Essential Question:

How does evidence from fossils help to understand early civilizations?

Research Information

Fossils are preserved remains of any organism from the past. Fossils provide important evidence regarding the organism’s past. Specimens typically have to be over 10,000 years old in order to be considered a fossil. Fossils show a progression of evolution, as the organisms from the past are not the same as those of today. Some examples of fossils include; shells, bones, stone prints, DNA remains, oil, coal, and hair. There are two main categories of fossils: body fossils and trace fossils. Both kinds are the remains of living organisms however, body fossils display the body formation of the organism whereas trace fossils show the action/life of the organism.

The study of human bones is not just restricted to the field of osteology but also includes other areas of research, such as archaeology and forensic anthropology. A major aspect of these studies is that they investigate human skeletal remains in order to recreate the past, comprehend human variation, and provide information about the deceased person such as age, sex, ancestry, mental illnesses or traumatic injuries.

A standard inspection will involve:

– Catalog of present skeletal elements

– Dental inventory

– Ancestry

– Pathology, cultural modifications

– Height and other metric data

– Non-metric data

– Aging data

Researchers use stratified random sampling to acquire a sample population that most appropriately depicts the population being studied. Stratified random sampling involves sampling a percentage of the population so that all groups within that population are represented. This methodology has its advantages and disadvantages. For example, it minimizes selection bias and ensures that subgroups within the population receive genuine representation within the data, but it also includes many conditions that must be met in order to be used correctly. Some of the conditions include that every member of a population must be studied and each individual must be categorized into a subpopulation. Finding an all-inclusive list is just one of many problems. Another is correctly categorizing each member of the population into a single class. Although this may seem fairly simple with definitive examples such as, male and female, this become much more challenging when you factor in race, ethnicity, religion etc. This selection process become increasingly difficult as you increase the number of variables, showing that this is an inadequate method.

With all of this in mind, my group decided not to use stratified random sampling and to do convenience sampling. The subjects for this experiment are all female students from Grade 9 – 12 at Elmwood School. Having an all female student sample simplifies the technique as we only need one equation since we would be removing the male factor. A disadvantage to this method is that it allows much more room for bias. For instance, you may know some people better than others and would be more inclined to ask those people for their measurements. Similarly, we collected much more data from grade 9 students which suggests that we were more comfortable asking people in our own grade than others (convenience), which, in turn affects our end results.

Raw Data:

Part I:

Figure 1: The graph above shows the relationship between females’ tibia lengths and heights.

The independent variable in the equation is the tibia length and the dependent variable in this equation is the height. The unit of measurement for both variables is in centimetres in order to keep the data constant.

The subjects’ heights were measured against a wall, with a book placed on their head. Their height would then be measured from the bottom of the subject’s foot to the bottom of the book. The subjects’ tibias were measured by a measuring tape being extended from the ankle bone to the indent where the tibia meets the femur. The measurements in this survey were collected using direct observation. This may be a source of error or inaccuracy.

The linear regression equation is H = 0.7558t + 137.17. The correlation coefficient value is r = 0.32. Using r2 that Excel provided my group and I calculated the square root to get the coefficient correlation. As shown in the graph above, the correlation between tibia length and height for Elmwood students grades 9-12 is a weak positive correlation.

I will be testing my equation on 5 different subjects to test whether or not the formulas given by the archaeologists are accurate.

Archaeologists Formula Formula from Graph

Trial #1: Trial #1:

Subject #6 Subject #6

Tibia length (cm): 38.1 cm Tibia length (cm): 38.1 cm

Height (cm): 171.45 cm Height (cm): 171.45 cm

H = 2.50t +74.70 H = 0.7558t + 137.17

H = 95.25 + 74.70 H = 28.8 + 137.17

H = 169.95 cm H = 165.97 cm

Trial #2: Trial #2:

Subject #18 Subject #18

Tibia length (cm): 36 cm Tibia length (cm): 36 cm

Height (cm): 171 cm Height (cm): 171 cm

H = 2.50t +74.70 H = 0.7558t + 137.17

H = 90 + 74.70 H = 27.21 + 137.17

H = 164.7 cm H = 164.38 cm

Trial #3: Trial #3:

