SOCI 217 Lecture Notes - Lecture 11: Stratified Sampling, Sampling Bias, Simple Random Sample
jbWeek 5- sampling
Why sample
• We sample all the time
• Need a process to select participants/observations
• Studying a whole population may be cumbersome/impossible
Probability sampling
• Samples selected in accord with probability theory
o Everyone has a known non-zero chance of being selected
o Random selection mechanism
• Typically involves some random selection mechanism
o Representative of larger population
o Equal chance
• Most common approach in quantitative research
Non-probability sampling
• Any technique for selecting samples that does not use probability theory
• May be unrepresentative
o Not equal chance of selection
• Most common approach in qualitative studies
Errors
• Sampling error
o Does’t atch ider populatio
• To reduce chance of sampling error we increased the sample size
• Another way is a homogeneous population
• Coverage error
o Target population not having non-zero probability of being included in sample
o Saple does’t coer the full populatio of iterest
o Go the extra mile to find hard to reach respondents
•
Methods of probability sampling
Basic probability sampling jargon
• Representativeness
find more resources at oneclass.com
find more resources at oneclass.com
Document Summary
Why sample: we sample all the time, need a process to select participants/observations, studying a whole population may be cumbersome/impossible. Non-probability sampling: any technique for selecting samples that does not use probability theory, may be unrepresentative, not equal chance of selection, most common approach in qualitative studies. In some cases it may be unrepresentative: sampling unit, who is considered, study population, all the elements put together, ex. All students of 217: population is called parameter, sample is called a statistic. Simple random sampling (srs: most basic method of random sampling, research participants selected at random from a sampling frame, computer assisted, table of random numbers, 1. take class list and number each of the students, 2. Determine number of digits you need: 158 -> is 3 digits, 3. Look at the table of random numbers and make a plan: 4.
