The researcher needs to decide upon the people to be studied, this is the sample. It is usually chosen in order to be representative of the particular population to be studied e.g. women over 50. Sampling can save time and money, because the larger population cannot be studied. There are different methods of producing a sample.
Random sampling is the simplest. Everyone in the population has an equal chance of being in that sample, and it gets rid of bias sampling. Random sampling involves selecting individuals randomly e.g. selecting every 10th person on a list.
Stratified random sampling divides the population into strata. These strata are based on significant variables e.g. social class, gender, and ethnicity. Samples are then randomly drawn from each strata, and then combined to form the final sample.
In Quota sampling, the researcher sets the quota outlining the number of people required, who meet certain criteria e.g. 10 men under 40. This is a quick and easy method because even if people refuse you simply find another. The main disadvantage is that it is not truly random or representative.
Multi-stage sampling involves taking a selection of a sample, which goes through several stages. Each stage involves the selection of a sample from a previous sample, until you finally arrive at the people you will be using. Multi-stage sampling is usually used for discovering how prospective voters are likely to cast their votes leading up to an election. The advantage of this method is that it is cheaper and less time consuming, but it is not truly random.
Snowballing is a method only used when other methods aren’t practical. In snowballing the researcher starts with a few people, and asks them to recommend more people to study, and eventually that leads to a big sample. This is participant observation. This method is used for particularly sensitive issues e.g. domestic violence or criminal behaviour.
Observation is carried out at all times with both quantitative and qualitative data. Overt observation can lead to taking part in illegal activities, where as when doing covert observation, people are unaware they’re being observed so they act naturally. This method would suit certain types of research e.g. criminal behaviour. The disadvantage of covert observation is that it may be difficult to record data. There also may be a danger of becoming too involved.
In Participant observation validity is greater then interviews and more subjective. The researcher can also see people in different contexts and can watch the change in behaviour, you wouldn’t in interviews. Participant observation can be time consuming and can also take years. Only small groups can be studied making generalisations difficult to assume.
Questionnaires can be very useful in carrying out research, because questionnaires use quantitative data, which is mainly about numbers and statistics. The advantages of quantitative data are, with facts, figures and numbers, you can see trends and patterns. This kind of data can be analysed using computers. You could also use large samples and surveys. Quantitative data is also useful in carrying out research, because it can be reanalysed, so sociologists can use it again, therefore it can be very reliable with this method of research.
After looking at the different factors that influence the sociologist’s choice of research methods, I found that the methods are very different to one another. The correct method should be used with the appropriate choice of research. Ethnical issues should also be considered when do research. Ethnical issues are important when carrying out research, because people give you information and this research can have an effect on people’s lives. This is why ethnical issues are important when carrying out research. Respondents must also be protected from harm, which may arise from the research.