Research Process 4- Sampling

Question 1

What is the total population ?

Answer 1

• Referred to as the study’s population, target or wider population.
• Have at least one population or characteristic in common in order to make them suitable for the study
• Identify the size of the population

Question 2

What is a sample ?

Answer 2

• A sub-group of the total population

- Participants that actually take part in the study

Question 3

What is representativeness ?

Answer 3

• Commonly associated with quantitative research.
• The intention is to ensure that the participants who take part in the study have the same characteristics as those in the study’s population.
• The sample is as close as possible to represent the total population
• Allows the researcher to generalise the study findings
• In qualitative research the word normally used is reflect.

Question 4

What is generalisability ?

Answer 4

• Closely associated with quantitative research
• When the study findings are generalised or applied to the study’s population
• In order to generalise the sample must be a representative of the population
• Generalisability requires representativeness

Question 5

What is transferability ?

Answer 5

• Qualitative research associated

- Not possible to generalise qualitative data

Question 6

What is data saturation

Answer 6

• Association with qualitative research and grounded theory
• New data or information is discovered both in terms of the depth and breadth of the data.
• The point at which saturation is reached can usually be pinpointed in retrospect, it is often used to determine the size of a sample.

Question 7

What is homo/heterogenous sample ?

Answer 7

• Homo- single or narrow range of characteristics

- Hetro- wide range of characteristics.

Question 8

What are the sampling strategies ?

Answer 8

• The researcher first needs to decide if it is feasible to involve the total population or if a sample of the population is required.
• It may be unethical and resoursely restricted to involve the whole population in a study if the findings can be confirmed by involving just a sample of a population.
• Two broad sampling strategies:
  o Probability sampling- quantitative
  o Non-probability sampling-qualitative
  o Also systematic sampling a strategy uses both probability and non-probability approaches.

Question 9

What is probability sampling ?

Answer 9

• Selection of a proportion of the population using random procedures
• Random sampling: everyone in the study’s total population has an equal chance to being included.
• This way the sample is representative of the population and so generalisable.

Question 10

What are the three common probability sampling ?

Answer 10

• Simple random sampling
• Stratified random sampling
• Cluster sampling

Question 11

What is simple random sampling ?

Answer 11

• each potential participant has an equal chance of being included in the sample
• usually selection of the sample done by computer program or a random table.

Question 12

What is stratified random sampling ?

Answer 12

• The population is divided into sub-groups from which the sample is selected randomly.
• Using this strategy can be more representative of the total population

Question 13

What is cluster sampling ?

Answer 13

• When the study population is divided into sub-groups and the clusters are then selected randomly
• Either the whole cluster participated in the study or participants may be randomly selected from the cluster.

Question 14

What is non-probability sampling ?

Answer 14

• Non-random procedures
• No intention to generalise the findings to the study population
• Sample size is usually determined by data saturation
• Perceived to be biased

Question 15

What are the common non-probability sampling strategies ?

Answer 15

• Convenience sampling
• Purposive sampling
• Quota sampling
• Snowball sampling
• Theoretical sampling

Question 16

What is convenience sampling ?

Answer 16

• Researcher recruits the most readily available participants who meets the study’s inclusion criteria.
• A risk that the sample will not reflect the characteristics of the population.

Question 17

What is purposive sampling ?

Answer 17

• Usually used in the early stages of the grounded theory or phenomenological strategy
• Participants who meet the study’s inclusion are recruited
• The researcher should constantly monitor the configurations of the sample and should attempt to recruit individuals to address any shortfalls of the sample
• Can be difficult if the research was very specific

Question 18

What is quota sampling ?

Answer 18

• The researcher pre-specifies the required characteristics of the sample to ensure the final sample includes a certain number with each characteristics

Question 19

What is snowball sampling ?

Answer 19

• Involves the identification of potential participants through referrals from earlier participants
• Useful way of accessing difficult to reach participants.

Question 20

What is theoretical sampling ?

Answer 20

• Researchers using concepts and ideas rather than participants.
• Enables the researcher to collect data from a range of sources

Question 21

What is systematic sampling ?

Answer 21

• Sometimes referred to as list sampling
• Strategy starts with the generation of a list of all participants in the population.
• The selection of the first participant is undertaken randomly and from then on, every nth participant is selected for the study.
• Therefore the inclusion of a participant is determined by their position of the list.

Question 22

What are the important questions ?

Answer 22

• Important question: who should take part in the study and who should not determined by the study’s aims and objectives.
• How many participants do I need?
• The sample size need to be large enough to achieve the research aims but a sample that is too large in unethical and costly.
• In quantitative research the sample size is known before the study starts, whilst qualitative research, the sample size is known once the study evolves.
• Quantitative research: the larger the group  the more accurate the findings are and therefore generalisable. Should be determined by a power calculation. Done by the statistical team. The calculation is based on outcome measures and measurement tools (etc).
• Qualitative data is usually determined by data saturation: therefore can only indicate the likely number of participants at the beginning. Determining the point at which data has saturated can be hard.