Studying Populations (DP IB Environmental Systems & Societies (ESS))

Sampling Strategies

Populations and samples

What is a population?

• A population refers to the whole set of things that you are interested in

  • e.g. if a teacher wanted to know how long pupils in year 11 at their school spent revising each week then the population would be all the year 11 pupils at the school

• Population does not necessarily refer to a number of people or animals

  • e.g.  if an IT expert wanted to investigate the speed of mobile phones then the population would be all the different makes and models of mobile phones in the world

What is a sample?

• A sample refers to a selected part (i.e. a subset) of the population that data is collected from

  • e.g.  for the teacher investigating year 11 revision times, a sample would be a certain number of pupils from year 11

• A random sample is where every item in the population has an equal chance of being selected

  • e.g.  every pupil in year 11 would have the same chance of being selected for the teacher's sample

• A biased sample is where the sample is not random

  • e.g.  the teacher asks pupils from just one class

What are the advantages and disadvantages of using a population?

• You may see or hear the word census - this is when data is collected from every member of the whole population

• The advantages of using a population include:

  • Accurate results - as every member/item of the population is used

  • All options/opinions/responses will be included in the results

• The disadvantages of using a population include:

  • Time consuming to collect the data

  • Expensive due to the large numbers involved

  • Large amounts of data to organise and analyse

What are the advantages and disadvantages of using a sample? 

• The advantages of using a sample include:

  • Quicker to collect the data

  • Cheaper as not so much work involved

  • Less data to organise and analyse

• The disadvantages of using a sample include:

  • A small sample size can lead to unreliable results

    • Sampling methods can usually be improved by taking a larger sample size

  • A sample can introduce bias

    • Particularly if the sample is not random

  • A sample might not be representative of the population

    • Only a selection of options/opinions/responses might be accounted for 

    • The members/items used in the sample may all have similar responses

    • e.g. even with a random sample, it may be possible that the teacher happens to select pupils for their sample who all happen to do very little revision

• It is important to recognise that different samples (from the same population) may produce different results

Random and systematic sampling strategies

• There are two different types of sampling:

  • Random

  • Systematic

• In random sampling, the positions of the sampling points are completely random or due to chance

  • For example, sampling points can be selected using a random number generator to create a set of random coordinates 

  • This method is beneficial because it means there will be no bias by the person who is carrying out the sampling that may affect the results (i.e. there will be no researcher bias)

  • Random sampling can be used when the population size or the individual sample size is relatively small, and all individuals have an equal chance of being sampled

• In systematic sampling, the positions of the sampling points are chosen by the person carrying out the sampling and a regular pattern is used to select sample points

  • There is a possibility that the person choosing could show bias towards or against certain areas

  • Individuals may deliberately place the quadrats in areas with the least species as these will be easier and quicker to count

  • This is unrepresentative of the whole area

• When a sampling area is reasonably uniform or has no clear pattern to the way the species are distributed, random sampling is the best choice

Transect sampling

• Systematic sampling allows researchers to investigate the effect of the presence of certain environmental features on species distribution e.g. by taking samples along a line that extends away from, or along, an environmental feature, such as a river

  • A line of this type is known as a transect

• Transect sampling is used when there is a clear change in the physical conditions across the area being studied

  • For example, there may be changes in altitude, soil pH or light intensity

  • Methods using transects can help show how species distribution changes with the different physical conditions in the area

  • A transect is a line represented by a measuring tape, along which samples are taken

• For a line transect:

  • Lay out a measuring tape in a straight line across the sample area

  • At equal distances along the tape, record the identity of the organisms that touch the line (e.g. every 2 m)

• For a belt transect:

  • Place quadrats at regular intervals along the tape and record the abundance or percentage cover of each species within each quadrat
    

    

Quadrat Sampling

• Quadrats are square frames made of wood or wire

• They can be a variety of sizes e.g. 0.25 m² or 1 m²

• They are placed on the ground and the organisms within them are recorded

• Non-motile organisms such as plant species are commonly studied using random quadrat sampling to estimate their population size

• Quadrats can be used to estimate population size by recording:

  • The number of an individual species: the total number of individuals of a single species (e.g. daisies) is recorded

  • Percentage cover: the approximate percentage of the quadrat area in which an individual species is found is recorded (this method is often used when it is difficult to count individuals of the plant species being recorded eg. grass or moss)

Estimating population size

• Quadrats must be laid randomly in the area to avoid sampling bias

  • This random sampling can be done by converting the sampling area into a grid format and labelling each square on the grid with a number

  • Then a random number generator is used to pick the sample points

• Once the quadrat has been laid on the chosen sample point the abundance of all the different species present can be recorded

Estimating percentage cover and percentage frequency

• Percentage cover is an estimate of the area within a given quadrat covered by the plant or animal being sampled

  • Percentage frequency is the number of squares in which the species occurs divided by the number of possible occurrences

% frequency = (number of quadrat squares in which species present ÷ total number of quadrat squares) × 100

• This can be useful, as it can sometimes be difficult to count individual plants or organisms within a quadrat

• For example, if grass is found in 89 out of 100 squares in the quadrat then it has a percentage frequency of 89%

  • This process could be repeated for a series of quadrats within a given sample area

  • This information could then be used to calculate the average percentage cover across all the sampled quadrats

  

Capture–mark–release–recapture & the Lincoln Index

• The sampling methods described above are only useful for non-motile (sessile) organisms

  • Different methods are required for estimating the number of individuals in a population of motile animals (i.e. animals that are mobile)

  • The capture-mark-release-recapture method is commonly used alongside the Lincoln index (a statistical measure used to estimate population size)

  • The Lincoln index can be used to estimate the abundance or population size of a species in a given area

  • First, the capture-mark-release-recapture technique is carried out

• For a single species in the area:

  • The first large sample is taken—as many individuals as possible are caught, counted and marked in a way that won’t affect their survival e.g. if studying a species of beetle, a small amount of brightly coloured non-toxic paint can be applied to their carapace (shell)

  • The marked individuals are returned to their habitat and allowed to randomly mix with the rest of the population

  • When a sufficient amount of time has passed another large sample is captured

  • The number of marked and unmarked individuals within the sample are counted

  • The proportion of marked to unmarked individuals is used to calculate an estimate of the population size (the Lincoln index)

• The formula for calculating the Lincoln index is:

Population size estimate = (M × N) ÷ R

• Where:

  • M = number of individuals caught in the first sample (i.e. number of marked individuals released)

  • N = number of marked and unmarked individuals caught in the second sample (i.e. total number of individuals recaptured)

  • R = number of marked individuals in the second sample (i.e. number of marked individuals recaptured)

Limitations of using the capture-mark-release-recapture method

• When using the mark-release-capture method, there are a few assumptions that have to be made:

  • The marked individuals must be given sufficient time to disperse and mix back in fully with the main population - this can be time-consuming 

  • The marking doesn't affect the survival rates of the marked individuals (e.g. doesn't make them more visible and therefore more likely to be predated)

  • The marking remains visible throughout the sampling and doesn't rub off - this is often difficult to ensure and so the accuracy of population size estimates may be negatively affected

  • The population stays the same size during the study period (i.e. there are no significant changes in population size due to births and deaths and there are no migrations into or out of the main population) - again, this is almost impossible to ensure, further affecting the accuracy of population size estimates