Professional Documents
Culture Documents
Investigation of The Relationship Between Leg Length and Lateral Jumping Distance
Investigation of The Relationship Between Leg Length and Lateral Jumping Distance
com by Hselim
Page 1/17
om
.c
ud
lo
ic
@
ky
Jumping Distance
lg
ae
ni
ha
y
tif
as
Cl
Downloaded from www.clastify.com by Hselim
Page 2/17
Introduction
In elementary school, I was an avid athlete. I loved every sport I tried. Physical Education was
my favourite subject. It was no surprise that every year, when our school would have an event
called track and field where it was a competition of physical finesse, I excelled. Everyone
competed in categories like running and jumping where stamina and fitness were crucial factors
in victory.
My favourite category was called the standing long jump. Everyone jumped one by one from a
stationary position and the winner was whoever could jump the furthest. I would come in the
om
top three every year amongst my classmates and represent my school along with two others at
.c
the annual regionals competition. It wasn’t until later in high school did I realize that the top
ud
lo
three winners in standing long jump were tall. I wasn’t short by any means; I was well above
ic
average in terms of height. Yet the people who jumped further than me were still taller. This led
@
ky
me to believe that there was very likely a relationship between leg length and jumping
sa
capabilities. Throughout this internal assessment, I hope to determine the answer to two
aw
questions:
lg
ae
1. Is there a correlation between leg length, measured in centimeters from the top of the
ni
femur bone to lateral malleolus of fibula, and the distance the person is able to jump, in
ha
2. If yes, is there an optimum point after which increasing leg length no longer increases
as
Cl
My research topic involves statistics, as I will plot the relationship between distance and leg
length on a scatter plot. By doing this, I will also be able to calculate regression. This will then
allow me to conduct statistical tests in order to give a quantitative value to the relationship
between leg length and lateral jumping distance.
I will find 30 female high school students in Toronto, Canada, as a sample for my data. It will be
quota and convenience sampling, as I am choosing 30 individuals that fit a certain characteristic
(female and high school student) and asking my friends rather than posting an online survey.
Downloaded from www.clastify.com by Hselim
Page 3/17
Everyone will have three chances to jump, and I will calculate the mean of the three jump
distances. I will be measuring leg length from the top of the femur bone to the lateral malleolus
of fibula in centimeters with measuring tape.
Definitions
Lateral malleolus of fibula: a knob on the outside of the ankle (Nemours Children Health
System, n.d.)
Femur bone: the only bone in the thigh
Optimal value: The highest or lowest point in a parabola
om
Outlier: an observation that lies an abnormal distance from other values in a random sample
.c
from a population (Engineering Statistics Handbook, n.d.).
ud
lo
ic
@
ky
sa
aw
lg
ae
ni
ha
y
tif
as
Cl
Page 4/17
The drawn diagram below shows the jumping and measuring process.
om
.c
ud
lo
ic
@
ky
sa
aw
lg
ae
ni
ha
y
After each jump attempt, the student was asked to remain still and not move their feet. I then
measured from the starting line to the student’s heels (or hand if the student has fallen) in
centimeters. The student was allowed to move after I recorded the measurements.
Using the method described above and shown in Diagram 2, I collected data from 30 highschool
females by convenience and quota sampling. The table below shows the raw data of the leg
length of the 30 students, as well as the jump distance they achieved for each of the three trials.
The data from this table will be used to calculate the average jump distance for each student:
Downloaded from www.clastify.com by Hselim
Page 5/17
Table 1: Raw Data of the Leg Length and Lateral Jump Distance of High School Females
om
81.4 192.6 190.2 188.8 71.8 160.6 164.8 165.8
.c
75.8 153.2 104.4* 148.6 90.8 180.2 182.6 179.4
ud
83.0 192.4 189.6 191.8 80.4 183.8 181.4 184.4
lo
ic
87.2 176.0 178.2 179.2 @ 72.0 158.8 164.6 166.2
ky
84.0 180.4 181.4 180.4 76.4 168.4 167.2 167.8
sa
Page 6/17
om
exclusion of outliers, would be the most appropriate approach in making the most accurate
.c
model possible.
ud
lo
ic
I used the following equation to calculate the mean jump distance. The calculated mean jump
@
ky
distances were adjusted to one decimal place.
sa
aw
𝑇𝑟𝑖𝑎𝑙 1+𝑇𝑟𝑖𝑎𝑙2+𝑇𝑟𝑖𝑎𝑙3
lg
The calculation below uses the jump distances from the first student.
y
tif
as
151.6±0.1+153.8±0.1+159.0±0.1
𝑀𝑒𝑎𝑛 𝑗𝑢𝑚𝑝 𝑑𝑖𝑠𝑡𝑎𝑛𝑐𝑒 = 3
Cl
464.4±0.3
𝑀𝑒𝑎𝑛 𝑗𝑢𝑚𝑝 𝑑𝑖𝑠𝑡𝑎𝑛𝑐𝑒 = 3
The mean jump distance for the first student is 154.8±0.1cm. The unit used was centimeters
(cm), as it is flexible enough to accommodate all possible jump distances. It also allows for a
precise measurement with relatively low uncertainty while also being part of the
universally-recognized metric system (Metric System, n.d.).
