet 38 150 Pepsi Vanilla 37 165 Pepsi Twist 38 165 Pepsi Wild Cherry, regular or diet 38 165 Coca-Cola Classic 35 145.5 Coke Black Cherry Vanilla, regular or diet 35 150 Pibb XTA 35 156 Coke Cherry, regular 35 156 Coke Lime 35 156 Coke Vanilla 35 157.5 Barq's Root Beer 23 166.5 Table of specific sodas with their corresponding data (omits diet sodas)

Calories per 12 oz. Calories per 12 oz. 5 Figure 3 200 180 160 Calories & Caffeine in Soft Drinks 140 120 100 80 60 y = 0.6573x + 78.448 R² = 0.0136 40 20 0 0 20 40 60 80 Caffeine Content per 12 oz. (mg) Scatterplot of soda data including diet sodas Figure 4 Calories & Caffeine in Soft Drinks (w/o diet sodas) 200 150 100 50 0 y = -0.2367x + 166.37 R² = 0.0244 0 20 40 60 80 Caffeine content per 12 oz (mg) Revised scatterplot of soda data when omitting diet sodas

6 Models Two models were presented as part of my data analysis. The first, with all of the gathered data, gave the average predicted calories in 12oz. of soda to be equal to 78.448 more than the product 0.6573 and the caffeine content (mg) of the same quantity of soda. This regression line had a coefficient of determination of 0.0136. This would have a corresponding r value of +0.1136. This indicates a very weak positive relationship between caffeine content and calories per serving of soda. However, in contrast to the first model, when the diet sodas are omitted, the relationship strengthens slightly, although it is still weak, and it changes from a positive relationship to a negative one. As discussed previously, this indicates that as the calories remain relatively constant, the caffeine content rises across a much broader range of values. The regression line of this second model indicated the average predicted calories in 12oz. of soda to be equal to 166.37 more than the product -0.2367 and the caffeine content (mg) of the same quantity of soda. This has a corresponding r value of -0.1562. This indicates a very weak negative relationship between caffeine content and calories per serving of soda. Predictions In this particular model, a high caffeine prediction would be inappropriate, primarily because by U.S. federal law, a beverage with more than 71 mg of caffeine per 12 oz. serving must be marketed as an energy drink; I confined my data to soft drinks. Taking this into account, for a caffeine level of 180 mg (about the same level in a Rockstar juiced), the first model predicts a calorie content of 196.762 while the second model predicts a calorie content of 123.764; a Rockstar juiced actually has about 165 calories per 12 oz. serving. In the introductory section, no claim of causation was given and upon analysis of the data, none is found. However, I did initially suspect an association, but also found no evidence for that. Error Evaluation Error may have been introduced through the limited sampling method chosen. Only one store was visited, and four of the sodas there were omitted. However, given the data and the homogeneous nature of calorie content among the studied soft drinks, as well as those not studied, I strongly suspect that this would not change the overall conclusions given. Also, when looking at the same sodas online, the calorie, and more so the caffeine content, changed from what was published on the product. This may be another source of error, and may change the specifics, but I do not believe it would change the overall conclusions either. Conclusions As reiterated throughout the paper and by the data itself, I find little association in soft drinks between caffeine content and calories. This would also indicate no causative relationship either.