I can also use Spearman’s rank correlation coefficient value to see the correlation between the two sets of ranks. A coefficient value of ‘0’ means that there would be no correlation between the 2 sets of ranks. A coefficient value of ‘1’ means that there would be perfect positive correlation between the 2 sets of ranks. A coefficient value of ‘-1’ means that there would be perfect negative correlation between the 2 sets of ranks. Any value in between can be described as strong negative/positive or weak negative/positive correlation.
I predict that as the size of the rock pools increase the diversity of the animals in the rock pools will increase. This is because small rock pools have harsher environments than larger rock pools. During hot weather the small rock pools are heated up by the sun more quickly than the large rock pools. Even though in large rock pools there would be more competition between animals, it shouldn’t have much effect on the diversity of the animals in the large rock pool.
Null Hypothesis – There is no correlation between the size of the rock pools and the diversity of the animals in the rock pools.
Experimental Hypothesis – There is a correlation between the size of the rock pools and the diversity of animals in the rock pools.
Equipment
- Tape measure – To measure the lengths and widths of rock pools.
- Meter ruler – To measure the depth of the rock pool.
- Fishing net – To take out unidentified species so they can be identified later.
- Bucket – To put the unidentified species in.
- Identification chart – To identify unknown species of animals
- Clip board and record sheet – To record the results on.
Method
- Collect the equipment listed above.
- Find a suitable rock pool on the middle shore. Try to find pools of various sizes. This is so that we have a number of sizes of rock pools to compare.
- Measure the largest length and the largest width of the rock pool using a tape measure. Then, using a meter ruler measure the greatest depth of the rock pool. Record these dimensions on to the record sheet. These values will be later used to estimate the volume of the rock pools.
- Identify all the animals of each species using the identification chart and list them on the record sheet.
- Count all the different animals of each species and record it on the record sheet.
- The bucket should be ¾ full with water. If there are any unidentified species of animals, take them out of the rock pool using a fishing net and put it in the bucket so they can be identified later on. If the species of that animal is still unidentified, name them species A, B, C … and so on.
- Repeat the procedure until a substantial number of rock pools have been investigated. An ideal number would be around 15 rock pools.
- The independent variable is the size of rock pools and the dependant variable is the diversity of animals in the rock pools.
Fair Test
To make it a fair test I will do the following:
- I will count one species at a time. This would be the most suitable way of counting species as it is difficult to count more than one species at a time and there is more chance of error.
- When I have counted one species of animals I will recount to make sure that I counted correctly the first time.
- I am going to be doing the experiment on the middle shore only. This is because there is going to be a change in diversity of animals in the middle, upper and lower shore. The diversity will change because of the different environment at upper, middle and lower shores and some animals can not survive in these environments.
- Any unidentified species will be put in the bucket so they can be identified later. This is important because animals of the same species might look a bit different sometimes (For example, might be different colours) and can not be counted as different species of animals. This would effect the calculations leading into errors.
Implementation
The experiment was carried out and the results are shown on the following page. From the data I have collected I am going to calculate the following three values:
1. Calculating the Volume of the Rock Pools
The length, width and depth of the rock pools were measured in metres during the experiment because these are the values which are required to calculate the volume of the rock pools.
Length x Width x Depth
The unit of the volume is ‘m3’.
2. Calculating the Diversity of the Rock Pools
Diversity is a way of describing the relationship between the number of individuals and the number of species in a community.
D = N (N – 1) a
∑ N (n – 1)
N = total number of organisms of all species
n = total number of organism of particular species
3. Calculating Spearmans Rank Correlation Coefficient
Spearman Rank Correlation Coefficient is a value, between -1 and 1, which is used to see how 2 sets of ranks correlate.
Spearmans Coefficient = 1 – 6 D2
N (N2 – 1)
The values of diversity of each rock pool (calculations shown) and the volume of each rock pool were summarized in the results table in the following page.
Spearmans Rank Correlation Coefficient
I have tabulated the information I need to calculate the Spearman’s Coefficient.
