Research Design, Sampling and Recruitment

Question 1

Sample

Answer 1

Fraction of a population that can represent the whole population.

Question 2

Sampling

Answer 2

Sampling is the process of selecting a sample from the population to be studied.

Question 3

2 Categories of Sampling

Answer 3

Probability sampling method

Non-probability sampling method

Question 4

Probability Sampling Method

Answer 4

All entities (persons) in the target population have equal chance of being selected. 
4 types:
- Simple random sampling
- Stratified random sampling
- Systematic random sampling
- Clustered random sampling

Question 5

Simple Random Sampling

Answer 5

The whole population is available for selection.

Use lottery/balloting or computer-generated random sample.

Question 6

Stratified Random Sampling

Answer 6

Same as simple random but samples are selected from subgroups e.g. age groups, levels of severity.
For example: 10 participants needed. Population: 30, 12 men, 18 women. Men and women separated, 4 men chosen and 6 women.

Question 7

Systematic Random Sampling

Answer 7

Selection at defined intervals. For example: 3rd patient to be discharged every hour. 1st patient to arrive on Mondays and Fridays.

Question 8

Clustered Random Sampling

Answer 8

For very large population that require subdivisions into geographic areas. For example: South is more populated than north. So out of 100, 45 will be chosen from south, 10 from north, 25 from east and 20 from west.

Question 9

Non-Probability Sampling Method

Answer 9

Selection does not guarantee equal chances of selection for all entities in the target population. 
3 types: 
- Convenience sampling
- Purposive sampling
- Snowball sampling

Question 10

Convenience Sampling

Answer 10

Selection based on availability.
Used in qualitative research.
May also be used in case series research.

Question 11

Purposive Sampling

Answer 11

Same as convenience sampling but participants have to be experts (as in experts in their thoughts, emotions, feelings, perceptions, etc.)
Selection is based on pre-defined characteristics and it is the judgement of the researcher that a potential participant is an expert.
Must be used in qualitative research, may be combined with random sampling for large populations.

Question 12

Snowball Sampling

Answer 12

Using a participant you already have to get more participants.
When the population is difficult to access.

Question 13

Delimitations of a study

Answer 13

Delimitations are the boundaries of your research study.
Identify potential weaknesses.
- Inclusion/exclusion criteria (sampling).
- Research instruments and procedure of administering them.

Question 14

What is the primary sampling method for qualitative research?

Answer 14

Purposive sampling.

Question 15

Cross-sectional study design

Answer 15

• Research aims to document the frequency of occurrence per unit of time.
• Driven by specific aims/objectives.
• Large-scale surveys approach.
• Research outcomes are used to shape and inform health and social care policy.
  Example: How many people have spinal chord injury every year?
  Sampling not required if population register is available.
  Otherwise clustered random sampling.

Question 16

Economic analysis

Answer 16

• Aims to quantify the resources required to deal with disease/injury as a healthcare problem.
• Used to shape health and social care policy and to determine insurance pay-outs.
• Aims to attach monetary value to everything, including caring.
  Example: How much does a spinal chord injury cost?
  Sampling not required.

Question 17

Action research

Answer 17

Simultaneous enquiry into local practices and practical action to change local practices.
A combination of qualitative and quantitative research also known as mixed methods research.
Example: How can local care be improved for spinal chord injury patients?
Qualitative: Sampling not required.
Quantitative: Purposive sampling.

Question 18

Recruitment methods

Answer 18

Cold Call
Advertisement
Direct Recruitment
Snowball Sampling

Question 19

Cold call

Answer 19

Letter, email, phone call to potential participants who have never heard about the study, inform them of the purpose and ask for consent.
Suitable for:
- Qualitative research.
- Qualitative part of action research.

Question 20

Advertisements

Answer 20

Place adverts in newspaper, internet, television, magazines etc.
Posters and fliers.
Target population must regularly access the medium that is being used for the advertisement.
Suitable for:
- RCT, qualitative.
- Any research that needs human participants.

Question 21

Direct recruitment

Answer 21

Invite patients directly from clinics and hospitals.
Researcher collaborates with healthcare professionals. Healthcare professionals recruit participants on behalf of researcher (researcher can be the healthcare professional).
Suitable for:
- Non-randomized control trial, qualitative.

Question 22

Snowball Recruitment

Answer 22

Ask current participants to refer their friends and relatives to you.

Question 23

Ethical Principals in Research Participant Recruitment

Answer 23

• Give information to potential participant.
• Capacity to consent.
• Participant gets time (2 weeks) to consider the information.
• Signing of informed consent.

Question 24

Information to give potential participant.

Answer 24

• What participant is required to do.
• How research will benefit participant and society.
• Compensation/incentives that participant will receive.
• Participants right to confidentiality and privacy.
• Participants right to withdraw consent at any time.

Question 25

Children and others ,that cannot give consent, as participants.

Answer 25

Child: Get consent from parents/guardian. Researcher and parent/guardian are sure child understood the information.
Others: Check for an advance directive, if not get consent from guardian.

Question 26

Recruitment difficulties

Answer 26

Modify or adapt:

• Research question.
• Research method.
• Reduce participant burden, e.g. transportation.
• Give incentive.