Representative data sample. How to do it? If I have population [x1, x2, x3 ... xn] and I want representative data sample. I need to provide equal chance for each x, will be into the representative sample. For example n = 1000, must to provide chance 0.1 percent for each x will be into the my sample. This sample will be representative.
Need to get reasonable size of it without mistakes and from complete population also. Reasonable if cost of this work isn't conflict with management' planing about costs, risk and profit.
Comparison between result of two representative sample possible no all time.
Data of people is one of the most difficult to get it. May be random isn't best choice here, but it's real good choice.
Long term research provide information to get causal relationships. I think it's about searching correlation.
Lie in the data is important mistake and can show lie in the result. It's about real lie into the data, we have about essence "false" but it's "true".
Repeatable research can help to fix various mistakes and help to planing, because I know that plan is nothing, planning it's all.
If I need to compare any research need to check that their data are equal and I wasn't lost important data in the past.
Conclusions
- Must provide chance 0.1 percent for each x of [x1, x2, x3 ... xn];
- Size of sample of [x1, x2, x3 ... xn] is reasonable where cost of this work isn't conflict with management' planing about costs, risk and profit;
- Lie in the data is important mistake and can show lie in the result. It's about real lie into the data, we have about essence "false" but it's "true";
- Repeatable research can help to fix various mistakes and help to planing, because I know that plan is nothing, planning it's all;
- If I need to compare any research need to check that their data are equal and I wasn't lost important data in the past;
- Long term research provide information to get causal relationships. I think it's about searching correlation;