Subject #21 Subject #21

Tibia length (cm): 33 cm Tibia length (cm): 33 cm

Height (cm): 160 cm Height (cm): 160 cm

H = 2.50t + 74.70 H = 0.7558t + 137.17

H = 82.5 + 74.70 H = 24.94 + 137.17

H = 157.2 cm H = 162.11 cm

Trial #4: Trial #4:

Subject #36 Subject #36

Tibia length (cm): 39 cm Tibia length (cm): 39 cm

Height (cm): 161cm Height (cm): 161cm

H = 2.50t + 74.70 H = 0.7558t + 137.17

H = 97.5 + 74.70 H = 29.5 + 137.17

H = 172.2 cm H = 166.67 cm

Trial #5: Trial #5:

Subject #53 Subject #53

Tibia length (cm): 36.5 cm Tibia length (cm): 36.5 cm

Height (cm): 177 cm Height (cm): 177 cm

H = 2.50t + 74.70 H = 0.7558t + 137.17

H = 91.25 + 74.70 H = 27.6 + 137.17

H = 165.95 cm H = 164.77 cm

Based on my trials, I can conclude that the correlation was close. The equation from the graph was closer to the actual height of the subjects. The correlation for the archaeologists’ formula was approximately 7.5 cm off and my formula was only 6.1 cm.

The archaeologists’ formula did not work with my collected data. This could be attributable to the fact that the archaeologists possibly creating these equations based on remains of a particular ethnic group. So, when we collected data from a diverse range of girls, it could have caused issues with using the equation determined by the archaeologists.

Another possible error could be within the methods we used to collect data for our survey. When measuring the subjects’ height, most students were wearing shoes and we also didn’t ask them to stand up straight. This could serve to create anomalies within our data. It was also quite difficult to measure subjects’ tibia when they were wearing tights or sweatpants as it was hard to locate where the tibia met the femur and where their ankle bone was situated.

Part II:

Raw data :

The dependent variable in this equation is the head length (cm) and the independent variable is the height (cm). In the equation, the subject’s head length will be represented by ‘h’ and height will be represented by ‘H’ (H = 7h). We used this equation to test the accuracy of Leonardo Da Vinci’s theory that the average human is 7 heads high.

Figure 2: Figure 2 displays the relationship between subjects’ height (cm) versus the equation (H = 7h) applied.

The linear regression equation was (H =7h). The correlation coefficient value is r = 0.3, this shows a weak positive correlation. This result tells me that the statement “the average human is 7 heads high” is only somewhat true.

Part III: Designing an Investigation

Raw Data:

Figure 3: The graph shown above is a representation of the relationship between hand length (cm) and head width (cm)

The independent variable is the hand length (cm) and the dependent variable is the head width (cm). In order to collect our data we measured the subjects’ hand from their middle finger to the bottom of their palm and measured their heads by asking them to cover their ears and measured from one palm to the other. When we input our data into a scatter plot we were given the value of r2 . We square rooted this value to get the correlation coefficient. The correlation coefficient displayed on this graph is 0.14. This relation is exceptionally weak.

Analysis

This experiment was intrinsically flawed. As displayed through the tables and graphs, there were many areas that presented as potential sources of error. For example, the formula that was created by archaeologists was designed for adult females. Our sample only included the measurements of students from grade 9-12 which created room for error as many students could still be growing. In other words, we are not comparing identical samples so it may not be an accurate comparison. Measuring was also a major issue. Whilst measuring subjects, much of the data collected was just from direct observation and as a result, the measurements may not be completely accurate. In addition, it was also quite difficult for my group and I to actually pinpoint where the tibia met the femur as some subjects were wearing sweatpants, tights etc. This may not have been an issue for the archaeologists as they were measuring an exposed tibia of a skeleton whereas we were using live subjects. If I were to do this experiment again, I would focus more on time management. Throughout the experiment I felt that I was struggling with managing my time and I was worried that I would not get to finish my report.