Report #4: The Correlation between Caffeine & Sleep
Report #4: The Correlation between Caffeine & SleepBy: Adrienne GebeleI am a student-athlete at Ohio University. I keep a very busy schedule with my extracurricular activities, my undergrad in Nutrition Dietetics, my being a member of the softball team. Caffeine, more specifically, coffee is a big part of my day and helps me maintain my busy life style, stay awake, and perform to the best of my ability both on and off the field. For my project, I am studying the effect of caffeine on the amount of sleep student-athletes at Ohio University get on average on an average weekday. My research questions for this study are: “Is there a relationship between the number of caffeinated beverages a student-athlete drinks per day and the number of hours of sleep a student-athlete gets” and “Does the number of caffeinated beverages a person consumes per day predict the number of hours a student-athlete sleeps at night.” I collected my data on Monday, November 18, 2014 on the 4th floor of Peden Stadium at Ohio University. The 4th floor of Peden Stadium is the student-athlete study hall area and is a universal place for all athletes to come and study, print, use the computer, get tutored and meet with their athletic academic advisors. All sports teams and all years of athletes utilize the 4th floor of Peden Stadium. The variables I am studying during this experiment include the amount of caffeinate beverage consumed as my predictor variable and number of hours slept a night as my criterion variable. In a correlation study, the predictor variable is the ‘cause’ and the criterion variable is the ‘effect’. For this experiment, I predict that the more caffeinated beverages a student-athlete drinks, the fewer hours of sleep they will get a night. During my experiment I controlled for many variables. For instance, I controlled for gender, current college attended, age, activity level, schedule, lifestyle, and definition of the amount of 1 ‘caffeinated beverage’. I controlled for gender by surveying 10 male student-athletes and 10 female student-athletes. Controlling for gender ensured that an even number of males and females were surveyed and that gender would not play a role in influencing my results. Also, I only surveyed current undergraduate Ohio University student-athletes. This means all my participants in my survey are between the ages of 17-23. I decided to only survey the student-athlete population to control for activity level, schedule and life style. Student-athletes live a different lifestyle from the normal population. Student-athletes’ daily schedules normally involve a high amount of physical activity a day along with school and studying. This leaves very little time in their schedule for a job, extracurricular activates, and social events. I determined that the student-athlete population across the board, no matter which sport or gender, have similar schedules, daily demands, lifestyles, and activity level. Finally when asking participants the amount of caffeinated beverages they consume per day, I defined what a caffeinated beverage was and gave examples of soda or pop, energy drinks, and coffee. I also answered questions if the participant was unsure if a beverage had caffeine. I also defined for the participant the measurement, a cup (8oz), and showed then an approximation. Defining some of the terms and criterion helped for consistency among participants’ definitions and measurements, which in turn will result in more accurate results.I collected data from a total of 20 student-athletes. I surveyed 10 female student athletes and 10 male student athletes specifically about how many caffeinated beverages they drink per day and how many hours of sleep they got each night. Below are my data results in an array followed by the output of the data produced by SPSS, which includes the mean, standard deviation, correlations, regression, variables, anova, and summary. Also, demonstrated below are the correlation and regression calculations done by hand. Data Array:Males:Hours of sleep (X)7.56686966.577# Caffeinated drinks (Y)0100010200Females:Hours of sleep (X)7575.5776667# Caffeinated drinks (Y)2623100011Summarized Data:ΣX: 20ΣY: 132.5ΣX2: 62ΣY2: 893.75ΣXY: 122.5Descriptive StatisticsMeanStd. DeviationNIntake of Caffeinated Beverages1.001.48720Number of Hours Sleep per Night6.625.915920Scatterplot:The scatterplot, produced by Excel, shows slope of my data is a slightly negative slope. This is shown: as the amount of caffeine increases per day the amount of hours of sleep decrease. I would predict that the data represents roughly a linear -.3 slope due to how closely aligned the points are to the line. This shows there is a negative relationship between the cups of caffeine a day and number hours of sleep. The cups of caffeine a day does not result in fewer hours of sleep.Correlation:CorrelationsIntake of Caffeinated BeveragesNumber of Hours Sleep per NightIntake of Caffeinated BeveragesPearson Correlation1-.387Sig. (2-tailed).092Sum of Squares and Cross-products42.000-10.000Covariance2.211-.526N2020Number of Hours Sleep per NightPearson Correlation-.3871Sig. (2-tailed).092Sum of Squares and Cross-products-10.00015.938Covariance-.526.839N2020Covariance:Covxy= ΣXY-ΣXΣY122.5-(20)(132.5) n = 20 = -0.5263 n-119According to SPSS and my calculation the covariance of the data is -.0526. The covariance statistic represents the degree of relationship between two variables, in this case: number of cups of caffeine per day and hours of sleep. The negative sign of the covariance shows there is a negative relationship between the number of cups of caffeine a day and