Nonprobability sampling methods Report person 164 S 17112
Non-probability sampling methods Report person: 164 S 17112 Mick 164 S 17113 Karl 164 S 17114 Smart 164 S 17119 William 164 S 17122 Chris
Content 1 Convenience sampling 2 Quota sampling 3 Purposive sampling 4 Snowball sampling
Non-probability sampling methods: • It's not a random selection • The distribution of sample statistics is not exact • Simple operation, fast aging, low cost
01 Convenience sampling
Convenience sampling means that a sample is chosen according to the Convenience of an investigator and conducted in a purposeless and random manner. Advantages: easy to implement, low cost of investigation. Disadvantages: Investigators tend to choose the nearest or most accessible person as respondents, which will affect the accuracy of sampling. The sample cannot represent a clearly defined population and the population cannot be inferred from the survey results.
02 Quota sampling
Quota sampling u. Quota sampling, refers to the sampling method in which the investigator classifies or stratifies the overall sample according to a certain mark, determines the sample amount of various (layer) units, and randomly selects samples within the quota. u. There are similarities between quota sampling and stratified random sampling. All units in the population are classified in advance according to their attributes and characteristics, which we call "control characteristics" For example, the gender, age, income, occupation and education level of consumers in the market survey. Then, the sample amount is allocated according to each control characteristic.
Quota sampling It is applicable to the case that the design investigator has a certain understanding of the relevant characteristics of the whole, but the sample number is large, the cost is not high so it is easy to implement, and it can meet the requirements of the overall proportion. It is easy to cover up the deviation that cannot be ignored, and it is very difficult to calculate. Sometimes it can also be used effectively for field research. For example, in a study of formal groups, you should interview leaders and members of the group. If you study a student organization, you should visit both the radicals and the gentler members of the organization.
03 Purposive sampling
Purposive sampling u means that investigators don't have to fill a quota out of a variety of people, as they did in quota sampling, but do n' t pick a nearby zealous group as they did in convenience sampling. Instead, researchers choose only those who are best suited for the purpose of the study according to their own judgment of the respondent to be chosen.
Purposive sampling Advantages (1) The method of purposive sampling is simple and easy to carry out, meets the purpose of investigation and special needs, and can make full use of the known data of the survey. (2) Researchers can use their research skills and knowledge to select respondents. For example, he can look for "average American housewife, or all. American boy ". One common technique for sampling general election forecasts by intent is to find the average or lowest common denominator constituency, and the other is to look for unusual cases instead of the average respondent, in order to examine carefully what makes them deviate.
Purposive sampling Disadvantages (1) The result of purposive sampling is greatly influenced by the tendency of researchers, and it is easy to cause sampling deviation once subjective judgment deviation; it can not be inferred directly from the survey population. To give full play to the positive role of purposive sampling, the basic characteristics of the population must be clear, so that the selected sample can be representative, typical, in order to understand grasp the overall situation. (2) Purposive sampling is suitable for special types of research (e. g. product taste testing) when the composition of the population is very different and the number of samples is very small, and the design investigators have a good understanding of the relevant characteristics of the population.
04 Snowball sampling
Snowball Sampling Snowball sampling is a sampling method that expands gradually from less to more by means of the naturally formed interpersonal network like snowball rolling. Snowball sampling is often used to investigate the phenomenon of public opinion in a special group of people. Snowball sampling is a sampling method between random sampling and non-random sampling. Its normal state is non-random. If snowball sampling is to become random sampling, it must be randomly selected from all sampling units in each stage. Snowball sampling was carried out in stages. The first step is to interview several people with the characteristics of the subject from people known to the researcher, and use these people as contributors, relying on them to identify other qualified people. The second step is to interview these people and have them lead you to the more people you want to interview in the third step. And so on, gradually expanding the number of visitors until the required sample size is reached. Snowball sampling is used especially for observational studies and is particularly suitable for community studies. It is also suitable for the study of special types of people (such as disabled people, drug addicts and bridge enthusiasts). However, the sample representativeness of snowball sampling method is relatively low. Smart Gao 高浩然 164 S 17114
THANKS
- Slides: 15