Downloaded from www.clastify.com by Hselim
Page 7/17
Below is the data table of the mean jump distance of each student, calculated as the sample
calculation shown above. This data table allows for me to graph the relationship between
average jump distance and leg length to determine the correlation between the two:
Leg Length ( Average Jump Distance Leg Length (±0. 1𝑐𝑚) Average Jump
±0. 1𝑐𝑚) (±0. 1𝑐𝑚) Distance (
om
±0. 1𝑐𝑚)
.c
82.2 154.8 76.2 171.6
ud
lo
79.6 174.7 81.2 184.3
ic
70.2 149.2 @ 73.6 162.0
ky
Page 8/17
Because the inclusion of outliers would prevent the graph from being as accurate as possible, I
excluded them from the calculation in Table 1 above. I used a box and whisker graph below,
plotted using http://www.alcula.com/calculators/statistics/box-plot/, to visually represent the
outliers amongst the data collected.
om
.c
ud
lo
ic
@
ky
sa
aw
lg
ae
ni
ha
y
tif
as
Cl
Using a GDC, I determined that the outliers in the data set collected are 104.4, 116.2, and 121.8
Downloaded from www.clastify.com by Hselim
Page 9/17
After calculating the mean jump distance for each student and excluding the outliers, I plotted
the data points in a scatter plot graph, including uncertainties, using Google Sheets, as shown
below:
om
.c
ud
lo
ic
@
ky
sa
aw
lg
ae
ni
ha
Looking at Figure 4 above, it can be seen that there is a trend: as leg length increases, so does
Cl
lateral jumping distance. Just by looking at the data points, the models I am considering are a
linear regression model and a polynomial regression curve. However, it seems that, with the
exception of a few data points, the trend seems to be a linear correlation. Therefore, I decided
to make a graph with a linear trendline, shown below.
Downloaded from www.clastify.com by Hselim
Page 10/17
om
.c
ud
lo
ic
@
ky
sa
Figure 5: Relationship Between Leg Length and Lateral Jump Distance with a Linear Trendline
aw
lg
From Figure 5 above, with the use of a linear regression line, it can be seen that there is a
ae
ni
relationship between leg length and lateral jumping distance. There is a linear positive trend line
ha
that represents their relationship. However, when{𝑥 ∈ 𝑅 | 𝑥 > 83}, the data points no longer
y
tif
follow the trend line. Rather, they stay at somewhat of a constant. I believe that this is the case
as
because there is an optimal value that, if surpassed, will no longer continue to display a positive
Cl
It is also important to consider the limitations within my collected data sample. I collected data
from 30 students. Of those students, only 3 people had a leg length of over 83 cm. That is not
enough data to allow for a good trendline or regression model. Therefore, the trendline in
Figure 5 is not an accurate representation of data for all people with leg lengths above 83 cm.
Downloaded from www.clastify.com by Hselim
Page 11/17
To measure the correlation between leg length and lateral jumping distance, I conducted a
statistical test to find the value of 𝑟. The closer the 𝑟value to − 1or1, {𝑟 ∈ 𝑅 | − 1 < 𝑟 < 1},
the stronger the correlation between two variables
om
medium-strong correlation between leg
length and jumping distance in the data
.c
ud
sample collected.
lo
ic
@
The linear regression equation for Figure 5 obtained using a TI-84 Plus CE graphing calculator is:
ky
𝑦 = 1. 72𝑥 + 35. 9
sa
aw
In the equation of the regression above, y represents the average horizontal jump distance in
lg
centimeters while x represents the leg length, also measured in centimeters. The regression
ae
line, however, should only be used to predict the y when{𝑥 ∈ 𝑅 | 0 ≤ 𝑥 ≤ 83} because there
ni
ha
is no longer a positive correlation between the two variables beyond the domain given. This
y
Due to this, I decided to graph a polynomial regression to see if it would be better suited to the
collected data set than a linear regression model. A polynomial regression curve would show a
clear optimal value, as well as model the trend more appropriately when
{𝑥 ∈ 𝑅 | 0 ≤ 𝑥 ≤ 83}. The graph below shows the relationship between leg length and
average jump distance with a polynomial regression curve instead of a linear regression line.
