• Like
  • Save
Upcoming SlideShare
Loading in...5







Total Views
Views on SlideShare
Embed Views



1 Embed 1

http://www.slideshare.net 1



Upload Details

Uploaded via as Microsoft PowerPoint

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
Post Comment
Edit your comment

    Sampling Sampling Presentation Transcript

    • Collecting Data. Sampling.
    • Why do we usually sample?
      • Is it possible to measure all the pebbles on a beach to work out the average pebble size?
      • Chances are the answer is no- therefore as geographers we can sample a number of the population to gain an insight into the patterns.
      • We can also sample in human geography as well.
    • Why is sampling desirable?
      • It is quicker than measuring every item.
      • It is cheaper
      • Often it is impossible to measure everything.
      • It is unnecessary to measure the whole population because a carefully chosen sample can give you a result which is close to the figure you would obtain even if you measured every pebble.
    • Why is sampling desirable?
      • We may wish to take a snapshot of the population at one moment in time- for example meteorological readings are taken simultaneously at only a limited number of sample sites in the UK and sent to the met office.
      • It is sometimes impossible to gain access to the complete population
        • For example if we asked a questionnaire to everybody in Plymouth about their shopping habits some could be quite in their rights to say no.
    • Why is sampling desirable?
      • We may not know the population size or location and therefore we would be forced to sample.
    • Avoiding Bias
      • Bias can arise for many reasons.
      • The data from which the sample is taken is biased- For example why would selecting the addresses of 100 people from a telephone directory be biased?
      • Insufficient care in the choice of sample may result in an unrepresentative data set- for example asking a spread of young and old people is better than only asking young or only old.
    • Avoiding Bias
      • 4. The time the sample was taken may produce bias- Why would asking a street questionnaire at 10:00 on a weekday be biased?
    • Sampling Methods.
      • Before selecting a sample it is necessary to decide on the sampling method.
      • When selecting samples from an area on the ground or from a map there is one decision to be made
        • Are you looking for points, lines or quadrats- these are known as the three spatial sampling methods
    • Point sampling
      • This involves choosing individual points and sampling those points, such as specific houses down a street or crop types on a land use map.
    • Line sampling
      • This involves taking measurements along a line- for example to sample vegetation across a series of sand dunes and note the dominant vegetation type along each part of the line.
    • Quadrat sampling.
      • Also known as area sampling
      • This involves marking a square on the ground and noting the occurrence of the feature you are interested in within the square.
    • How do we decide what to do?
      • There are three commonly used methods.
      • Random Sampling.
        • Decide how many sample points you want.
        • Obtain random numbers- either from a published sheet or from a calculator which generates random numbers.
        • Overlay a map of the area with grid lines and number them.
        • Use random numbers to read off grid references.
        • Find out the land use at each points chosen by visiting the places on the ground or from a land use map.
    • Random sampling.
      • Can be used for most types of sampling
      • The provide a means by which we can select samples in the knowledge that no human bias is involved.
      • The disadvantage is that if the sample size is small we might obtain an unrepresentative result.
    • 2. Stratified sampling
      • With this type of sampling we start by asking the question Are there subsets of the pattern being measured that must be included within our sample?
      • For example- we might think older people shop differently to young- therefore our sample must contain old and young people.
      • These subsets are known as strata.
    • 2. Stratified sampling
      • In order to sample this way you plot the points on a map as in random sampling but do not allow more than the allotted number of points to fall on each area of study
        • you then ignore an excess points which are chosen by random numbers.
    • 2. Stratified sampling
      • This has the great advantage in that it helps to reduce any bias which might possibly arise if samples were chosen completely at random.
      • The only problem with the method is identifying what the strata should be.
      • The difficulty is knowing for certain that the strata are relevant and how many strata to create within each- how many age division for example.
    • 3. Systematic sampling.
      • In this method the sample is chosen according to some agreed interval- i.e we visit every 5 th house or measure every tenth pebble.
      • The advantages of this method are great
        • It ensures complete coverage of the map.
        • It is simple to do.
      • The danger is that the systematic sample might inadvertently pick up some underlying regularity.
    • Sample size.
      • The size of the sample will usually be dictated by the time available.
      • The larger the sample the better the quality of the results.
      • Also your capacity to handle the data collected influences the size of the sample- for example computers now allow us to collect more data and produce accurate analysis and graphs.
    • Required sample size.
      • There is a way to calculate the best possible sample size- known as the required sample size.
      • We will use measuring pebbles on a beach to calculate this.
    • Required sample size.
      • Collect a pilot sample of 30 pebbles- Measure their lengths.
      • Calculate the standard deviation of the results- this is how far the results are away from the mean- another lesson for this.
      • Decide on a tolerable margin of error- how close you wish to be to the result you would obtain if you measured all the pebbles- e.g 0.5 cm
      • Decide on the level of certainty you wish to achieve- e.g 90% sure that the average length of the pebbles measured lies in the 0.5 cm margin or error.
      • Calculate using this formula
        • zs
        • n= d
      2 N= required sample size. Z= z score (a different table you aquire as a table of numbers) S =standard deviation. d= Tolerable margin of error.