At a recent meeting, a committee was presented with some data. It took only a cursory glance for all members to recognize that the data were wrong and for these views to be vocally expressed. The Chair calmly said, “Yes, we know the data are wrong. Is there anything else to say or shall we move on to the next item on the agenda?”
The committee just moved through the agenda and the incorrect data remained stored in the organization’s systems waiting to be accessed again to produce more inaccurate and meaningless data. The tacit acceptance of the committee members that the data presented to them were wrong demonstrates the poor state of information management in the organization.
It is likely that the IT system being used to capture the data did not have the flexibility for the users to categorize the data in the way in which they needed the data to be categorized. The coding mechanism was therefore abused and resulted in inaccurate data being reported. These inaccurate data have been immortalized in the organization’s databases and will be used over subsequent years for reporting, benchmarking and data mining. The data are wrong, so how can meaningful information ever be derived from mining poor quality data?
The data collected have been rendered a useless and dangerous resource, providing the basis for inappropriate decisions to be taken. But, what is going to be done about it? Who is going to stand up and say that this cannot continue? Who is going to take responsibility for the problem? It is easy to blame the person who coded the data incorrectly or the IT team who developed the system that lacked the flexibility to meet the needs of the organization. However, the reality is that data are wrong and the decisions based on the data are going to be wrong both now and in the future.
Tackling the problem of poor quality data is not easy, but as someone once told me, nothing worth achieving is ever easy. Organizations cannot afford to ignore poor quality data. Some organizations hope that the problems will be resolved when a new system is developed. A new system may improve the quality of data captured in the future (though the new system may bring with it other data problems too…) but a new system is not a magic wand to improve the quality of all the existing data. The development of a new system may be a catalyst for cleaning some of the existing data but there are still risks. For example, old data may be cleaned but still be of poor quality, or old data may just be discarded as mistakes of the past from which lessons have not been learnt.
We all have responsibility for the data we capture, maintain, and use. We must accept and act upon this responsibility to ensure that high quality information is available to support effective decision-making in organizations.
Further Reading: data quality is discussed in Chapter 7 and responsibility for information is discussed in Chapter 16.
Please use the following to reference this blog post in your own work:
Cox, S. A., (2014), ‘“We Know the Data are Wrong but…”, 30 May 2014, http://www.managinginformation.org/poor-data-quality/, [Date accessed: dd:mm:yy]