Subscribe to the Compass blog on Telegram: there you will find articles, tips and cases on how to study potential customers and create a successful product.
In quantitative research
Representativeness of the sample is very important when conducting quantitative research. These include:
all types of surveys - online and offline;
alternatives testing or A/B testing;
hall tests and home tests;
All types of quantitative analysis answer the question "What proportion of the audience has this opinion?" Therefore, it is extremely important to comply with the basic requirements for selecting respondents. For example, if a company produces cosmetics intended for women aged 30-40, then there is no point in interviewing men, schoolgirls or pensioners - such a sample will not be representative.
To obtain statistically significant results, you need to survey at least 100 people, and preferably 200 or 300. In this case, the error size is moderate.
Interesting fact: the accuracy of measurements depends on the number of people south africa phone number list surveyed. But a large sample size does not guarantee the correctness of the results. Representativeness is ensured only by the correct selection of study participants and the use of appropriate data collection methods.
An example of a study with a bad sample from a joke: an online survey showed that 100% of the population uses the Internet.
The only case where you don't need to use a representative sample is if you have a very narrow audience. For example, a business operates in the B2B sphere and cooperates with factories producing steelmaking equipment. There are few such enterprises in the country - it is easier to directly communicate with representatives of each company, rather than study them selectively.
Representativeness of the sample is important in all types of research where the audience is large.
Sample types
Samples can be random or targeted.
A random sample is considered the most accurate and representative, but it is extremely difficult to create. To do this, you need to have access to all elements of the general population. In the case of surveys, this is the contacts of all people who are part of the target audience .