Data Quality Definition

3222 Words7 Pages

. Data Quality.
It is very difficult to say what data quality means. The word quality itself has different meanings for different people. Even for an individual the word may has a different meaning regarding the circumstances. I will use an example to try to make it clearer. If we ask three people to tell us which car they believe is a quality purchase, we will get three or even more different opinions. Some people will say that acceleration matters, others will say security. Some would prefer an environment friendly car, others a low price one. This is the reason why there is not a universally agreement of what data quality means, so quite few definitions exist. In my opinion Joseph Moses Juran’s definition is the most representative and summarizes the most of the existing definitions. According to Joseph Moses Juran “Data are of high quality if they are fit to their intended uses in operations, decision making and planning” [1].
So how we know if our data are of good or bad quality? Following the car example, in order to answer this question we have to define which characteristics we should take into consideration and how much each of them weights. It is also important these characteristics to be measurable. A continuous research in this field providing us a great range of data attributes along with a ranking, according to their importance. In data quality literature these attributes referred as dimensions. From now on we will use this term when we talk about data quality characteristics. In chapter 2.2 we will present the data quality dimensions in detailed.

2.1 Why good data quality is critical.
Back in 2006, Clive Humby a mathematician from Sheffield said that “Data is the new oil”, in an attempt to highlight the signific...

... middle of paper ...

...such as the “heterogeneity of their components” and security issues.

d. Cooperative IS. According to Massimo Mecella et al. “is a large scale information system that interconnects various systems of different and autonomous organizations, sharing common objectives” [30]. The main problems with these information systems are the many copies of the same objects (duplicate copies) and the possibility of poor data quality from one source to spread through the cooperative systems. Thus it is very important for the individual information systems to be trusted.

e. Web IS. The importance of web led the classical information systems to transform in order to integrate with web technologies. This means that a web application can access an organization’s dataset. And as we mentioned above, this integration creates new data issues, like security and accessibility.

f. P2P IS.

Open Document