Spearmans Coefficient = 1 – 6 D2
N (N2 – 1)
= 1 – (6 x 446)
15 (152 – 1)
= 1 – 2676
3360
Observed Value = 0.203 (to 3 s.f)
For a 1 tailed test, at 0.05 significance with sample size N = 15 has the critical value of 0.443. This can be gathered from the ‘Critical Values of Spearman’s Rank Order Correlation Coefficient’ table. The observed value is 0.203 which is lower than the critical value.
Conclusion
Since the calculated value of spearman’s rank correlation coefficient is less than the critical value we can accept the null hypothesis and reject the experimental hypothesis. Therefore, the results have shown no pattern between the size of the rock pools and the diversity of animals in the rock pools. However, the data does indeed show that the size of the rock pool and the diversity of the animals in the rock pool have a weak positive correlation. This is confirmed by the scatter diagram.
As you can see in the results table, the smallest rock pool has the smallest diversity value. This matches my prediction as I said that bigger the rock pool, bigger the diversity. But the biggest diversity value was calculated in a fairly small rock pool. This doesn’t match my prediction. This has happened because the population of that rock pool is low and it has got more different number of species. Since diversity is not just the relationship between the numbers of individuals, it involves relationships between the different species as well the diversity value for that particular rock pool was high.
There are some rock pools which follow my prediction but there isn’t a general pattern in the results that I have obtained. This could be because of possible errors in my experiment or my prediction could be wrong. The size of the rock pools isn’t the only factor which affects the diversity of the animals in the rock pool. However, the size of the rock pool has effect on the some abiotic and biotic factors.
Evaluation
I believe that my experiment on “the effect of the size of the rock pools on the diversity of the animals in the rock pools” went reasonably well. However, there were some errors in the experiment which could have been improved ensuing in more accurate results.
There was one anomalous result. When the size of the rock pool is 0.087 cm3, the diversity of the rock pools is 7. Even though the population of the rock pool was 7, which is very small, there were more different species of animals: the rock pool was more diverse. Since, Diversity is a way of describing the relationship between the number of individuals and the number of species in a community.
Firstly, the sizes of the rock pools were measured by using the formula for a cube. The biggest length, width and depth of the rock pools were taken. Therefore, rock pools weren’t a perfect cube and so the calculations of the volume of the rock pools aren’t accurate. This can be improved by taking various lengths, widths and depths and then calculating the average length, width and depth for a particular rock pool. This would be more accurate than the method I used but still not 100 % accurate. I could use ‘high-tech’ equipment to calculate the exact volume of the rock pool but it would be expensive and time consuming.
Although I was very careful while counting the different number of species in the rock pool I could have easily made a mistake. It was difficult to measure the animals in the rock pool because sometimes there were too many of one species. I might have counted the same animal twice and I might have overlooked some animals in the rock pool. There might have been some miscalculations in measuring the different number of species in the rock pool which would have effect on the diversity index of animals in the rock pools.
More rock pools could have been observed because increasing the sample size increases the chance of getting accurate results. I had limited time to carry out the experiment because the experiment had to be done on the same day. Therefore, only 15 rock pools could be observed. However, if I had assistance from someone else, more rock pools could have been observed and this would give me an even wider range of rock pools to compare.
Another way to get more accurate results is that I could have taken a different approach to select the rock pools. I could set different ranges which the sizes of the rock pools had to be in between. For example, I could set 5 different ranges: 0.00 – 0.30 cm3, 0.40 – 0.60 cm3, 0.70 – 0.90 cm3, 1.00 – 1.20 cm3, 1.30 – 1.50 cm3. Then I could select 3 rock pools for each of these ranges that I have set, i.e., 3 rock pools with sizes between 0.00 and 0.30 cm3 and 3 rock pools for each of the other ranges. This would give more accurate results as I have observed rock pools which vary in sizes more. In the experiment that I carried out I selected any rock pools on the middle shore. This way, I observed rock pools which didn’t vary much in sizes and this change in size might not be enough to have any change in diversity of the animals in the rock pools.