Finally, we met our target of 30 subjects! After some persuasion done to our classmates that the confidentiality of their height and weight will not be revealed. :P Had a bit of fun comparing our heights with one another. We can now move on to our data analysis!!
a little progress at 4:04 PM.
About us
[zhu ting]
Hi. I’m Seet Zhu Ting.
My admin no is 086488C. I’m not very tall nor do I have a perfect figure but I’m glad my BMI falls within a healthy range.
[lori]
Heh...
this is just a very short intro of Lori. =D This adventurous gal loves the sun and water and she thinks it's best to get to know her by interacting with her directly.
Heh. So much for intro huh..
Lori
[vicky]
Hello! :D
This is Vicky here. I’m a Physiotherapy student.
I love math and therefore I LOVE STATISTICS!!! <3
check out our beautiful blog for more details on our project!!!
[aneesa]
Heys! Okey.
Firstly I hate the topic of height and BMI.
( If you take a look at me, you’ll know why.)
So surprise surprise..Our group is doing on the very topic that I dread.
Oh well I sure do hope that our null hypothesis is rejected. Till next time..
Cheers
aneesa
[atiqah]
HI! I’m atiqah..
I was studying in pioneer junior college and I have a lot of interest in physiotherapy.
So here I am pursuing my course and hopefully I will be a good physiotherapist one day.
I love soccer too although I only started soccer during my first year of poly but I do hope I can improve my skills and play well in future.
Well that’s it basically about me!
1) Introduction
Body Mass Index (BMI)
The Body Mass Index (BMI) was invented between 1830 and 1850 by the Belgian polymath Adolphe Quetelet during the course of developing "social physics".
It is calculated as
BMI = Weight (kg) / Height (m) x Height (m)
Due to its ease of measurement and calculation, it is the most widely used diagnostic tool to identify weight problem within a population including: underweight, overweight and obesity.
Why Height as an independent variable?
Studies have shown that changes in height is often associated with changes in BMI. Weight, in fact, does not have the impression of making a significant change in BMI.
Take this example: If Robin weighs 68kg and is 1.75m tall, his BMI is 22.2.
If he gains 1kg, his BMI would be 22.5.
However, if he were to gain 1cm in height instead, his BMI would be 19.8.
Why is our study important?
Our study hopes to prove if there is any trend between height and BMI.
Since BMI is a used as an indicator of health, knowledge of the relationship between height and BMI (if there is any), would be useful in predicting the health threats one faces just by looking at his height.
To be able to make sensible predictions may encourage appropriate beneficial actions to be taken as precautionary measures.
Our hypothesis
Null Hypothesis: There is no relationship between one's BMI and Height.
Alternative Hypothesis: There is a relationship between one's BMI and Height.
2) Literature Review
We have found a number of literature reviews that shows the relationship between height and BMI.
Trends in Height and BMI of 6-Year-Old Children during the Nutrition Transition in Chile
Juliana Kain*, Ricardo Uauy*,†, Lydia Lera*, Marcela Taibo‡ and Cecilia Albala* 9 September 2005
Their hypothesis: Both stunting and tallness are potentially associated with increased risk of obesity. This interaction may have changed over time as the nutrition transition progressed.
Justification: Stunted children have lower lean body mass and lower energy needs per unit body mass; thus, they are at an increased risk of energy excess.
In tall children, earlier maturation may bring about more weight gain; if weight gain exceeds the corresponding gain in height, the number of those with excess BMI for age may rise.
Results Analysis: Seen from the result analysis, the probability of obesity among tall boys and girls was significantly greater than that for normal height children (median +/- SD); the upward trend became evident as height approached the median.
The probability of a stunted boy being obese progressively increased from 0.06 to 0.15 for each survey year (over 15years) compared with that of a tall child, which rose from 0.12 to 0.25 in 2002.
In short: Tallness is significantly associated with increased obesity risk in children, while stunting is also associated, but to a lesser degree.
