Country has population of 40 000 000 people. I have sample with data of 1 500 000 people from this country. In this sample 1.9% of people have "pair" (one or more) who has the same first name, last name and date of birth (but is not the same person). Can i compute with this data how many people in whole population have pair with the same name and date of birth?
Computing number of people with the same name and date of birth base on sample data.
1
$\begingroup$
probability
combinatorics
statistics
probability-distributions
statistical-inference
-
0When you say "pair", do you mean "at least one other person"? Note that there could be more than one, which complicates matters. – 2017-02-28
-
0Thanks for your question. I have meant one or more. – 2017-02-28
-
1Whether you can make an inference from the sample depends on how that sample is chosen. For instance if all of the $1 500 000$ people come from one region in the country, which has a culture that uses a limited pool of names, then the $1.9\%$ value is unlikely to apply to the country as a whole. But if the sample was chosen uniformly at random from the entire country, then it is probable that the $1.9%$ value is close to the percentage for the entire country. – 2017-03-01