Downloaded from www.clastify.com by Hselim
Page 12/17
om
.c
ud
lo
ic
@
ky
sa
Figure 6: Relationship Between Leg Length and Lateral Jump Distance with Polynomial Trendline
aw
2
y
model below.
2
𝑅 ≈ 0. 615
Page 13/17
To compare both models and determine which one is a better fit for the data sample collected, I
2
compared the 𝑅 values, which determines how suitable a regression line is to the data it’s fitted
2 2 2
{ }
to (Hershy, 2019). The larger the 𝑅 value, 𝑅 ∈ 𝑅 | 0 < 𝑅 < 1 ,the better the model is
suited to the data sample collected.
2 2
The 𝑅 value in the linear regression model (Figure 5), calculated using a GDC, is 0.543. The 𝑅
value in the polynomial regression model (Figure 6) is 0.615.
om
.c
ud
2 2
Because the 𝑅 value of the polynomial regression model is larger than the 𝑅 value of the linear
lo
ic
regression model, a polynomial trendline would be a more accurate representation of the data
@
ky
displayed. The regression falls to the left and falls to the right as well for the end behaviours.
sa
aw
Conclusion
lg
The graph and the trendline are not good choices for extrapolations. Data when
ae
ni
{𝑥 ∈ 𝑅 | 𝑥 > 83}becomes extremely unreliable due to lack of information, but both the linear
ha
regression model and the polynomial regression curve are good fits when
y
tif
2
{𝑥 ∈ 𝑅 | 0 ≤ 𝑥 ≤ 83}. This is also indicated by the 𝑅 values, 0.543 and 0.615 respectively, One
as
could use the regression equations to fairly accurately predict lateral jumping distance (y) when
Cl
However, it is important to note that if one were to extrapolate data, the linear regression
trendline will continue to display a positive correlation between leg length and jumping
distance. The polynomial trendline will display a negative correlation between leg length and
jumping distance after the optimal value. I find that these are both unrealistic predictions of
possible trends. It is impossible to make a good extrapolation with such little data. I think that if
Downloaded from www.clastify.com by Hselim
Page 14/17
I had found more students with longer legs, I would have been able to come to a more concise
conclusion and make a prediction with evidence supporting my theory.
Because there is a correlation between leg length and lateral jumping distance, as determined
by Pearson’s Correlation Coefficient, r, it can be said that one has a greater chance to jump
further with longer legs. I understand that my height was likely a prominent factor of my success
in the standing long jump. Even so, there were more than two people who were taller than me,
but I still placed in the top three every year. This lead me to believe that there were other
factors that influence one’s success in the standing long jump, such as:
om
.c
1. Athleticism
ud
lo
Some people, like me, have been involved in sports and physical exercise for a very long
ic
time. This means that their muscles are more developed and are likely stronger
@
ky
(University of Southern Denmark Faculty of Health Sciences, 2016). They also have more
sa
muscle mass, which allows them to excel in physical activities such as the standing long
aw
jump. An athlete with shorter legs could very well jump further than a normal person
lg
ae
While the body mass index (BMI) is not a good indicator of health, several studies have
as
shown that there is a correlation between higher BMI and reduced physical performance
Cl
(Shen et al. 2015). Because I did not set a BMI range when collecting data, there were
people with varying BMI. This indicates that those who have a higher BMI likely did not
perform as well as their counterparts that have lower BMI. This also ties in with the first
possible factor, as athletes tend to have lower BMI.
3. Race
It has been scientifically proven that some races have a natural athletic advantage over
others (Encyclopedia.com, n.d.). Those of West African ethnicity have significantly more
Downloaded from www.clastify.com by Hselim
Page 15/17
fast-twitch fibers and anaerobic enzymes than Caucasians. This will result in stronger
muscles and increased physical performance. For example, if one were to ask two
people, with all other factors being the same (gender, age, BMI, mentality), one of
african descent and the other of caucasian descent, it is extremely likely that the person
of West African descent will perform better.
