Researchers are not interested in the sample itself, but in the understanding they can potentially infer from the sample and then apply to the entire population. 2.A good sample should be an accurate representation of the entire universe or population. It is important to understand why we are sampling the population. For example, studies are underway to investigate the relationships between risk factors and disease. In other words, we want to know if it is a true association, while aiming for minimal risk of errors such as coincidence, bias or confusion. (2) Accurate representation of the universe: A sample must be an accurate representative of the universe from which it was taken. There are several methods of selecting a sample. It will only be truly representative if it represents all types of units or groups of the total population in appropriate proportions. In short, the sample must be selected carefully, as an inadequate sample is a source of error in the survey. A good sample should be representative of the population we want to study, with each item having an equal chance of being randomly selected for the study. A good sample has the following characteristics: Disadvantages: When there are very large populations, it is often difficult to identify every member of the population, and the pool of subjects is distorted.
Dialing numbers from a telephone directory, for example, may not be entirely random because, although random, the numbers would correspond to a localized region. A sample created in this way may leave out many parts of the population that are relevant to the study. If each entity in the population has some non-zero probability of being included in the sample, the sample is called the probability sample. However, it would not be possible to experiment with the entire population, a good sample would have to be taken and the risk of errors would have to be reduced by an appropriate sampling technique. How: As the name suggests, items in such a sample are selected solely based on their convenience in terms of availability, reach, and accessibility. The samples do not need to be larger than what is representative of a material and are limited in size for our instrument by the physical dimensions of the table and sample holders. Samples cannot be thicker than about 13 mm (~1/2″), but a variety of sample dimensions can be supported. We have mounts that can hold samples up to about 2 inches in diameter. Among these, we currently have brackets for up to three polished petrographic thin sections of 27 x 46 mm, up to six round samples of 1″ (thick profiles or cut cylinders), two cylinders of 1.25″ or 1.5″ in diameter, one cylinder of 2″ and several devices that can hold small samples that we usually mount (by pouring epoxy resin) in brass tubes of 1/4″ (up to six per device) or 3/32″ (up to 43 per device). Small free samples for surface analysis or imaging can be glued to stumps with a diameter of 1″ (with or without a central well for particles larger than 1 mm in height). We also consider using certain criteria such as: time (at least X number of villages within a given tehsil were included in X number of surveys over time), geography (villages in a given tehsil are spread over the tehsil), village resampling (villages were not oversampled and if they were, we will select data from a single survey body), Urban/rural (villages are adequately distributed between urban and rural areas).
What do you think of this tehsil selection approach, which would give us fairly representative data for analysis? Note that we also do not have a population size at the tehsil level. Thank you very much! (5) Economical: A sample must be economical. The objectives of the survey should be achieved with minimal cost and effort. I would like to ask if I would like to sample NGOs in my country, which is located in Malaysia, whether I should use stratified random sampling (probability sampling) or targeted sampling (non-probability sampling). My target group are NGOs that are exclusively dedicated to the preservation of biodiversity. Probability sampling techniques are superior, but the cost can be prohibitive. For the early stages of a study, non-probability sampling techniques may be enough to give you an idea of what you are dealing with. For detailed information and results you can rely on, switch to more sophisticated techniques as the study gains speed and adopts a more concrete structure.
2 REQUIREMENTS FOR A GOOD SAMPLEreproduce characteristics of interest to the population as accurately as possible Representative: Each sampling unit represents the characteristics of a known number of units in the population. How: A stratified sample essentially attempts to reproduce the statistical characteristics of the population on a smaller scale.