18.2.317-18 · Selection, Evolution, Classification and Conservation
Random Sampling — Cambridge AS & A Level Biology 9700 Revision Notes & Topical Questions
5 min read 10 parts Lesson 25 of 36 1 quiz
Lesson summary
18.2.3 Random Sampling — CAMBRIDGE AS & A LEVEL BIOLOGY 9700 — REVISION NOTES notes: Explain why sampling is used and how random sampling differs from systematic sampling.Measuring biodiversity needs a species list and population sizes, which means identifying and counting all or…
- Other methods
- Percentage cover
- Hypothesis
- Measuring biodiversity needs a species list and population sizes, which means identifying and counting all organisms
- This is possible for very small areas or very large species such as trees
- It is impossible to find, identify and count every organism in large complex ecosystems like rainforests
Section 1 of 10Why sampling is needed
1
Measuring biodiversity needs a species list and population sizes, which means identifying and counting all organisms.
2
This is possible for very small areas or very large species such as trees.
3
It is impossible to find, identify and count every organism in large complex ecosystems like rainforests.
4
Samples of the area are used to estimate total species numbers.
5
A sample must be representative, free from bias and large enough.
Section 2 of 10Types of sampling
Types of sampling
Sampling
- Sampling a method of investigating the abundance and distribution of species and populations.
Random sampling
- Random sampling the positions of the sampling points are completely random, decided by chance.
- Gives every part of the habitat an equal chance of selection.
- Avoids bias by the person sampling.
Systematic sampling
- Systematic sampling the positions of the sampling points are chosen by the person sampling.
- The person could be biased for or against certain areas.
- Quadrats may be placed where there are fewest species because they are quicker to count, which is unrepresentative.
Section 3 of 10When to use each method
1
Use random sampling when the area is reasonably uniform or species have no clear pattern of distribution.
2
Use systematic sampling (transects) when conditions change gradually, e.g. up a shore or away from a path.
3
Both can be combined, e.g. choosing the starting point of a transect at random.
4
Transects sample only a narrow strip, so several should start at random points.
Section 4 of 10Random sampling with quadrats
1
Mark out the study area with two long tapes at right angles forming the axes of a grid.
2
Label each grid square with a number, or use a random number generator or table to produce pairs of coordinates.
3
Place a quadrat (0.25 m2 or 1 m2) with one corner at each coordinate pair.
4
Record the species present and abundance in each quadrat, as a count or percentage cover.
5
Repeat for enough quadrats (at least 10 to 20) and calculate mean values.
Section 5 of 10Sample size and reliability
1
The sample size should be large enough for the mean to stabilise.
- Plot the running mean against number of quadrats and stop when the line levels off.
2
More quadrats reduce the effect of anomalies and allow statistical tests but take longer.
3
Total population can be estimated: mean density × total area.
4
Quadrat size should suit the organisms: small for dense grassland, larger for scattered shrubs.
5
Decide counting rules for plants on the quadrat edge before starting.
Section 6 of 10Sources of error and how to reduce them
| Source of error | How to reduce it |
|---|---|
| Bias in placing quadrats | Use random coordinates |
| Too few samples | Take more quadrats and calculate a mean |
| Identification errors | Use a key, check with an expert and use one observer |
| Trampling damage | Plan routes and avoid disturbing the area first |
Section 7 of 10Other methods and evaluation
Other methods
- Other methods pitfall traps, pooters, beating trays, sweep nets, kick sampling and mark-release-recapture.
Random placement may miss a rare patch by chance.
Percentage cover is quick for clumped plants such as grass or moss but more subjective than counting.
Keep quadrat size, time of year, observer and weather constant.
Random samples are needed for valid statistical tests such as Simpson's index and correlation tests.
Handle animals gently, return them and limit damage to the habitat.
Section 8 of 10More from the specification
Spatial sampling in an ecological niche
- Spatial sampling in an ecological niche deploying random quadrat coordinates ensures unbiased estimation of organism populations within each designated ecological niche.
Spearman's rank and Pearson's correlation
- Statistical correlation coefficients quantify the strength and direction of linear or monotonic associations between two paired ecological variables.
Lincoln index for mobile animal populations
- The mark-release-recapture method estimates population size using marked individuals, assuming random distribution without altered mortality.
Systematic vs random sampling with quadrats
- Random coordinates avoid investigator bias when estimating abundance,
- while systematic line transects investigate changes in species distribution along environmental gradients.
Spearman's rank and Pearson's correlation
- Statistical correlation coefficients quantify the strength and direction of linear or monotonic associations between two paired ecological variables.
Lincoln index for mobile animal populations
- The mark-release-recapture method estimates population size using marked individuals, assuming random distribution without altered mortality.
Systematic vs random sampling with quadrats
- Random coordinates avoid investigator bias when estimating abundance,
- while systematic line transects investigate changes in species distribution along environmental gradients.
Hypothesis
- A testable prediction with a scientific explanation.
Correlation coefficient
- Value of r or rs showing strength and direction of correlation.
Genetic diversity
- The range of alleles within a population.
Standard deviation
- Measure of spread around the mean.
Species richness
- The number of species in an ecosystem.
Lincoln index
- N = n1 × n2 ÷ m2, the estimate of population size from mark-release-recapture.
Section 9 of 10Key terms
| Term | Meaning |
|---|---|
| Sampling | Investigating abundance and distribution by examining parts of an area. |
| Random sampling | Sampling in which positions are decided by chance, avoiding bias. |
| Systematic sampling | Sampling in which positions are chosen by the investigator or at regular intervals. |
| Quadrat | Square frame used to mark off a sample area. |
| Bias | Distortion of results by the choices of the person sampling. |
| Percentage cover | Estimate of the percentage of area covered by a species. |
Section 10 of 10Exam technique
1
Calling quadrat placement random when positions are chosen by eye — random means generated by chance.
2
Using systematic sampling in a uniform area when random is best — and the reverse where there is a gradient.
3
Using too few quadrats — the sample must be large enough to give a stable mean.
4
Forgetting to control variables such as observer and quadrat size.
Marked practiceQuizzes for this lesson
Last updated: