Sampling & Setting

Question 1

What is the difference between population and sample?

Answer 1

Population is the broad population we want to study (e.g. palliative patients), and sample is a group of people from said population to gain insight from (e.g. palliative patients at Mission hospice)

Question 2

What factors need to be considered in determining sample size?

Answer 2

1) Type of design
2) Type of sampling
3) Heterogeneity of attributes under investigation
4) Frequency with which phenomenon of interest occurs (rare or common?)

Question 3

What is eligibility criteria?

Answer 3

Criteria the researchers determine necessary to participate in the study in order to give the researchers greater confidence in saying “this definitely caused this effect”

Question 4

What are the types of sampling?

Answer 4

1) Probability (random): every person in population has an equal opportunity to participate; random sample; simple; stratified; cluster; systematic
2) Non-probability: not random; convenience; quota; purposive; network sampling

Question 5

Describe the types of probability sampling:

Answer 5

1) SIMPLE RANDOM: population defined, sampling frame listed, and a subset from which the sample will be chosen is selected. Better representation of population being studied.
2) STRATIFIED: population is divided into subgroups; an appropriate number of elements from each subgroup are randomly selected based on their proportion in the population (e.g. ethnicity)
3) CLUSTER: a successive random sampling of units; units progress from large to small (multi-stage)
4) SYSTEMATIC: probability sampling strategy that involves selection of subjects randomly drawn from a population list at fixed intervals; comes with the most bias and most problems.

Question 6

Describe the types of non-probability sampling:

Answer 6

1) CONVENIENCE: uses the most readily accessible persons or objects as subjects of a study; who are the easiest to bring into the study?
2) QUOTA: identifies the strata (levels) of the population and proportionally represents the strata in the sample
3) PURPOSIVE: researcher selects subjects who are considered to be typical of the population (good for qualitative studies)
4) NETWORK (aka. snow ball): strategy for samples difficult to locate; uses social networks and facts that friends tend to have characteristics in common; subjects who meet eligibility criteria are asked for assistance in getting in touch with others who meet the same criteria

Question 7

What is theoretical sampling?

Answer 7

• Sample identified, data collected, but realize that there are gaps; have to fill the gaps and compare with what they already have
• Found in qualitative grounded theory studies (key to develop theories, requires all the gaps are filled)
• Purpose to discover categories and offer interrelationships that occur in substantive theory
• Used to strengthen and understand the evolving theory

Question 8

If there is increased homogeneity in a sample being studied, we need a ____ sample size

Answer 8

Small, because there likely will not be much fluctuation in results d/t homogeneity

Question 9

If there is increased attrition in a study, we need a _____ sample size

Answer 9

Increase, as once a participant drops-out of a study their data is unusable

Question 10

What is a power analysis?

Answer 10

• Power = capacity to detect differences or relationships that exist in a population
• Acceptable level = .80
• an important aspect of experimental design; it allows us to determine the sample size required to detect an effect of a given size with a given degree of confidence.
• 20% chance of making a type 1 error (saying there is a difference when there isn’t)

Question 11

What is a type 1 error?

Answer 11

• Says there is a difference when there isn’t
• Difference d/t chance and need to consider reliability and validity of instruments
• Rejection of a null hypothesis that is true

Question 12

What is a type 2 error?

Answer 12

• Says there is not a difference when there is one
• Caused by small sample size, acceptance of a null hypothesis that is actually false (e.g. we think an intervention isn’t effective, but it actually is and can only be seen with larger samples)

Question 13

If there is increased variables being studied, there is a _______ sample size

Answer 13

Increased to ensure all variables will be studied

Question 14

If there is precise measurement sensitivity, there is a _______ sample size

Answer 14

Small, as likely to have highly accurate results

Question 15

How do we critique a sample?

Answer 15

• Is target population to which findings will be generalized defined?
• Is the sample representative?
• Are inclusion and exclusion criteria clearly presented?
• Is sample size sufficient?
• Use quantitative study results with caution when sample of small and sampling is non-probable

Question 16

How do we critique a quantitative sample?

Answer 16

• What type of sample?
• Is it representative?
• Are inclusion and exclusion clearly presented?
• Is sample size sufficient?
• What population can findings be generalized to?
• What are limitations of generalization?

Question 17

How do we critique a qualitative sample?

Answer 17

• What type of sample?
• Is it appropriate to design?
• Is the sample size appropriate?
• How is it substantiated?
• What are limitations to generalizabilty?