Learn something new every day
More Info... by email
There are a range of different sampling methods used when selecting a testing panel for research. This research can involve testing either a theory or a specific product, carrying out an opinion poll, or any other research which aims to cover a particular group in its entirety. This group is known as the population, though it can involve any type of group, not just the citizens of one country.
With a small population, such as staff who work in a particular office, it's usually possible to question or test everyone involved. This is known as a census study. With most populations, such as "everyone in China aged 65 or over," it's impossible to question or test everyone, so a sample group must be selected. The different ways of choosing these participants are known as sampling methods.
Sampling methods fit into one of two main categories: probability and non-probability. In a probability sampling method, everyone has a known likelihood of being selected, though this likelihood can vary from person to person. In a non-probability sampling method, some people have no chance of being selected as the participants are chosen from specified sections of the population. This can be more convenient, but comes at a price: unlike probability sampling, non-probability sampling makes it impossible to estimate how accurately the sample group represents the entire population.
The simplest form of probability sampling is to randomly select people from a list of the entire population. A variation of this method, systematic sampling, involves picking out people at fixed intervals along the list, for example every hundredth person. Both of these sampling methods have flawed as the resulting sample group may not represent the make-up of the population. For example, the sample group may have three children and seven adults, which is clearly not representative if the entire population is 20% children and 80% adults.
This can be solved by using stratified sampling, in which the population is broken down into particular groups sharing common factors and participants are selected randomly from these groups in the appropriate proportions. In the example above, the researchers would randomly select two people from a list of all children and eight people from a list of all adults. Naturally this can be extended to cover other types of group, such as by gender, to make a sample group which more accurately reflects the entire population.
The simplest forms of non-probability sampling is known as convenience sampling. The researchers simply choose the participants who are easiest to get hold of. Clearly there is a strong risk of this being very unrepresentative of the population. For example if researchers knock on doors during the day they will be less likely to get participants who are in full-time employment.
Quota sampling combines stratified sampling and convenience sampling and usually involves researchers setting out to find participants to fill quotas. In the example above, the researchers might knock on doors until they had spoken to a total of two children and eight adults. Although this method means the sample group is in the right proportions, the selection process makes it impossible to know how representative it is. In our example, the eight adults might all be unemployed, which would make them unrepresentative of the opinions of the whole population in a question about social security benefits. Because of this, quota sampling is classed as a type of nonprobability sampling.
@Bhutan - I just fill out the form and be done with it. I think that the best sampling methods in research involve offering a survey along with follow up questions like many market research companies do.
When they call you they often verify if the answers that you initially gave were correct and if the answers do match then they offer you an opportunity to join a focus group.
They are very selective because you really have to match the customer’s demographics in order to give valuable feedback. I was involved in a focus group and it was really a lot of fun.
I was asked questions about my experiences with a particular retailer and its competitors. It was about two hours and I think I was paid like $150.
I did notice a substantial difference in the customer service level in this particular retailer so maybe they really listened to our suggestions.
@Cafe41 - I remember that. I also saw a story on television that talked about the sampling problems with some of the major pollsters.
I think that the Census sampling method is the most accurate because you have to fill out your demographic information and if you don’t they actually come to your home and ask you the questions in person.
This is what happened to me and they asked me detailed questions.
I think that random sampling methods with respect to political polling can sometimes be misleading because people sometimes the methods of sampling are not even.
For example, some polls will display a liberal point of view as being more prominent when in reality there was an oversampling of Democrat voters which skewed the results to favor the Democrat point of view. This is why polling data has to be taken with a grain of salt cause sometimes the research methods were not as accurate as they would have you believe.
I also think that sometimes there might be leading questions that would cause a respondent to state a certain opinion but if the question were worded differently
the respondent would have chosen a totally different response.
I think that data integrity is really important and random sampling methods such as exit polling might not reveal the information that the pollster wants because people can lie.
They can also oversample people from one party over another. This is what happened in the Presidential election between President George W. Bush and John Kerry.
The exit polling data suggested that John Kerry would win by a margin of five points, but it was actually President George W. Bush that won by that same margin.
One of our editors will review your suggestion and make changes if warranted. Note that depending on the number of suggestions we receive, this can take anywhere from a few hours to a few days. Thank you for helping to improve wiseGEEK!