Sampling method can be divided into two types: Non probability sampling and probability sampling. Non probability sampling does not follow the probability principal. It is based on human’s subjective experience or other conditions and is commonly used for explorative research. Probability sampling involves the selection of a sample from a population, based on the principle of randomization or chance.
Starmass International will design the sampling methods scientifically and practicably in accordance with our clients requirements in market entry, objectives of the research and characteristics of the selected market object.
Non-probability Sampling
1.Haphazard sampling
Haphazard sampling is sometimes referred to as convenience or accidental sampling. A haphazard sample is used by simply stopping anybody in the street who is willing to answer the questions. In other words, the sample comprises subjects who are simply available conveniently to the researcher.
The obvious advantage is that the method is that it is low in cost, saves time and expenses. However, this advantage is greatly offset by the presence of bias. Although useful applications of the technique are limited, it can deliver accurate results when the population is homogeneous.
2. Judgement sampling
Selection of sample is based on certain judgement about the overall population. The underlying assumption is that the investigator will select the subjects according to the characteristic of the population. The accuracy of judgement sampling will be affected by researcher’s biases which may be more than haphazard sampling. Biasness can be introduced if the researchers have their own preconception regarding the research as they are reflected in the sample. If these preconceptions are inaccurate, large biasness would occur.
3. Quota sampling
Sample units in the population do not have a known chance of selection in quota sampling. Interviewers are required to find cases with similar characteristics. A quota of particular types of people will need to be interviewed. The information gathered will be organized and the final sample would be representative of population.
4. Snowball sampling
Snowball sampling is a type of non-probability sampling where initial respondents are selected at random and subsequent respondents are then selected by referrals or information from the earlier respondents. The selection of one element leads to the identification and selection of others and these in turn to others, and so on like a rolling snowball which would increase in size.
Probability Sampling
1. Simple random sampling
In simple random sampling, each member of a population has an equal chance of being included in the sample. Also, each combination of members of the population has an equal chance of composing the sample. This is the easiest method of sampling and it is most commonly used. Advantages include no requirements of any additional information on the frame other than the complete list of members of the survey population along with information for contact.
2. Stratified Sampling
Using stratified sampling, the population is divided into homogeneous, mutually exclusive groups called strata, and independent samples are then selected from each stratum. Stratified sampling ensures an adequate sample size for sub-groups in the population of interest. This method is a frequently used method that is superior to random sampling because it reduces sampling error.
3. Cluster sampling
By dividing the population into clusters, a number of them will be randomly selected to represent the total population except those in the non-selected clusters. This method is low in cost and it is convenient as all units in the population are not always available but a list of all clusters saves time.
4. Interval sampling
Interval sampling is sometimes referred to as systematic sampling, it means that there is a gap, or interval, between each selected unit in the sample, for example, measuring depth in a stream every 50 meters or interviewing every tenth person as part of a survey. The advantages of interval sampling in market research are that the sample selection cannot be easier and that the sample is distributed evenly over the listed population.
5. Multi-stage sampling
Multi-stage sampling, as the name implies, involves drawing from several different samples. First, large groups (includes more units that required) are selected. The process of selecting population units within the groups continues until there is a final sample.
Starmass International processes comprehensive database as sampling frames, which include:
- sampling frame of China major cities
- database of China main industrial players
- sampling frame of China main industrial products
- sampling frame of China media database
- latest database of China sixth population census