Collecting accurate data is fundamental to a good information system. If the data doesn’t reflect reality collecting it is pointless. If reduced to its most basic structure data is 1’s and 0’s as suggested by the picture. However, the only time data is collected in this format is when communication is either between two computers or a computer and a digital sensor. So nobody is suggesting that any human is likely to accurately enter these ones and zeros.
In any case, one of the greatest impediments to collecting accurate data, even in a more manageable format is people! So, one of the main approaches we at Orpheus Technology recommend is to minimise or, if possible, totally eliminate human involvement and leave the communication up to the machines and sensors. Minimisation of human involvement in data entry is relatively straight forward – for example rather than typing values in from scratch where spelling mistakes can be made or keys mis-hit the accuracy can be virtually guaranteed by offering a list of values (provided the list is finite) from which the user must choose data values.
Another method to maximise accuracy in human data entry is to employ data validation in each field on the data entry form. Primarily the data type has to be checked. If numeric data is accidentally entered into a character field clearly it is not useful data! Also if you’re collecting data that you know contains a certain character, such as “@”, while not guaranteeing that the value is correct, you could scan for that character and if it doesn’t exist the data is certain to be in error.
If you want to maximise collecting accurate data then you must opt for automation and remote sensors to supply the data to your system. Once you have made the up-front investment in whatever sensors you need and the software to interpret the data you can go about your everyday activities while the sensors and computers don’t just start collecting accurate data but collect it in bulk 24/7.