It's not.
Chat with our AI personalities
its better because we often don't have to survey a large population, so a sample is quicker, easier, requires few ressources, little time and can be more accurate if a person is not there to answer it because a sample could represent that person.
Sure it can. But in the survey business, the trick is to select your sample carefully so that they'll be equal, i.e. a sample that is accurately representative of the population.
Sample is subset of the population so sample size and population size is different.However, as a subset can be the whole set, if the sample size equals the population size, you have sampled the entire population and you will be 100% accurate with your results; it may cost much more than surveying a [representative] sample, but you get the satisfaction of knowing for what you surveyed the population exactly.Using a sample is a trade off between the cost of surveying the whole population and accuracy of the result.A census is a survey of the whole population and could be considered that the sample size = population size; in this case the results are 100% accurate.The television viewing figures are calculated using a sample of the whole population and then extrapolating them to the whole population; depending upon how the same was chosen, including its size, will affect the accuracy of the results - most likely not more than 95% accurate.With a carefully selected (that is properly biased) sample you can prove almost anything!
You can estimate a population's size when counting individuals if the density in a sample is greater than the population density.
A random sample is better than a census because it takes less time and costs less.