Height-Related Changes in Body Mass Index: A Reappraisal
Dror Mandel, MD, Eyal Zimlichman, MD, Francis B. Mimouni, MD, FACN, FAAP, Itamar Grotto, MD, MPH, and Yitshak Kreiss, MD (2004)
Their hypothesis: BMI is influenced by height and that the degree of adiposity is not uniformly distributed in the population.
Results Analysis: There was a positive correlation between BMI and height in men (regression slope = 0.00717, r = 0.015, p = 0.03), while the correlation was negative in women (regression slope = -0.02811, r = -0.05, p < 0.0001).
In multiple regression analysis, when BMI was used as the dependent variable and height, gender, ethnic origin, smoking, oral contraceptive use, and level of recreational exercise as the independent variables, only height, gender, and ethnic origin remained significant in the final analysis (R-square 0.0205, p < 0.0001).
To view table with results please click: Result table
In conclusion: In young adults, BMI is affected in a subtle, but opposite manner in males and females.
In males, BMI increases with increasing height, while in females, BMI decreases with increasing height.
BMI rebound, childhood height and obesity among
adults: the Bogalusa Heart Study
DS Freedman1*, L Kettel Khan1, MK Serdula1, SR Srinivasan2 and GS Berenson2 2001
Their objective: The beginning of the post-infancy rise in the body mass index (BMI, kg=m2) has been termed the adiposity
rebound, and several studies have found that an early rebound increases the risk for overweight in adulthood. We examined
whether this relation is independent of childhood BMI levels.
Their results and findings: it is possible that childhood height may help identify children who are likely to become overweight adults.
for example Among
relatively heavy (BMI.16.0 kg=m2) 5-y-olds, a child with a height of 120cm was estimated to be 1.2 kg=m2 heavier in
adulthood than would a 104cm tall child.
Today my group met to start with our taking of measurements. We met around 11.
We were discussing more about the project and finally we were on our way to the next step of what is needed to be done.
How was the data collection done?
Data collection will be done by collating all the results of the individual subjects.
We have decided that each of us was given task to do. As there are a total of 5 people in the group, we delegate the responsibilities.
2 will be taking the measurement s of the subject’s height, another 2 will take care of the weight while the last member will collate all the results at the
end of the measurement. We decided to take about 30 subjects randomly from the school and thus our task began in J204, School of Health Sciences, Nanyang Polytechnic.
We agreed to take 3 measurements of both the height and weight respectively and take the average of the entire 3 results we found.
These 3 measurements will be done 3 days consecutively and by the same person so as to avoid errors and maintain reliability of the results.
Height Station
Weight Station
How did you gather the samples? (Eg measure from where to where?)
We took random people to be involved in our project. They don’t necessarily have to be of someone we know of.
Hence we used the simple random sampling method.
Once getting the subjects, we brought them to our physiotherapy lab, J204 where all the instruments are and thus get them to go to the 2 different stations;
the height station and the weight station.
At the height station, we measure them basically from their feet all the way to the head.
Our person in charge stands on a chair so as to avoid unnecessary errors.
We used a flat surface like a file to measure the height of the subject.
While one of us stand on a chair, the other is holding onto a file and paper to write the measurements down as well as to make sure the subject don’t cheat.
We ensure that the subject is bare-footed.
As for the weight station, our group member ensure that the arrow always point to zero before we get the subject to stand on the weighing machine.
The 2 people assigned to this station will ensure that the subject emptied his pockets and takes off the shoes before stepping on the weighing machine.
The same routine, one will take down the result and the other will read the number.
Throughout the 3 days, we ensure that those who are doing the results writing and reading of measurements remain the same to maintain reliability and validity.
What did you use to measure?
We decided to hold the measurements taking in one of the physiotherapy lab, J204.
The measuring tape and weighing scale is brought by one of group members.
We paste the measuring tape against the wall.
Since the measuring tape wasn’t long enough, we measured up to 1m and pasted the measuring tape on the wall after 1m.
We did not remove the measuring tape for the 3 consecutive days to avoid unnecessary errors.
Nearby we placed the weighing machine. Same thing, to avoid errors we decided to use the same weighing scale for the consecutive 3 days so as to get a consistent
results and avoid minor errors as well accuracy of the results is higher.
