Academia.eduAcademia.edu
University of Illinois at Chicago Anthropology 594 — Spring 2012 Social Network Analysis (SNA) in Archaeology and Anthropology N o te s fo r W e e k 3 H OW TO BU ILD A SOCIAL N ETW ORK AN ALYSIS 1. D e fin e your research question, i.e., say specifically what it is you want to discover 2. State unequivocally why answering this research question is laudable and im portant 3. D e cid e what type of an analysis you want to do: For a p la ce -b a s e d SNA analysis, e.g., m apping the Departm ent of Anthropology pick your nodes, i.e., d e fin e a population (e.g., everyone considered by the adm inistration to be in the departm ent) Step # 1: or pick a node to begin with, i.e., s e le ct som eone in the m useum to observe Step # 2: pick a t y p e of relationship, e.g., A tells y ou B is a “friend” or pick a k in d of relational e v e n t , e.g., you see A and B gossiping For a r e la t io n s h ip -b a s e d * SNA analysis pick a t y p e of relationship, e.g., friendship Step # 1: or pick a k in d of relational e v e n t , e.g., gossiping Step # 2: pick your nodes, i.e., d e fin e a population or pick a node to begin with, i.e., s e le ct som eone to observe Qu e s tio n s : 1. What role should theory play in deciding which type of analysis to do? 2. What role should theory play in determ ining what relationships should constitute edges (linkages), and whether these are valued, directional, etc.? 3. What is the distinction m ade here between d e fin e and s e le ct (hint: a priori or a posteriori)? 4. Between t y p e and k in d (hints: em ic or etic, categorical or behavioral)? 5. Is a place-based analysis m ore arbitrary (or subjective) than a relationship-based analysis? 6. From an anthropological (i.e., cross-cultural) perspective, which type of analysis m ight be easier to do? _____ * May be also called an in te ractio n al analysis University of Illinois at Chicago Anthropology 594 — Spring 2012 Social Network Analysis (SNA) in Archaeology and Anthropology 2. Co n s tru ct a data m atrix For a place-based SN A analy sis Step # 1: Construct a blank data m atrix (spreadsheet) for the predeterm ined set of nodes Step # 2: Build a database recording relevant inform ation on each of these nodes Step # 3: Enter values sum m arizing this inform ation into the data m atrix For a relationship-based SN A analy sis Step # 1: Build a database recording relevant inform ation on each node as it is added to the analysis Step # 2: Construct a blank data m atrix (spreadsheet) for the set of nodes to be included in the analysis Step # 3: Enter values sum m arizing this inform ation into the data m atrix Qu e s tio n : 1. What role should theory play when deciding what kinds of data are relevant to your research question? 2. What role should theory play when deciding what kinds of sum m ary values should be entered into your data m atrices?