Home > Statistics Every Writer Should Know > The Stats Board > Discusssion
sample of the sample
Message posted by kevin on August 11, 2000 at 12:00 AM (ET)
I run a database for a rocknroll band called X. Each person who goes to the web site can enter their name,address, gender and age. I've got about 60,000 kids in the database. 45,000 of those kids told me their age. If I know (I'm guess righ now) that 30% of the kids who told me their age are between 15-17. Can I assume that 30% of the kids are between 15-17 for the entire 60,000 in my database?
READERS RESPOND:
(In chronological order. Most recent at the bottom.)
Re: sample of the sample
Message posted by Lonnie on August 15, 2000 at 12:00 AM (ET)
I'm pretty sure you can only assume that 22.5% of your audience is between 15-17 yrs. old. This is because 45k out of the 60k file is 75% who have an age. Furthermore, of this 75%, you assume 30% are between 15-17 years old. Thus 75% (with an known age of something) times the 30% (in this age group) represents 22.5%. I'd only say less than one in four of your full database are age 15-17 years old.
Re: sample of the sample
Message posted by Bill on August 15, 2000 at 12:00 AM (ET)
The problem may be that the 15,000 people who did not give their age may be significantly older or younger than the 45,000 people who revealed their age. With any research, you do the best you can while acknowledging the limitations of your approach. In the absence of other data, your best GUESS is that 30% of the 60,000 are between 15-17.
Your $5 contribution helps cover part the $500 annual cost of keeping this site online.
