If you publish data yourself, you can specify in a licence how others are allowed to use your dataset and what is prohibited. It is recommended to make the licence clearly visible on the webpage from where the data can be downloaded and, if possible, also include it in each of the data files.
If you have found a dataset which you would like to use in your own research or teaching, the conditions of use for this dataset are important. The licence is usually found in the dataset description on the website of the repository.
An existing dataset may be combined with another existing dataset, or supplemented by data collected by yourself. If you wish to make the resulting dataset available for reuse, you must take the licence(s) of the original dataset(s) into account. The licence you select may not be more restrictive than the licence of the original dataset.