Sampling Flashcards
What is a population or universe?
Any complete group with common characteristics
What is a population element?
Single member of a population
What is a census?
Investigation of all individual elements that make up a population
What is a sample?
- Subset of the larger population of interest
- This is the subset of group that the researcher will actually study or investigate
What is a population frame?
A list of all the elements in the population
What is a sample frame?
A list of all the elements in the population from which the sample may be drawn
What is sampling frame error?
An error which occurs when certain sample elements are not listed or are not accurately represented in a sampling frame
What is a sampling unit?
An element or group of elements subject to selection in the sample
What is inclusion/exclusion criteria?
The criteria potential participants must meet in order to be included in the study
What are the core statistical concepts?
- Descriptive statistics
- Inferential statistics
- Population parameters
- Sample statistics
What are descriptive statistics?
- Describe the data
- Measure of central tendency, frequencies, dispersion
- Trends
What are inferential statistics?
- Project characteristics of a sample to an entire population
- Make an inference about a population from a sample
- Used in hypothesis testing
What are population parameters?
- Characteristics of the population
- Variable in a population or measure characteristics of the populaiton
- Greek letters as notation
- μ = ∑X/N
What are sample statistics?
- Estimates of population parameters
- Variables in a sample or measures computed from sample data
- English letters for notation
- X̅ = ∑X/N
What is the cost associated with sampling and the solution to this?
- A loss of information
- To make up for this loss we have to ensure that the sample is representative of the population
What does representativeness determine?
The representativeness of the sample determines the extent to which generalisable inferences can be made
What does the CLT imply?
- Suggests that the sampling distribution of the sample mean produces a normal curve
- As the sample size increases, the means of random samples taken form the population approach a normal distribution
- This means we have a representative sample - and our sample mean will be within close range of the true population mean
What is probability sampling?
Chance of selecting any particular member is known and is equal for all units (probability is non-zero)
What methods of probability sampling are there?
- Simple random
- Systematic
- Stratified
- Cluster
What is simple random sampling?
A sampling procedure that ensures that each elements in the population will have an equal chance of being included in the sample
What is systematic sampling?
Every nth name from a list (sampling frame) will be drawn
What is stratified sampling?
- Subsamples are drawn within different strata
- Strata are subgroups of elements that may be expected to have different parameters on a variable of interest
- Each stratum is more or less equal on some characteristic
What different types of stratified sampling are there?
- Subjects drawn from each stratum can be either dispropriationate or proportionate to the number of elements in the stratum
- Proportionate stratified sampling (20% of members from each stratum)
- Disproportionate stratified sampling (% of members disproportionate across stratum)
What is cluster sampling and what is the important aspect of it?
- Purpose is to sample economically while retaining the characteristics of a probability sample
- Primary sampling unit
- No longer the individual elements in the population
- Instead, a larger cluster of elements located in proximity to on another
What is non-probability sampling and when is it useful?
- Chance of selecting any particular member is unknown
- Useful when sampling frame cannot be created
What kinds of non-probabilty sampling are there?
- Convenience methods
* Purposive Methods
What types of convenience methods of sampling are there?
- Convenience
* Snowball
What is convenience sampling?
- The sampling procedure of obtaining the people or units that are most consistently available
- Least reliable of all sampling designs in terms of generalisability, but may be the only viable alternative when quick and timely information is needed
What is snowball sampling?
- Initial respondents are selected by probability methods
- Additional respondents are obtained form information provided by the initial respondents
What types of purposive sampling methods are there?
- Judgement
* Quota
What is judgement sampling?
- An experienced individual selects the sample based on his or her judgement about some appropriate characteristics required of the sample member
- May curtail generaliablity, but it is the only viable sampling method for obtaining the type of information from specific sub-groups
What is quota sampling?
- Ensures that the various subgroups in a population are represented on the pertinent sample characteristics
- Are basically stratified samples from which subjects are selected non-randomly and on the basis of convenience
- Needed to adequately represent minority groups
What are confidence and precision in regard to sampling errors?
- Confidence and precision refer to the extent and degree to which legitimate inferences can be made about a target population from a sample
- Confidence and precision will be influenced by sampling errors
What do sampling errors do?
- Sampling errors limit the extent of generalisability because they diminish representativeness
- Sampling errors can cause differences in the sample statistic and population parameters
What types of sampling errors are there?
- Random sampling error
* Non-random or systematic sampling error
What are random sampling errors?
- Difference between the sample result and the result of a census conducted using identical procedures
- Statistical fluctuation due to chance variations
- Large number of untypical subjects
- Outliers and extreme values
- Inversely related to sample size, thus only way to manage is through large sample
In sampling, what are non-random or systematic sampling errors?
- Sampling error not due to chance i.e. biased selection
- Non-response error
- Pattern of responses are atypical of target population: self-selection bias; social desirability
- Study design errors or imperfections in execution
- Managed through sampling design
In sampling, what kinds of non-random or systematic sampling errors are there?
- Self-selection bias
* Social desirability bias
What is self-selection bias?
- The bias that creeps into results when the participants of a study are people who choose to participate
- The key component is that research subjects (or organisations) volunteer to take part in the research on their own accord. They are non approached by the researched directly
- It’s a bias because a group of people who choose to participate is not the same as a random sample of the population
- They may also differ in important ways from those who do not choose to partipcate
- They motivations and decision to participate in the study may reflect some inherent bias
- This can either lead to the sample not being representative of the population being studies, or exaggerating some particular finding from the study
What is social desirability bias?
- The tendency of subjects to attribute to themselves statements which are desirable and reject those which are undesirable
- In other words, some people like to portray themselves positively - tending to exaggerate or inflate their strengths and achievements, and often deny or trivialise their deficiencies and failures
- Particularly problematic with self-reports of performance
- To manage this it is important to collect data from multiple sources of use a scale to measure social desirability
