Explaining how properties at the level of individuals translate into properties at the level of collectives is a core objective of sociology. Because the social world is characterized by complex webs of social interdependencies, establishing how micro and macro are related to one another requires a detailed understanding of how individuals are influenced by their social environments and the consequences that such influences have for the dynamics of the social process. However, until very recently, it has been difficult to conduct detailed empirical investigations of micromacro linkages due to the lack of large-scale data containing information on how individuals interact with one another. In the absence of such data, substantive research has tended to (a) focus its attention elsewhere: studying how social factors influence individual outcomes, rather than how actors in interaction with one another bring about collective outcomes, or (b) propose models of micro-macro linkages that-for reasons of parsimony and tractability-often assume artificially high levels of homogeneity. Against this background, this thesis sets out to investigate, first, how the data and tools that have emerged from the digital and computational revolution can help sociologists construct empirically well-founded mappings from the micro to the macro level, and second, how the conclusions about the role of social interdependencies and networks change when the analysis is informed by real-world heterogeneities.A central theme of this thesis is the often-surprising effects of social interaction. As it happens, the thesis itself is an unexpected result of long line of such interactions. My journey from statistics to sociology and the result, this dissertation, would not have been possible without the input and support from the many people I have had the privilege to interact with along the way.First, I would like to thank my terrific supervisors, Peter Hedström and Marc Keuschnigg. Thank you for giving me the confidence, support, and inspiration to develop beyond my methodological interests. Thank you for all the insightful discussions. Thank you for the many rounds of excellent feedback that I have received over the years and thank you for our collaborations. Working and coauthoring with both of you has been an absolute pleasure, and I have learnt a tremendous amount. Your influence has certainly been of the wide kind! I feel very lucky to have done my PhD at such a great workplace as the Institute for Analytical Sociology (IAS), surrounded by amazing colleagues and researchers. During these years, there has never been a shortage of interest in discussing ideas and providing feedback. Countless of such discussions, formal and informal, small and large, at seminars, in the lunchroom, in the hallways, at the pub, have undoubtedly shaped my thinking and made this thesis better than it otherwise would have been. A special thanks to