www.igroup.org – project consortium
a multi-disciplinary project consortium.
addresses new interfaces between humans and the real and virtual environment.
defines itself through project work.
The scale construction is described in our article The Experience of Presence: Factor Analytic Insights (Schubert, Friedmann, & Regenbrecht, 2001, Presence). We briefly describe the construction here.
We distinguish theoretically between immersion and presence. While immersion is a variable of the technology and can be desribed objectively, the sense presence is a variable of a user's experience. We therefore rely on self-reports for measuring presence. We think that so-called objective measures of presence, like postural sway measures, tap mediating processes like the cognitive processes involved in understanding the virtual environment and creating a mental model of the VE.
In two survey studies on presence and immersion experiences, questions from previously published questionnaires, questions from own past research and newly designed questions on experiences of immersion and presence were combined. In the two studies, data from 246 and 300 participants were collected. respectively.
The manuscript describes how different factors emerged for presence and immersion experiences - that is, subjective experiences of being in the virtual space, and subjective evaluations of technology and interaction possibilities. We will here focus on the presence scales exclusively. If you need further information on the immersion scales, please contact us.
Three different presence components were identified in the factor analyses. On the basis of the study 1 sample, a three-component presence scale was developed using structural equation modeling (SEM). The goal was to develop short and orthogonal subscales. The resulting model was then tested on the independent dataset from study 2, which is a crucial step for scale development. The model did indeed fit the second data set.
The scale has an unorthodox approach since it combines two different strategies: On the one hand, the SEM had the goal of developing independent subscales, which do not strongly intercorrelate. On the other hand, the resulting model features a general item loading on all three subscales. This item is, not surprisingly, the general definition of the sense of presence: "I had a sense of being in the virtual environemnt" (Slater & Usoh, 1994). Note that this general item was identified on purely empirical grounds, but loads together with the subscales on the second-order factor General Presence. This fact solidly validates our scale. The structure of the SEM model is illustrated in figure 1.
(For readers not familiar with factor analysis and structural equation modeling, we recommend Kline, P. (1994). An Easy Guide to Factor Analysis, London, New York: Routledge).
The scale development process identified one general item, five items for Spatial presence, four items for Involvement, and three items for Experienced Realism. We felt it necessary, however, to add one more item for the last scale for practical purposes. We chose one item that showed a mild double loading on Involvement. The resulting IPQ therefore includes 14 items.