Activity 2: Introduction to Literature Searching

HCI... not as it should be: inferential statistics in HCI research

Paul Cairns
University of York
Heslington
York, YO10 5DD
+44 1904 434336
p.cairns@cs.york.ac.uk

http://www-users.cs.york.ac.uk/~pcairns/papers/Cairns_HCI07.pdf

This paper surveys the use of inferential statistics over the last two BCS HCI conferences and the last year (2006) of two leading HCI journals. Of the 80 papers covered, 41 used some form of inferential statistics. However, all but one had some form of problem of reporting or analysis that undermined the value or the validity of the statistical testing and hence the research findings. This paper discusses the implications of such widespread issues for HCI research and considers approaches for improving the use of statistics in HCI.

This study is broken into distinct sections such as

     1.HCI AND STATISTICS
     2.APPROACH TO THIS STUDY
     3.PAPERS CONSIDERED
     4.SURVEY FINDINGS
     5.IMPLICATIONS
     6.RECOMMENDATIONS
     7.ACKNOWLEDGMENTS
     8.REFERENCES

The aim of this paper is to push those involved with HCI to adopt a standard for adequate reporting and execution of statistical analysis. Additionally, in these times of free and easy flow of data, it should be possible for journals and conferences to set up a repository of data associated with each paper and it is a requirement of publication that such data is submitted. Thus, whilst authors and referees strive to do good statistics to the limit of their knowledge, all readers are ultimately able to check the quality of the statistical analysis by doing it for themselves.

The underlying problem seems to be due to a broad lack of adequate statistical education. The basic knowledge of how to do good statistics is not at all ingrained in HCI researchers.



Research Literature Sample |  Introduction to Literature Searching | Answer an Online Survey
Creating an Online Survey | Data Analysis