Hello everybody, hello for the last time!
So this week was the final week of my 8-week research placement here at the University of Kent. I was told to write a ‘2 DIN A4 pages long blog’ for the final week to sum up my experience and what we found, so I will try my best to fulfil this.
Looking back to the beginning seems longer than 8 weeks to me now. Looking back I remember thinking: ‘what the hell is a zygion?’ and also ‘oh yeah, let’s do a questionnaire on our own’… Well, don’t mess around with questionnaires. I’m not joking, don’t even try to use them. They are evil. And I really mean it. So what did we find?
Descriptive Statistics
Our sample consisted of 55 men and women (men=21, women=34). The mean age was 61.84 years altogether, although men were slightly younger (58.19 years) than women (64.09 years). We had 7 female smokers and 5 male smokers. 5.9% of the females were underweight, 32.4% normal and the remaining 61.7% were overweight or obese class I,II,III according to BMI categories. No single man was underweight, however 33.3% were normal, and 66.7% fell into the overweight and obese categories. BMI categories do not take age into account, but weight correlates with age so it is somewhat unsurprising that quite a few people are indeed overweight as our sample consisted of an older population.
Significant differences between men and women
Unsurprisingly, men were significantly** (two ** means significant value p<0.01, one *means p<0.05) stronger in their best hand grip strength than women were (similar finding for each hand).
They are also significantly** taller, heavier, have bigger zygions (face width), bigger nasale lenght (length of nose) and wider jaws.
Finally, men also show significantly** higher values for their best peak expiratory flow, which indicates lung capacity and represents to some extent a measure of upper body strength.
All our other measurements were insignificant: D2:D4 both hands (D2= index finger in relation to D4= ring finger), Body Mass Index, all three face ratios (Width:Height, Jaw:Height, Width:Jaw), Systolic and Diastolic blood pressure and pulse.
Significant Correlations
First of all, I need to say that we didn’t find what we thought we would. There were no significant relationships between any of our facial ratios (Width:Height, Jaw:Height, Width:Jaw). Boooo. However, we found some other interesting things (I won’t bore you with the obvious significant correlations such as weight and height).
Best Peak Expiratory Flow (PEF) is positively correlated (when PEF increases then the following increases as well) with height**, weight*, hand grip strength** (right, left and best out of 6 attempts), zygion**, nasale*, and jaw width**. PEF was negatively correlated (when PEF increases the following increases, or vice versa) with age**, sex**. So the older the person the less PEF.
Best Hand Grip Strength (best out of three on each hand, =HGS) is positively correlated with height**, weight **, zygion**, nasale*, jaw**, and diastolic* blood pressure. HGS was negatively correlated with age** and sex**(means men were better).
Other correlations: Age was positively correlated with systolic blood pressure**. Means the older the higher the blood pressure. Normal. Unfortunately, different to previous literature, we did NOT find a correlation between facial width-to-height ratio and BMI. However, we found a positive correlation* between BMI and jaw-to-height ratio, and a negative correlation* between BMI and cheekbone dominance (width-to-jaw ratio).
Also, for some reason systolic and diastolic blood pressure are correlated with finger length and D2:D4 ratios. I have not figured out yet why, so I will update this as soon as I know a bit more.
A new regression model
A regression is a statistical model for predicting a certain measure/variable by the use of one or more different variables. We found in previous literature that PEF was used to get a rough overview over a person’s current and future health. However, there has not yet been a model for predicting PEF that included hand grip strength. We found a model that predicts PEF to 72.9% (adjusted R^2=69.7%). This means it can explain 72.9% of the variance of the predicted values for PEF.
PEF is predicted by sex, age, height, weight, and new in the model left hand grip strength.
What we didn’t test
Unfortunately, we did not record a few variables that seemed to be important in previous literature: level of education, marital status, body fat (important for accurate BMI), more health backgrounds (e.g. childhood health such as birth weight, operations, illnesses, arthritis in fingers, etc.), income, dominant hand, gait speed (time to walk a 98.5 inch span), insulin levels and amount of time spent standing on one leg (seems to be a good predictor of core muscle strength).
Well, we can’t test for everything…
Overall implications and summary
We haven’t finished analysing the data so a summary will follow here soon.
My own experience
Overall, I really enjoyed being responsible for my own research project. It definitely made me realise that I need to either get a bigger screen to stare at all day, or try to go for a lunch walk/gym session. However, I can imagine doing research as a career and that’s good, otherwise I wouldn’t have a clue what to do as I assumed that I would like doing research.
Things I disliked was the lack of exercise, the slow process, the annoying search for participants and sharp comments about my lack of ‘proper’ English and talking to people who always take the mick.
Things I liked were the gain of knowledge, the feeling of responsibility for my own project, data analysis, feeling nerdy while learning how to use MatLab, the family-like atmosphere at our Psychology department, scoring football goals at the inter-department evening football, and playing bat n trap.
I hereby thank everybody involved making my time enjoyable at Kent this summer, and supporting my daily request for scratch parking tickets (sorry Esmé!).
So long,
Anna B.
(update with the findings will follow)
