Reginald Golledge (2002) pointed out „the open door of GIS“ for environmental psychology. Geoinformatics, digitally storing and analyzing spatial information, GIScience, and place-based GIS (Goodchild, 2010) have opened up gradually to Social Sciences data. Public health, biology, criminology, and environmental protection already use GIS data on a regular basis. The Detroit Area Study 2001 probed the GIS potential for a Quality of Life (QOL) analysis and stimulated studies in other places (Marans & Stimson, 2011). After a visit of the first author to Michigan in 2006, translated and edited QOL items were used in a local QOL study at Salzburg City with 293 field interviews over 55 hectar. A cooperation was established with geoinformatics at Salzburg University (third and fourth author), which led to a 2007-2012 field survey of 16 Salzburg districts with 802 interviews. The statistical analysis of this dataset was done in 2013 in cooperation with geoinformatics and psychology.

Well-being in the city, the urban interface of environmental and health psychology, has a diversity of possible predictors. A synthesis of objective GIS indicators and subjective geocoded items on urban quality of life (QOL) was considered as useful for a mixed methods approach. 41 US QOL items were validated for Austria by an embedded qualitative survey that gave very similar main subjective QOL dimensions of the Salzburg residents. The QOL items were geocoded, matched with objective GIS data of the Salzburg City planning department, and tested for general predictors and subjective-objective interactions.

The 41 US QOL items, when factor analyzed, gave a reliable 26 item 3-factor solution delivering three dependent variables (environmental/social quality, social roots, subjective infrastructure). Multiple linear regressions used them as DVs, sociodemography and GIS variables as IVs. Relevant predictors were housing density (GIS), residential duration, social relations, a green factor (GIS), and subjective district center perception. The variable social roots showed no GIS predictors. 25.8-47.7% AV-variance was explained by the predictor loadings. District differences for the DVs were tested by MANOVA and found to be highly significant, which underlines the role of city district images, further supported by MANCOVA findings. However, approximately 50% of the total variance is not explained by IVs and will need additional predictors.

At the end of 2013, an official psychology-GIScience cooperation platform was formed and supported by Salzburg University, Z_GIS and iSPACE. First discussions formulated the following topics of interest: 1. Replication of the Salzburg factor analysis and regression results in a city of similar size. 2. Testing whether the US QOL item list validated for Austria is validated for other countries. 3. Further mixed methods evaluation of geocoded QOL items versus GIS data – after housing density and greenery, e.g. infrastructure (shopping, restaurants, public transport). 4. Further study on the QOL effects of city districts, finding more items that explain QOL variance. 5. Introduction of psychological constructs (e.g. stressors) as predictors of QOL dependent variables. A parallel QOL survey in the city of Timisoara is underway and will be reported in a contribution to IAPS23.