Sampling

• What is sampling?


  Gathering opinions from a number of people chosen from a specific group to find out about the whole group
• Why is data not collected from the entire population of a market?


  Too expensive and time-consuming
• The chosen group should be _______ of the target market


  Representative
• What happens to the sample size and design complexity of the sampling design if the market research REQUIRES PRECISION?


  More precision market research requires= More complex design=Larger sample size
• What can the units in the population be identified by?


  Existing information e.g government registers
• How is a target population sampled?


  Using a sampling frame
• A good sample design ensures specification of _____ population is as _____ as possible to ensure all elements within _____ are ________.


  Target, clear, population, represented
• Give an example of a geographic sampling frame?


  E.g. using postcodes
• Do sampling design covers have to be proportionally representative- why/not?


  NO because for example a larger sample may be drawn from a minority group to ensure sufficient data collected for analysis.
• What are the pros of random sampling?


  -Reduced sample bias (in a way selecting by judgement or convenience cannot)

  -Allows justifiable inference from the sample to population to be made
• SAMPLE SIZE- what will assumptions about a population have?


  An attached margin of error
• How can the margin of error be reduced?


  Using a larger sample size
• What is the confidence level?


  The likelihood the results obtained from sample lie within a required precision
• What confidence level to statistians use to provide strong conclusions?


  95%
• Does population size normally affect sample size?


  NO!!

  (In fact larger population= lower proportion of population needs to be sampled to be representative)
• TYPES OF SAMPLING: Non-probability- what does this include?


  Cluster, quota and convenience sampling
• What is non-probability sampling?


  Researcher deliberately picks items/ individuals for sample based on non-random factors like convenience or cost
• What is cluster sampling?


  -Units in population can be found in certain geographic groups or 'clusters' (e.g primary school children in Harrogate)

  -A random sample of clusters is taken and all units within cluster examined
• What is convenience sampling?


  Uses those willing to volunteer and easiest to involve in study
• What is quota sampling?


  Population divided (stratified) by MOST IMPORTANT variables (e.g age)- the required quota sample is then drawn from each stratum
• Non-probability sampling: What is quota samplings aim?


  To obtain a sample that is representative of overall population
• What is probability sampling?


  AKA random sampling

  Randomisation used instead of deliberate choice, each member of population has a known, non-zero chance of being selected

  Ensures every member has equal chance of selection
• What are the pros of sampling?


  -Indication of likely behaviour of whole market

  -Avoid expensive marketing errors (e.g before launch of new product)

  -firms can learn about its market quickly (act quickly and get ahead of competitors)

  Reliable info from small cross-section of people
• What are the drawbacks of sampling?


  -Samples may be unrepresentative

  -may be bias in questions/answers they encourage (e.g Do you brush your teeth everyday?).
• What type of research is non/probability sampling used in?


  Qualitative- non- probability

  Quantitative- probability