random sampling techniques

simple random sampling

• where every member of the population has an equal chance of being selected for the sample

• using a sample frame, the sample can be drawn randomly by , for eg taking names out of a hat

• strengths : there’s no researcher influence and no bias in selection

• weakness: there is a chance that the sample obtained may not be truly representative for eg the sample could be all male or all female

systematic random sampling

• involves randomly choosing a number between one and ten , say seven, and then picking out every tenth number from that number - that is, 7, 27, 37,47 and so on - from the sampling frame until the required number in the sample is reached

• the technique doesn’t guarantee a representative sample

• however, the larger sample , the more likely it is to be reasonably representative and the less likely it is to be reasonably representative and the less likely it is to be biased in favour of any age group

stratified random sampling

• separate sampling frames are constructed for men and women so that 500 people can be sampled from each group using a systematic random sampling technique

• if the researcher wanted to stratify further because the hypothesis included reference to ethnicity and age, more sampling frames could be complied so tat the right proportions of each group could be systematically and randomly sampled

samples and representativeness

• however, when judging a sample these are not the only way you judge them

• a sample may be chosen using a technique based on

• bias : if you have a random technique then all have an equal chance if being picked and the sample will be picked in an unbiased manner

• access :some groups are harder to access than others so the technique used to pick them may be more about that issue than whether the sample is representative

• theoretical constraints : positivists who wise to measure , look for patterns and trends and look for social facts , are most likely to want a large representative sample

• interpretivists who wish to explain , understand and empathise with their sample , will be less worried about representativeness and more worried about gaining precise type of sample

selecting a sample

• a research population refers to all of those who could be included in the survey

• there’s thousands and millions of people in the research population meaning that they cannot all be included in the survey

• the researcher cannot deliver it to them all, still less interview them face to face

• therefore a sample must be chosen

• the main principle of sampling is to choose a small cross-section of the research population , because it is quicker and cheaper , but the sample must be representative of the population as a whole

• many sampling techniques requires a sampling frame which is a list of all of the members in the research population

• common sampling frames include the electoral register or the post-code address file.