BMI Statistical Report
The aim of this project is check female and male BMI data to help in the development of the Obesity program that will draw information from the results of this analysis.
Chan and Jean Woo in their paper Prevention of Overweight and Obesity: How Effective is the Current Public Health Approach, define obesity as a condition of excess or abnormal accumulation of adipose tissue to a level of health impairment. BMI is considered and used as the best level of measuring obesity and is the simplest index that determines whether someone is overweight or underweight. With BMI, it is considered that if the index ranges from 30.0 to 39.9 a person is considered obese above which a person is characterized as extreme obese. Body mass index (BMI), waist-to-hip-ratio (WHR), and waist circumference (WC) which are used in the diagnosis of visceral obesity but WC presents more accurate data.
Variables of the study:
BMI and Gender
Sample size: 40
With the help of excel, to start the analysis, there I need to separately calculate the BMI and the 95% confidence intervals for male and females as shown below.
|BMI for females -CI OF MEAN
|Confidence Level (95.0%)
|BMI for males-CI FOR MEAN
|Confidence Level (95.0%)
Secondly, Using the excel MIN and MAX functions, the maximum and minimum values of BMI and confidence intervals for males and females are calculated separately.
From the results, it is evident that the BMI for female is 21.9 compared to the male BMI which is higher at 24.2. Also, the minimum BMI index for female is 19.6 which are lower than that of their male counterpart which is 23.8. Also, from the data, it is evident that the maximum and the minimum BMI of the males fall outside the confidence interval which implies that the values can be ignored in the development of the obesity program and more focus be placed on the confidence intervals.
The next step is trying to concentrate the maximum values of BMI while trying to reduce the male and female counts in both cases.
|MAX BMI MALES
|MAX BMI FEMALES
From the results, it is evident that out of 822886 females, 20278 have max BMI, on the other hand, out of 403140 males, 5600 have max BMI. These accounts for 1% and 2% of population respectively which have max BMI. This results show some good news which shows an evident that the obesity program can be effective in reaching out both the population of males and females.
Histogram provides a pictorial overview of the variability in the counts of BMI for the confidence intervals obtained above for males and females respectively.
|Bin for BMI in CI FEMALE
|Frequency of BMI
Bin for BMI IN CI MALE
|Frequency of BMI
From the result, we can conclude that pitting more emphasis on the BMI of male and females is essential because they have more counts. Next we plot a pie diagram with percentage of each BMI for females greater than 27.7:
Also, there is need to concentrate ob. the 6% and 7% BMI levels as shown in the above pie charts.
Plotting a pie chart with the percentage of BMI greater than 27.1 is as shown below.
Therefore, there is need to focus more on 8%-9% BMI values mentioned above.
The pie charts present the percentage in the range >27.7 and >27.1 BMI but not overall.
The correlation that need to be checked is for females with BMI >27.7 and males with BMI>27.1. Correlation is used to develop significance of this analysis. Oppenheim summaries his findings that effort and cost of the Research Assessment Exercise may not be justified when a cheaper and simpler alternative than correlation exists.
Therefore the results become:
When the sign is ignored, the correlation weakens as it approaches zero for both females and males for relation between count and BMI. This implies that more factors should be put into consideration to make the program more effective and accurate.
The obesity program derived on the provided data should be rely only on the BMI for males with >27.1 and for females with >27.7 BMI and the rest of the analysis remains the same.
There is need to find out other factors that contribute to age wise obesity like some drinks and junk foods then develop a regression model be on the factors.
Prediction of abdominal visceral obesity from body mass index, waist circumference and waist-hip ratio in Chinese adults: receiver operating characteristic curves analysis by Jia WP , Lu JX , Xiang KS , Bao YQ , Lu HJ , Chen L.
Ruth S.M Chan and Woo: Prevention of Overweight and Obesity: How Effective is the Current Public Health Approach.
CHARLES OPPENHEIM: THE CORRELATION BETWEEN CITATION COUNTS AND THE 1992 RESEARCH ASSESSMENT EXERCISE RATINGS FOR BRITISH LIBRARY AND INFORMATION SCIENCE UNIVERSITY DEPARTMENTS by