Sampling

Sampling

1. To test a hypothesis socioogists need to think about the reasearch population they are studying and how it should be sampled

2. Reasearch Population- The group that will be researched

3. The Sample- reasearchers select a sample that is represntative of the reasearch population

   • With a representative sample it is possible to generalize to the wider research population
   • What is true of the sample should be true of the research population as a whole

4. Two main sampling techniques can be used to ensure the sample is representative of the wider research population: 1) Random Sampling 2) Non-random Sampling

5. Random Sampling

   • Involves selecting names randomly from a list or sampling frame
   • Every member of the research population has an equal chance to be included in the sample
   • Various types of sampling frames can be used: -The electoral register -The postcode address file -The telephone directory -school attendance registers -Patient records of general practitioners
   • All sampling frames are unsatisfactory in some respect: -Out of date -Some groups may be over represented -Some groups may not be included

6. Different types of random sampling techniques

   • A simple random sample does not guarantee a representative sample
   • Therefore to produce a representative sample, sociologists have developed three variations on the random sample
   • 1) Systematic Sampling: Involves randomly choosing a number between one and 10, say "seven", and then picking out every 10th number from that number from that number- 7,17, 27, 37... on the list until the sample is complete. This does not always guarantee a representative sample but the larger the sample, the more likely it is to be reasonably and the less likely is to be biased
   • 2) Stratified Sampling: is the most common form of sampling used in sociological research. It involves dividing the research population into a number of sampling frames, for example, by gender, age, ethnicity and social class, which represent their proportion in the research population, and then randomly selecting, say every 10th name until the required number is reached for each sub sample.
   • 3) Cluster Sampling: is often used when no specific list of people is available. The researcher uses a map to randomly select a couple of areas, and then streets within those areas. The researcher then randomly targets a further sample of people or households within the streets.

7. Non-random sampling methods

   • 1) Quota sampling: is often used by market research companies to target people in the street to talk about consumer products. The researcher is told by his/her company how many participants are needed in each category and goes looking for them, usually in a city center. This sort of sampling technique is often used by television news companies and newspapers to find out people's voting preferences before an election
   • 2) Purposive Sampling: involves researchers choosing individuals or cases that the fit nature of the research. For example, a researcher might visit a church because he or she needs to interview religious people
   • 3) Snowball Sampling: is used mainly when it is difficult to gain access to a particular group of people because there is no sampling frame available or because they engage in deviant or illegal activities, which are normally done in isolation or secret. This sampling technique involves finding and interviewing a person who fits the research criteria and then asking her or him to suggest someone else who might be willing to be interviewed. The sample can grow as large as the researcher wants

8. All sampling is a compromise between representativeness and practicality, and researchers often have to make do with samples that are not fully representative