In short, a sample size calculation is required. If you want to assess the quality of life (QoL) of a specific group of people (elderly residents of a city, people with a specific disease), unless you are surveying everyone, you need to consider sample size and sample design, as well as what is known re the variance/sensitivity/norms for the quality of life measure you are using. Similarly, if want to compare quality of life of a group of people to population norms or track quality of life over time, you always ought to start with a sample size calculation. The only time you might not do a sample size calculation is if you are testing QoL of a 100% sample, but even there, I'd look at the research questions and I'd want to know my statistical power to answer it.
As mentioned by others, you will have to do power analysis to derive sample size calculations, so you can get meaningful data with statistical signficance