Paper Rejected - Human-Computer Interaction and Innovation in Handheld, Mobile and Wearable Technologies - page 327

Figure 1. Three plots of the same data - six tasks on two systems (means only; means plus 95% confi-
dence interval error bars; and means plus scatter plots)

CONCLUSION perimental work in mobile-HCI to read the papers

in our, deliberately short, bibliography.

This short paper has aimed to raise awareness in
the mobile-HCI community, and perhaps the wider
HCI community, of the fundamental concerns that
other disciplines have raised over null-hypothesis
testing. While we do not appear to be as bad at
“statistical sinning” as other domains, we cannot
afford to be complacent - particularly as confer-
ences and journals tend to become monotonically
harder to publish in, poor understanding and
treatment of null-hypothesis testing may seriously
affect the types of papers and results that make it
through to publication. We therefore encourage
anyone involved in writing up or reviewing ex-

REFERENCES 1

Cairns, P. (2007). HCI... not as it should be:
Inferential Statistics in HCI Research. Proceed-
ings of HCI 2007 (People and Computers XXI).
Lancaster, UK.

Cliner, J. A., Leech, N. L., & Morgan, G. A.
(2002). Problems with null hypothesis signifi-
cance testing (NHST): What do the textbooks
say? Journal of Experimental Education , 71 (1),
83-92. doi:10.1080/00220970209602058