hours of sleep. In other words, as the amount of caffeine increases per day the amount of hours of sleep decrease. Pearson’s Correlation Coefficient:r=Covxy= -0.5263 = -0.3864 Fail to Reject (sx)(sy) (1.487)(.9159)Pearson’s Correlation Coefficient is calculated because the absolute value of Covxy is also a function of the standard deviation of X and Y. To resolve this problem we use Pearson’s Correlation Coefficient and divide Covxy by the standard deviations of X and Y. According to SPSS and my calculations the covariance of the data is -0.387. The negative sign of the correlation is demonstrates that the relationship between cups of caffeine and hours of sleep is a negative correlation. While the numerical value demonstrates the size or magnitude of the relationship since the numerical value of -0.387 is far from the value of 1 this shows there is a weak relationship between the variables, cups of coffee and hours of sleep.Hypothesis Testing:Ho: The variables, cups of caffeine and sleep hours, are not significantly correlated in the population.H0: p=0H1: The variables, cups of caffeine and sleep hours, are significantly correlated in the population.H1: p≠0α=.05, two-tailed df=18 r crit(.05)18= ±0.444When comparing the Pearson’s Correlation Coefficient value of -0.387 to the r critical value of ±0.444 we would conclude that we would fail to reject the null hypothesis and conclude the variables, cups of caffeine and sleep hours, are not significantly correlated in the population.Coefficient of Determination:Model SummaryModelRR SquareAdjusted R SquareStd. Error of the Estimate1.387a.149.102.8678a. Predictors: (Constant), Intake of Caffeinated Beveragesr2= (-0.3864)2=0.14930.149 is the proportion of variance in cups of caffeine per day due to the relationship with hours of sleep a night.Regression:Variables Entered/RemovedModelVariables EnteredVariables RemovedMethod1Intake of Caffeinated Beverages.Entera. Dependent Variable: Number of Hours Sleep per Nightb. All requested variables entered.ANOVAaModelSum of SquaresdfMean SquareFSig.1Regression2.38112.3813.161.092bResidual13.55718.753Total15.93819a. Dependent Variable: Number of Hours Sleep per Nightb. Predictors: (Constant), Intake of Caffeinated BeveragesSlope:b=Slope=Covxy= -0.5263 = -0.2392 S2x 2.2y-intercept:a= ?-(b)(x?)= (132.5/20)-(-0.2392)(20/20)= 6.8642Regression Equation: ?=-0.24+6.86Scatterplot with Regression Line:The line of regression shown on the scatterplot, produced by Excel, was determined using the regression equation, ?=-0.24+6.86. As demonstrated the slope of the line is .24 and the y-intercept is 6.86.Standard Error of the Estimate:Model SummaryModelRR SquareAdjusted R SquareStd. Error of the Estimate1.387a.149.102.8678a. Predictors: (Constant), Intake of Caffeinated BeveragesSy-? =Sy √(1-r2) n-1= .9159√(1-0.1493)19= 0.8679n-218SPSS and my calculations determined a standard error of the estimate at 0.8679. This value can be interpreted as the standard distance between the actual point and regression line is 0.8679.Predicted Values:Subject 7:X= 0 Y= 6 ?=bx+a=-0.2392(0)+ 6.8642=6.8642The actual y value for subject 7 is 6 while the predicted y value for subject 7 using the predictive equation is 6.86. The predictive equation estimated a y value close to the actual value for subject 7. Since the correlation of the x and y value is weak this is why the predictive equation did not calculate the exact y value for subject 7. The stronger the correlation of the two variables the more accurate the predictive equation is for generating one of the values. In conclusion, this study shows there is a weak correlation between the number of cups of coffee consumed by a student athlete and the hours of sleep they get each night. In this experiment gender, current college attend, age, activity level, schedule, lifestyle, and definition of the amount of 1 ‘caffeinated beverage’ were all controlled for. On the other hand, GPA, major, college class, and if the athlete was in-season or out-of-season was not controlled for and may have influenced the results. Even through the study showed there is no significant difference in the number of caffeinated beverages consumed and number of hours of sleep a night, it is interesting to see a pattern in the data collected. The study controlled for gendered to ensure gender did not skew the data results. Looking at the raw data 70% of the male student-athlete subjects do not consume caffeine on a normal weekday while 30% female student-athlete subjects do not consume caffeine on a normal weekday. I found this to be very interesting while collecting my data. I did not include prework, which is a NCAA legal performance-enhancing supplement, as a caffeinated beverage. For future research it would be interesting to determine what factors influence why male student-athletes consume less caffeinated beverages than female student-athletes. ................
................
In order to avoid copyright disputes, this page is only a partial summary.
To fulfill the demand for quickly locating and searching documents.
It is intelligent file search solution for home and business.
Related searches
- correlation between education and religion
- calculate the correlation coefficient r
- correlation between two variables
- compute the value of the correlation coefficient
- correlation between two variables examples
- examples of correlation between non related variables
- find the distance between the points calculator
- find the distance between the two points
- correlation between two numbers
- correlation between a1c and glucose
- correlation between hypertension and diabetes
- correlation between multiple variables excel