Table 1 Raw data; density of plantains and dandelions in
two different localities, number of plants per ¼ m2
From the table 1 difference between means within one population in different localities could be found:
Δ between dandelions in two localities: 11,65 – 4,55 = 7,1
Δ between plantains in two localities: 10,2 – 3,6 = 6,6
As differences between two localities are quite large, these results imply that there is a strong influence of growing (soil) conditions on the populations. Of course, there is a possibility of error due to sampling (error caused by observing a sample instead of the whole population), although, difference is too large to be kept as error of sampling. To obtain more reliable results, more trials should have been done as well as more localities should have been investigated.
Δ between dandelions and plantains in sandy soil: 4,55 – 3,6 = 0,95
Δ between dandelions and plantains in clay soil: 11,65 – 10,2 = 1,45
If compared only arithmetical means of two populations, they seem to be quite small, however, to make more meaningful comparison between two populations, further comparison between SD’s and difference in means should be done:
SDs in sandy soil are 1,6 and 1,2 → they are bigger than 0,95;
SDs in clay soil are 1,63 and 1,6 → they are bigger than 1,45.
The value of calculating standart deviation
SDs can be used to help to decide whether the differences between the two related means (such as population of dandelions and plantains in same localities) are significant or not. Comparison between difference in means and SDs show that differences between average means are not significant as SDs are larger than the difference between means. This means, that two populations could be kept as of similar distribution. Data sets overlaps are shown in further processing of raw data:
Table 2 Frequency in sandy soil. Table 3 Frequency in clay soil.
Conclusion and evaluation
Results from the experiment proved my hypothesis that distribution of certain species (dandelion and plantain) is quite similar. Both species showed wider spread in clay damp soil than in sandy dry soil as the difference between dandelions populations (between average means) in different soils is 7,1 and between plantains – 6,6. Comparison between difference of means and SDs suggested that difference in means is not very significant as SDs were larger than the differences between means. What’s more, frequency charts (chart 2 and chart 3) indicated that two populations within the same locality do not differ much as well. However, although the difference between means might be kept as not significant, further statistical tests, such as t-test, could be used to reprove the hypothesis.
The limitations of the experimental method basically count weaknesses of sampling. Results might not be very fair due to that quadrate used was only ¼ of square meter; quadrate of square meter should be used instead. Besides quadrate’s size, randomness principle could also be erroneous, a map of a lawn could be draw and then places of measuring could be chosen. A human error could also have altered the results – sometimes it is too difficult to find the real number of organisms within the quadrate as some of the are hidden under higher neighbors or some are even to small to be noticed. Population depends not only on humidity of a soil and light intensity, there are such factors as people activity, pollution, characteristics of soil and even near-growing plants. Hence, characterizing the lawn more precisely would let make more accurate conclusions. There may be even unknown factors that have an effect on populations, therefore, more lawns and fields of similar features should be tested in order to obtain more reliable results. One more factor that may have changed the recordings is the time of a season. Generalizations that two popularities are quite similar could be erroneous as I fulfilled an experiment at a particular time and did not take into the account that maybe after a period of a few months one of the populations will have been grew into 3 times larger. For more precise results trials should be taken at least every week for a period of a month.