Researchers sometimes get very involved with their research topic and feel that it is necessary to form many survey questions to collect very detailed data.
The survey is consequently too long and of below-average quality; but the most awkward fact is that many of the questions are never used during analysis. Therefore it is important to carefully consider if we really need all of the questions.
This is especially true for demography, when we, for example, ask about the region even though it is completely irrelevant to the research problem. It is also not necessary to go into too many details when forming questions (e.g. age) if we know that we are only interested in analysing five (e.g. each spanning 10 years) or even only three age groups (e.g. less than 30, 30-50, 50+), as is customary in marketing.
In such cases, it is much better to ask which age group a person belongs to instead of asking respondents to enter a numerical answer (age, year of birth). This way we avoid time-consuming recoding and make survey participation easier as well. It is also crucial to remember that answering questions about an exact year of birth makes some respondents feel uneasy due to privacy concerns.