There were no rejection of results.
Here's a copy of our data entry form!
Stay tuned to check out on our statistical analysis results! Aren't we excited to find out if they have any strong correlation? :D
Find your BMI NOW!!
Fill in your height and weight in the blanks
Before applying Pearson's Correlation Coefficient, the following assumptions must be made.
Assumption 1: All observations must be independent of each other. Assumption 2: The dependent variable should be normally distributed at each level of the independent variable. Assumption 3: The dependent variable should have the same variability at each level of the independent variable. Assumption 4: The relationship between the dependent and independent variables should be linear.
Assumption 1 is fulfilled as samples were taken randomly and independently.
Assumption 2
The dependent variable, BMI, is distributed normally at each value of the independent variable, height.
This can be done by plotting a QQ plot.
From the QQ plot, we can see that the variables are normal.
Assumption 3
The dependent variable, BMI has the same variability at each value of the independent variable, height.
Assumption 4
The relationship between the dependent variable, BMI, and the independent variable, height, should be linear.
From the scatter plot, we can observe that there is a positive linear relationship between the variables.
Applying the Pearson's Correlation Coefficient
Now that we have ensured that all the assumptions have been fulfilled, let us take a look at our data's Pearson Correlation Coefficient!
From the above table, a Pearson's correlation coefficient of 0.172 indicates a weak positive relationship between height and BMI.
Linear Regression
Knowing that a relationship between a person's height and BMI exists, linear regression quantifies the relationship by having an equation predicting a person's height and BMI.
The linear equation is
BMI = 7.093*(height) + 9.581
So what do all these results mean? Check them out in our conclusion! :)
4) Conclusion
Now let us refresh our memory.
The main objective of this project is to find out if there's any correlation between BMI and height.
We have obtained the height and weight values, of where we calculated and obtained our BMI.
Hypothesis:
H0 : There is no relationship between one's BMI and height.
H1: There is a relationship between one's BMI and height.
From the analysis of our statistics, with r= 0.172, we can conclude the following that:
- There is a positive linear relationship between BMI and height, with r>0
- The relationship between BMI and height is weak, as r<0.5
- The relationship is insignificant with r<0.5 and p>0.05
Since p>0.05, we fail to reject our null hypothesis. The relationship between BMI and height is not significant.
Haha, what an interesting conclusion! Now we all know. :D
But we must also note that this could also be because of our relatively small sample size of 30!
5) Reflections
the process
The process of data collection went rather smoothly.
Good Parts:
1. There was no difficulty getting the required number of subjects.
2. Height and weight were measurements that could be recorded wihout much hassle or time.
3. Recording was done easily as we printed out table formats to fill in the details neatly.
4. Reliability was ensured as the one person took all measurements of height and another took all of weight.
5. With 5 people in the group, we did not have to play multiple roles - photographer, height measurer, height recorder, weight measurer and weight measurer.
Bad Parts:
1. We had to make sure that the subjects turn up for 3 consecutive days in the morning to take their measurements.
2. The weighing machine was not digital, hence the results may not be as accurate.
3. The measuring tape did not start from the floor but 1.00m from the floor instead, hence every morning we had to measure the distance of 1.00m from the ground.
the putting together..
Putting the bits and pieces together was tediously and painstakingly done due to the unforseeable temperaments of the internet and blogspot.
Within minutes our hours of hardwork would disappear and be replaced with blank screens.
We didn't realise till much later that that was due to all of us saving our templates at different times and thus leading to each of us cancelling each other's efforts.
But alas..the job was done; all neat and perfect in our own ways.
I think that we were only too glad to accept that our masterpiece was in a good form before retiring to bed after a good deal of hours past bedtime.
Now we leave it in the hands of those more capable of judging our attainment...
Tuesday, July 14, 2009
Finally, we met our target of 30 subjects! After some persuasion done to our classmates that the confidentiality of their height and weight will not be revealed. :P Had a bit of fun comparing our heights with one another. We can now move on to our data analysis!!
