The first sample surveys in the latter parts of the 19th century were an intellectual social movement. They were motivated by the intention to improve the economic and political conditions of workers. The quantitative survey was considered an ideal because it would present data about the workers as facts, i.e. establish a scientific authoritative truth. In a case study from Denmark, the paper shows how the first survey - a study of seamstresses - was carried out by bringing several cognitive and organizational elements together: a network of researchers, a method for sampling, the construction of a questionnaire, a procedure for coding, and analyzing the data. It was a trial and error process where the researchers lacked relevant concepts and methods but relied on their intuition and on inspiration from abroad.