A population by definition is a group of people with one or more characteristics in common. A population of people can also be defined as people who share a certain interest (eg. Cricket fans), have a common feature (eg. Left-handed people). The sample of that population is what’s recruited in a research or experiment.
The demographic taken should represent the population it's taken from so the findings of the research are representative and later generalizable. Target population of the study should also be recognized early on so that the sample the researcher chooses should be relevant and representative.
Important things to consider when sampling:
• Sample details such as age, ethnicity, gender. They are basic essentials that should always be considered.
• Sample details such as socio-economic standing, employment, education, occupation, geographical location.
• Sample size. (Should be balanced in terms of being representative)
• Small samples usually are less reliable and less representative and thus generally less valid to the clause of research.
An opportunity sample involves the researcher approaching people who are easy to find and easily available, such as students who are studying mathematics in the same university department. If a researcher is however interested in the ‘general demographic/public’ then it is possible for the researcher to approach people in places such as parks, student common rooms, shopping malls and etc.
Volunteer sampling (self-selecting sampling):
This usually revolves around researcher and experimenters advertising for participants. An advert could usually appear in a newspaper or on notice boards, online too. The people who reply are ‘self-selecting’ - that is they willingly volunteer themselves for the research. Sometimes volunteers are not given incentives at all, neither credentialled nor paid, often they are given a small amount of both or one of the other.
In this type of sampling technique each participant is given the equal chance of being selected from the target population. If the target population is ‘factory workers’ and there are 800 of them – the only way to actually randomly select the sample is to put all 800 names together and pick out the first 20, 30 names (depending on the required sample size of the study).
- It is relatively quick and easy to recruit participants. A large, representative sample can be obtained without a lot of effort.
- It is convenient for the researcher and the participant as in some cases – there are also incentives.
- Participants in the study are unlikely to be actually representative of the target population in the sense that it could be biased (if they’re paid or given credits to), when the researcher chooses the sample.
- They are useful when the research requires participants that are specific to the needs of the experiment.
- Recruiting participants is easier because the advert can easily be placed in print media, social media and digital media which has a large enough reach.
- It is less time-consuming.
- This is expensive (adverts in media would cost a lot of capital investment) and in some cases it would take even more effort to convince the participant to be in the study.
- People may not see the advert, they might ignore it, they might not reply even after seeing it.
- Extraneous variables such as the actual eligibility criteria of the participants being different from what’s actually required.
- No way to assure the representativeness of the target population.
- Plausible demand characteristics and social desirability is risked to effect as the findings in the case where an incentive is involved when they volunteer.
- They are more likely to be representative than opportunity or self-selecting samples as the clause for bias does not exist as the selection is up to chance, random.
- It is efficient if a certain demographic is to be studied and out of that lot the participants are selected at random.
- It is however often time-consuming when a large target population is considered. If for example not all names of the potential participant pool, it would be difficult to conduct a random sample.
- The chance of equal opportunity in the random selection process if often too idealistic in the sense that not everybody might be inclined to fully participate. It is possible for them to leave and then for the researcher to recruit and replace a new participant.
- This might bias the sample.