Sampling Techniques

·        This usually involves the sociologist selecting the research subjects randomly from a list of names or addresses called a ‘sampling frame’.

·        The sampling frame is a list of names of the people who make up the research population that the sociologist wishes to research.

·        It is usually divided into ‘sampling units’.

·        Random sampling means every member of research population has an equal chance to be chosen.

·        However, it is possible it can end up selecting a disproportionate number of people from one particular group, creating ‘bias’.

·        Sociologists have developed two variations of the random sample in order to produce representative results.

·       Involves randomly choosing a number between one and ten, say seven, then picking out every tenth number e.g. 7, 17, 27, 37, 47, 57.

·       It does not always guarantee a representative sample

      However, because it uses a larger sample it is more likely to be representative and less likely to be bias of one group.

·       Involves dividing the research population into a number of different sampling frames and then using systematic random sampling to select the group that will be the research sample.

       E.g. if a town has 10,000 people who are exactly half male and half female, if you were to take a sample of say 1000, it is unlikely you would get exactly 500 men and 500 women.

·       So instead, the sociologist will put men and women into separate sampling frames and then systematic random sampling will be used.

·       Then a fair number of men and women will be chosen.

·       You can even break this down further and add in other sampling frames if you wished, such as age and ethnicity.

The research decides how many of each category of person should be included in a sample and then goes out looking for these people until the sample is done.

E.g. if a sample if 500 and the quote of women aged between 30-40 is 22, the researcher will look out for 22 such women and when they have found that the quota is filled.

This is used by people such as market researchers.

However, it lacks randomness, there is a danger that bias might creep into the selection of the sample as the researcher may only look for people who ‘suit’ their research.

Choosing individuals or cases that fit the nature of the research.

Goldthorpe (1969) wanted to study manual workers with high incomes to see whether they developed a middle-class way of life.

His team purposely studied workers employed by a car manufacturer in Luton because these manual workers were known to be well paid.

Making the most of situations or opportunities in which the research population is likely to be found.

E.g. if you wanted to research African people in the UK experiencing racism, you would have to identify opportunities in which there is more chance of coming into contact with African people.

Researchers may head down to community centres or churches that cater for African people and ask them to take part in research.

This involves finding and interviewing a person that fits the research needs and then asking them to suggest someone else who might be willing to take part.

This way the small sample grows by adding people to it in stages, much like how a snowball can be built up by rolling it along the snow on the ground.

The sample can grow as large as the researcher wants.

This ishandy when trying to study deviant or illegal activity.

Variation of snowball.

Sociologists may advertise for volunteers in magazines and newspapers or internet.

However, both snowball and volunteer sampling may fail to produce a representative sample.

The people who take part in research may not be typical of the research population that the sociologist is interested in.