Methods of sampling Flashcards
What is a population?
Large group
Area to be studied
Whole
This must be clearly defined
What is a sample?
Subset of a population
Must be representative of the population of interest
Need to minimise sampling error and variability
What are the problems of the Hackney sample example and how could this be improved?
One area only, urban, different foods, immigrants , lower SES
List of all the primary schools and randomly sample from that, sample from GP practices, could sample from the census but quickly out of date
What is the theory behind sampling?
Statistical estimation - point or estimate
Testing of a hypothesis - accept or reject the null
Statistical inferences - General statement about the population
What are the features of sampling method?
Economy
Reliability
Detailed study
Scientific base
Greater suitability in most situations
What are the limitations of sampling?
Less accuracy
Changeability of units
Misleading conclusions
Need for specialised knowledge
When sampling is not possible
What are the characteristics of an ideal sample?
Representativeness
Independence
Adequacy
Homogeneity
What are methods of sampling?
Probability sampling methods (random) - Simple random sampling, stratified, systematic, multistage
Non- Probability sampling methods (non random) - Judgement or deliberate sampling, convenience, snowball, quota
What are the advantages of probability sampling methods?
They require detailed information about the population to be effective
Provides estimates which can be measured precisely and are unbiased
Possible to evaluate the relative effectiveness of various sample designs
What are the disadvantages of probability sampling methods?
It requires a high degree of skill and expertise
Time to plan and determine a probability sample
High cost
What is simple random sampling?
Individual units constituting the sample are selected at random
Guarantees each element in the population has an equal chance of being chosen
Use a random number generator to generate numbers or SPSS can generate random samples
What are the advantages and disadvantages SRS?
Adv
Simple, free from bias and prejudices, more representative as each unit has an equal chance of being selected, errors are easy to detect
Disadv
Selection according to strictly random basis is not possible , lack of control of research
What is stratified sampling?
Each stratum in the population should be large enough so that selection of items can be random
Perfect homogeneity among units
Ratio of number of items to be selected from each unit should be the same as the total number of units in the strata
Should be well defined and clear cut
What are the types of stratified sampling?
Proportionate stratified sampling
Disproportionate stratified sampling
Stratified weighted sampling
What are the strengths and weaknesses of stratified samples?
Strengths - greater control of the investigator, easy to achieve representative character, replacement of units is possible
Weaknesses - Possibility of bias, difficult to achieve proportion, difficulty in making the sample representative, difficulty in placing cases under level
What are the strengths and weaknesses of systematic sampling?
Strengths - Simplicity in drawing a sample, variances are smaller, with the ordered population it reduces the variability compared to SRS
Weaknesses - Interval related to ordering of the population may increase variability, estimates of error likely to be high where there is stratification effect
What are the strengths and weaknesses of multistage sampling?
Strengths - A complete list of the population is not required , lists ID and numbering are required only for sampling units selected in the sample, if units are geographically defined then costs are reduced
Weaknesses - Errors are likely to be larger than in SRS or systematic, errors increase as number of selected units decrease
What are the strengths and weaknesses of judgement sampling?
Strengths - A small number of sampling units in the population it enables inclusion of important units, a more representative when the study looks into unknown traits of the population, practical method
Weaknesses - Involves the risk by including items in the sample which conform to researcher’s preconceived ideas, no objective way of evaluating the reliability of sample results
What is convenience sampling?
Used when population is not well defined
Sampling unit is not clear
When complete list of the source is not available
What is snowball sampling?
Researcher contacts small number of people in the target group then use these people to establish new contacts
Problems - No accessible sampling frame , sample will be unrepresentative - difficult to generalise
What is quota sampling?
Classifies the pop into types relevant to the characteristics researched. Determine the proportion of the pop falling into each type based composition of the pop. Quotas set for each interviewer so the total sample interviewed contains the proportion of each level
What factors determine the reliability of a sample?
Size
Representativeness - must possess all characteristics of all units
Parallel sampling - another sample for testing
Homogeneity of the samples - all same features present in the population
Unbiased selection
What are the 3 types of errors?
Sampling variability - different samples from the same population do not always produce the same mean and SD
Sampling error - the mean of a sample wont be the same as the population mean, can be minimised but not eliminated using good selection criteria
Non sampling error - Errors not connected with sampling method e.g. leading questions
How large should a sample be?
Populations with greater variability need a larger sample size
Greater precision requires larger sample size but not proportional increases in precision with sample size
Needs to fit the budget and resources
