HCI AS MIS - Human-Computer Interaction and Management Information Systems: Foundations - page 24

1985), with little cross-fertilization across areas. MIS is more theoretically advanced (Dillon,
chapter 2 in this volume), and this is where MIS researchers can make another important contri-
bution. Theories can help structure and organize the field, allow knowledge to accumulate, and
provide opportunities to share knowledge across disciplines. MIS can draw on theory to provide
a common language and to help to integrate the multiple HCI disciplines.
Methods and Analytic Techniques
MIS researchers are trained in qualitative methods (such as ethnographies) and quantitative meth-
ods (such as controlled experiments). Because of their knowledge of measure development,
including reliability and validity issues, they can provide valuable contributions to the HCI com-
munity by enhancing HCI measures (Zhang et al., 2002). They also value large-scale sampling.
MIS researchers have been trained to use a variety of analytical techniques, ranging from qual-
itative analyses (such as NVivo) to quantitative analyses (such as structural equation modeling).
For instance, quantitative techniques that take into account multiple levels of analysis, such as
WABA (within-and-between analyses) and HLM (hierarchical linear modeling), are particularly
appropriate for HCI research.
WHAT'S HOLDING MIS BACK?
Given that MIS appears to be in an excellent position to contribute to the development and growth
of HCI, one key question remains: What is holding HCI back from taking advantage of this
opportunity? Several potential reasons have been suggested.
First, MIS is a young field itself, and not long ago it faced many of the same criticisms that HCI
currently faces concerning issues such as methodological rigor. This, combined with the fact that
it would be one of the first times for MIS to act as a reference discipline, forces us to struggle with
the question of whether MIS is really ready for this challenge. However, MIS's ability to come so
far in such a short period of time is testament to its growth and ability to learn from previous mis-
takes. In addition, its recent successes in developing theories and utilizing rigorous methods
demonstrate its ability to address important research issues (Baskerville and Myers, 2002).
Second, HCI has been criticized for “testing, rather than inventing, new ideas. It looks at what
has been, not what could be” (Zolli 2003, p. 65). One could argue that this is even more true of MIS
research. Much MIS research is based on a behavioral, rather than a design, science paradigm
(Hevner et al., 2004). MIS researchers are not the developers of emerging HCI technologies, and
large-scale field applications of emerging interfaces are not available for testing. Therefore, some
MIS researchers are willing to simply wait to evaluate technologies after implementation.
However, years of software and technology development indicate that simply testing after devel-
opment will not ensure that products meet end user needs. End-user needs must be considered from
requirement gathering through to implementation. Therefore, if we recognize that computer sci-
ence focuses more on the technology and does not consistently develop products to meet market
needs, then simply testing products from an MIS perspective after development will only continue
to facilitate the current concerns regarding HCI's research relevance and business sensitivity.
MIS's influence on HCI must be felt well before the testing phase. Ultimately, MIS should assist
in deciding which technology-related projects are important and which research questions are
relevant to practitioners and future researchers.
A third issue relates to MIS's identity. One may question if the discipline of MIS has bought into
Carr's argument that IT (and thus, by extension, HCI) is a commodity and does not matter (Carr,
2003). According to Carr (2003), the new rules governing IT management include a decrease in IT