64th ISI World Statistics Congress - Ottawa, Canada

64th ISI World Statistics Congress - Ottawa, Canada

What makes statistics valuable and of high quality? – an attempt to understand user opinions


64th ISI World Statistics Congress - Ottawa, Canada

Format: CPS Abstract

Keywords: quality

Session: CPS 54 - Teaching statistics I

Tuesday 18 July 4 p.m. - 5:25 p.m. (Canada/Eastern)



As statisticians, we have a simple set of quality components for our statistical products and we strive to objectively measure and achieve these at a high level for our users. Some problems or concepts more abstract than products and services, and the feelings and perceptions associated with them, are more difficult to measure and explore. One of these more elusive themes is satisfaction with quality and its various components and the concept of value. Communicating the purpose and justification of producing statistics has been forcing us, statisticians to repeatedly ask ourselves: is that what we do good enough, is it valuable enough for users? What is more: can we take it granted that quality means the same for them as for us?
In parallel with the data revolution and digitalisation the concept of being valuable as a resource has become widely accepted. However, the question remains, what factors make data valuable, does the data being official of unofficial play a role in this and are data quality and data value linked.
The work of the Task Force on the Value of Statistics has also inspired us to ask our users. Our main research question is: What is the difference between value and quality for our users?

Methods / Problem statement

We want to make an effort to learn what quality components our users are looking for. Our goal is to get feedback from our users by ‘speaking the same language’ and associating the same meaning to the quality dimensions.
In our research we organise focus-groups between homogeneous user groups. A focus-group primarily for test purposes was organised with a heterogeneous group of professional users belonging to different sectors. We now intend to examine the question within more homogeneous groups. Furthermore, in this first focus-group, we have not yet discussed the concept of value, but only the qualitative components and their rank. Our assumption is that the two concepts are very similar, yet different, and that this difference has not yet been clearly captured and tested by users in any of the works we are aware of. We would like to define the present concepts from the perspective of our users.

Results and conclusion

It is hoped that the results of the focus-groups will provide new insights into the differences between what value and quality of the statistics mean.
This research can help us better understand our users and bring our definitions and quality dimensions (in the light of our own professional views) closer to theirs. In the long term, this and any future similar research can help us develop products that are truly what users need and to weigh up the characteristics for which there is typically a trade-off. These results could also serve as a basis for strategic issues of dissemination.