Thus, if we assume our sample population is roughly representative of the Full Population, we can, with 95% confidence, expect the Full Population's response to vary by up to 12-14% from these poll results. (i.e., only 5% chance that more than 34% of people would respond "like" and less than 42% of people would respond "dislike.") If we *don't* assume that the forum sample roughly accurately reflects the full population of BG3 players, then we can't use this poll to say anything and it's basically useless.
That's the problem, we can't just assume that the views of the forum population are an accurate representation of what the total population will be like. From what I've seen, and I guess most people would agree, this forum is a bubble comprised mostly of players that are already familiar with D&D that also seek to actively partake in discussions regarding the development of the game. The amount of active members of this forum that could be considered casual surely pales in comparison, and isn't proportional to that of the total BG3 casual audience.
This forum just isn't a good population to sample if you want actual representative data of the whole BG3 population, I agree with @Alodar.
Originally Posted by mrfuji3
Also, polls are regularly collected that use ratios MUCH smaller than 10%. US election polls don't survey 15 million people after all. Other sources I'm seeing suggest that "A good maximum sample size is usually 10% as long as it does not exceed 1000."
Obviously no one expects you to survey 100% of the population in order to get reliable data, but if you work with a biased sample you'll get skewed results.
