4theloveofhorsz
New Member
- Joined
- Sep 23, 2014
- Messages
- 2
I have a list of 14,561 people. We need to do a survey and we want to include about 5,000 random people total, but weighted based on the group they belong to (defined by L4 leader).
I calculated really quickly that if I include about 35% of people for each L4 I get to the 5,000ish I need. So if an L4 group has 1,000 people I include a random list of 350 people from that L4 group. If a group has 100 people I include 3 random people from that group.
How can I write some formula that lets me choose a certain number of random people from each L4 group?
I have a table with how many people each L4 has (L4 name frequency), and how many the 35% would be (desired size of each random sample). I have 50 L4s.
This does not need to be exactly perfect, but the sample really needs to be representative of the population of each group. Any ideas?
I calculated really quickly that if I include about 35% of people for each L4 I get to the 5,000ish I need. So if an L4 group has 1,000 people I include a random list of 350 people from that L4 group. If a group has 100 people I include 3 random people from that group.
How can I write some formula that lets me choose a certain number of random people from each L4 group?
I have a table with how many people each L4 has (L4 name frequency), and how many the 35% would be (desired size of each random sample). I have 50 L4s.
This does not need to be exactly perfect, but the sample really needs to be representative of the population of each group. Any ideas?
