It's also important to design data collection methodologies while keeping in mind data leakage or confounding variables 

e.g. Block Design for Classification Experiments is not smart (e.g. think of this word for 10 seconds, think of another for 10 seconds)

It's also important to design data collection methodologies while keeping in mind data leakage or confounding variables 

e.g. Block Design for Classification Experiments is not smart (e.g. think of this word for 10 seconds, think of another for 10 seconds)

https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9264220