A double blind study is one where neither the subjects, nor the people recording the data, know which subjects are on the test procedure and which are in the control group. Additionally, the participants in the study also do not know in advance which group they are in; the testing group with the actual treatment or the placebo group where the actual treatment to the group is unknown. Key to this is the concept of randomization. That is, all people participating in the study have the exact same chance of being in either the treatment, placebo, or control group.
Again: All parties are unaware of which group is which for the purpose of eliminating bias in the study.
The best examples of double-blind tests or experiments are clinical studies of new drugs.
Participants are randomly allocated to one of two groups - one taking the drug, the other taking a placebo, with identical test preparation, but with no drug in it (placebo) and a control group. Often the control group serves as the group in the study which receives the placebo.
Participants are unaware of the group to which they've been allocated at the commencement of the study.
Researchers record the symptoms reported by the patients and take measurements. The researchers do not know which patients are in which group, which prevents their observations being biased.
The above is why the study is referred to as "double-blinded."
At the end of the study, however, it is imperative that all participants are unblinded, that is informed about whether they were or were not in the treatment or control group.
Studies when conducted in this manner are considered the gold standard, especially when the researcher conducting the study has sufficient participants in all groups allowing that researcher to reach statistical findings which can be inferred upon the population versus just the sample of participants.
I hope I have assisted you with this question and that you have an excellent Monday. Please feel free to provide any feedback, especially if you have further questions, below the response in the comment section.