This week our GIScience Heidelberg team member Rene Westerholt most successfully defended his PhD! Congratulations! Very well deserved!
The thesis is located at the interface between spatial analysis methodology and the characteristics of spatially superimposed random variables. Three types of contributions are presented:
(i) the interactions of spatial analysis techniques with spatially superimposed random variables
are investigated;
(ii) novel methods for their analysis and characterization are put forward; and
(iii) the broader context of the discussed matters is explored, including a discussion of similar methodological issues in different fields.
The empirical contribution focuses on estimators of spatial autocorrelation and hot-spot statistics. Thereby, the impacts of mixed geographic scales and adverse topological arrangements on spatial analysis results are investigated.
The methodological contribution is two-fold:
First, a modified hot-spot statistic is proposed that takes account of the geometric characteristics of superimposed random variables. This statistic allows the detection of hot-spots when multiple scales are present in datasets.
Second, a statistical test is derived that allows to investigate the local interactions between the arrangement of random variables in geographic space and their local variance. The latter test can be used to investigate how places, like those represented in geosocial media, are characterized in terms of their endogenous variability.
Exploring the broader context reveals that questions in the analysis of spatially superimposed random variables are not limited to geosocial media, but extend to other areas such as socio-ecological psychology. In summary, the obtained methodological results are an important step towards the notion of a place-based GIS, which is a long-term goal in
GIScience.
The following papers are part of the cummulative PhD thesis in addition to an extensive introduction with synopsis and discussion:
Westerholt, R, Resch, B & A Zipf (2015). ’A Local Scale-Sensitive Indicator of Spatial Autocorrelation for Assessing High- and Low-Value Clusters in Multi-Scale Datasets’. International Journal of Geographical Information Science, 29 (5), 868-887. DOI: 10.1080/13658816.2014.1002499.
Westerholt, R, Steiger, E, Resch, B & A Zipf (2016). ’Abundant Topological Outliers in Social Media Data and Their Effect on Spatial Analysis’. PLOS ONE, 11 (9), e0162360. DOI: 10.1371/journal.pone.0162360.
Westerholt, R, Resch, B, Mocnik, F.-B. & D Hoffmeister (2018). ’A Statistical Test on the Local Effects of Spatially Structured Variance’. International Journal of Geographical Information Science (IJGIS), 32 (3), 571-600. DOI: 10.1080/13658816.2017.1402914.
Bluemke, M, Resch, B, Lechner, C, Westerholt, R, & JP Kolb (2017). ’Integrating Geographic Information into Survey Research: Current Applications, Challenges and Future Avenues’. Survey Research Methods, 11 (3), 307-327.
DOI: 10.18148/srm/2017.v11i3.6733.
Steiger, E, Westerholt, R, & A Zipf (2016). ’Research on Social Media Feeds – A GIScience Perspective’. In: European Handbook of Crowdsourced Geographic Information. Ed. by Capineri, C, Haklay, M, Huang, H, Antoniou, V, Kettunen, J, Ostermann, F & R Purves. London: Ubiquity Press, 237-254. DOI: 10.5334/bax.r.
Westerholt, R (2018). ’The Impact of Different Statistical Parameter Values between Point Based Datasets when Assessing Spatial Relationships’. Proceedings of the 21st AGILE Conference. Lund. Sweden.