4. Mentality
Those who have a naturally more competitive mindset would have tried harder than
those who don’t particularly care (Neat., 2018). During my data collection process, it was
om
easy to tell which people cared about their results and which people simply did it as a
.c
favor to me. Those who had a competitive mentality usually jumped progressively
ud
lo
further at each trial, whereas those who didn’t care jumped haphazardly.
ic
@
ky
While I do believe that this investigation was conducted well, there are always limitations. My
sa
first limitation was data sample size. Due to COVID-19 and the new protocols, I had a hard time
aw
finding even 30 girls to collect data from. I believe that, given a larger sample size, I would have
lg
ae
been able to draw a more reliable conclusion. However, to do this, I would have had to gather a
ni
large number of people at once, which is simply not possible. If I had asked every student to
ha
measure themselves, I would have had a larger sample size, but also another limitation
y
tif
regarding the consistency of the measurement process. Therefore, I limited myself to collecting
as
data from only 30 people. The second limitation occurred during the data collection process.
Cl
Instead of allowing the students to rest after every jump, I asked them to perform three
consecutive jumps immediately one after the other. This would have caused muscle fatigue and
strain, hindering them from their peak performance. The final limitation was the type of
sampling I used. I only reached out to those I already knew to ask to collect data from them.
Instead of using quota convenience sampling, I should have used quota random sampling. It
would have been beneficial to have a wider variety of students to collect data from, as it
provides a more insightful perspective on how the other factors listed above could influence
lateral jumping distance as well.
Downloaded from www.clastify.com by Hselim
Page 16/17
Based on my collected data, the longer the leg length, the greater the jumping distance. Even
though the end behaviour indicates that jump distance is no longer directly influenced by leg
length when {𝑥 ∈ 𝑅 | 𝑥 > 83}, people with longer legs still jumped further than those with
shorter legs. Overall, it can be concluded that there is a correlation between leg length and
lateral jumping distance, as I expected. The 𝑟value indicated that there was a medium-strong
correlation between leg length and lateral jump distance, which makes sense when I realized
that there were many other factors that contributed to success in the standing long jump.
om
I’d like to do further research on whether leg length affects other physical performances as well,
.c
such as sprinting and high jump. Both of these demand a lot from the athlete’s legs, therefore I
ud
lo
believe that there will be at least some sort of correlation. However, they also require the use of
ic
other parts of the body, such as stamina and momentum, unlike the lateral long jump. Because
@
ky
of this, I think there will be less of a correlation between leg length and better performance, but
sa
References
ni
Neat. (2018, August 22). How Your Mindset Can Influence Your Performance. Neat
ha
Nutrition.https://www.neat-nutrition.com/blogs/news/how-your-mindset-can-influence-your-p
y
tif
erformance
as
Institute. https://corporatefinanceinstitute.com/resources/knowledge/finance/correlation/
Encyclopedia.com. (n.d.). Genetics and Athletic Performance. Retrieved February 4,
2021, from
https://www.encyclopedia.com/social-sciences/encyclopedias-almanacs-transcripts-and-maps/ge
netics-and-athletic-performance
Engineering Statistics Handbook. (n.d.). 7.1.6. What are outliers in the data? Retrieved
February 4, 2021, from
https://www.itl.nist.gov/div898/handbook/prc/section1/prc16.htm#:~:text=An%20outlier%20is%
Downloaded from www.clastify.com by Hselim
Page 17/17
20an%20observation,random%20sample%20from%20a%20population.&text=Examination%20
of%20the%20data%20for,often%20referred%20to%20as%20outliers.
Hershy, A. (2019, October 15). Calculating R-squared from scratch (using python) -
Towards Data Science. Towards Data Science.
https://towardsdatascience.com/r-squared-recipe-5814995fa39a
Innerbody Research. (n.d.). Muscles of the Leg and Foot. Innerbody. Retrieved February
3, 2021, from https://www.innerbody.com/anatomy/muscular/leg-foot
metric system. (n.d.). Encyclopedia Britannica. Retrieved February 4, 2021, from
https://www.britannica.com/science/metric-system-measurement
om
Nemours Children Health System. (n.d.). A to Z: Fracture, Lateral Malleolus (for Parents) -
.c
Nemours. Retrieved February 3, 2021, from
ud
lo
https://kidshealth.org/Nemours/en/parents/az-fracture-malleolus.html#:~:text=The%20knob%
ic
20on%20the%20outside,called%20a%20lateral%20malleolar%20fracture
@
ky
TeachMe Anatomy. (n.d.). The Femur - Proximal - Distal - Shaft - TeachMeAnatomy.
sa
https://teachmeanatomy.info/lower-limb/bones/femur/#:~:text=The%20femur%20is%20the%2
lg
ae
0only,%3B%20proximal%2C%20shaft%20and%20distal
ni
University of Southern Denmark Faculty of Health Sciences. (2016, November 2). New
ha
research on the muscles of elite athletes: When quality is better than quantity. ScienceDaily.
y
tif
https://www.sciencedaily.com/releases/2016/11/161102132208.htm
Cl