Researchers and scientists conducting surveys and performing experiments must adhere to certain procedural guidelines and rules in order to insure accuracy by avoiding sampling errors such as large variability, bias or undercoverage. Sampling errors can significantly affect the precision and interpretation of the results, which can in turn lead to high costs for businesses or government agencies, or harm to populations of people or living organisms being studied.
TL;DR (Too Long; Didn't Read)
To conduct a survey properly, you need to determine your sample group. This sample group should include individuals who are relevant to the survey's topic. You want to survey as large a sample size as possible; smaller sample sizes get decreasingly representative of the entire population.
A small sample size can also lead to cases of bias, such as non-response, which occurs when some subjects do not have the opportunity to participate in the survey. Alternatively, voluntary response bias occurs when only a small number of non-representative subjects have the opportunity to participate in the survey, usually because they are the only ones who know about it.
In the case of researchers conducting surveys, for example, sample size is essential. To conduct a survey properly, you need to determine your sample group. This sample group should include individuals who are relevant to the survey's topic.
For instance, if you are conducting a survey on whether a certain kitchen cleaner is preferred over another brand, then you should survey a large number of people who use kitchen cleaners. The only way to achieve 100 percent accurate results is to survey every single person who uses kitchen cleaners; however, as this is not feasible, you will need to survey as large a sample group as possible.
Disadvantage 1: Variability
Variability is determined by the standard deviation of the population; the standard deviation of a sample is how the far the true results of the survey might be from the results of the sample that you collected. You want to survey as large a sample size as possible; the larger the standard deviation, the less accurate your results might be, since smaller sample sizes get decreasingly representative of the entire population.
Disadvantage 2: Uncoverage Bias
A small sample size also affects the reliability of a survey's results because it leads to a higher variability, which may lead to bias. The most common case of bias is a result of non-response. Non-response occurs when some subjects do not have the opportunity to participate in the survey. For example, if you call 100 people between 2 and 5 p.m. and ask whether they feel that they have enough free time in their daily schedule, most of the respondents might say "yes." This sample - and the results - are biased, as most workers are at their jobs during these hours.
People who are at work and unable to answer the phone may have a different answer to the survey than people who are able to answer the phone in the afternoon. These people will not be included in the survey, and the survey's accuracy will suffer from non-response. Not only does your survey suffer due to timing, but the number of subjects does not help make up for this deficiency.
Disadvantage 3: Voluntary Response Bias
Voluntary response bias is another disadvantage that comes with a small sample size. If you post a survey on your kitchen cleaner website, then only a small number of people have access to or knowledge about your survey, and it is likely that those who do participate will do so because they feel strongly about the topic. Therefore, the results of the survey will be skewed to reflect the opinions of those who visit the website. If an individual is on a company's website, then it is likely that he supports the company; he may, for example, be looking for coupons or promotions from that manufacturer. A survey posted only on its website limits the number of people who will participate to those who already had an interest in their products, which causes a voluntary response bias.