-
Chapter 6. Data and Analysis
- University of Ottawa Press
- Chapter
- Additional Information
97 6 DATA AND ANALYSIS This chapter presents the results and interpretations of Integrative Distance Analysis (IDA) and related analyses of the data classes.It begins with an outline of data-collection procedures and moves on to discuss the primary data classes of artifact assemblages, faunal assemblages, and ethnographic languages. Primary data results are presented in IDA stages as outlined in Chapter 5. Next, the important supplementary data classes are discussed and compared to primary results. Primary Data Classes Primary data from 149 Gulf of Georgia archaeological sites are listed inTable 6.1 and plotted in Figure 6.1.Variation in artifact assemblages,faunal assemblages,and ethnographic language were compared to each other and to geographic location using IDA.A supplementary set of data classes is included in the analysis but was not in a form readily applicable to IDA statistical analysis.The supplementary data classes are art style and mortuary architecture. These data are relatively rare, have low sample sizes per site component, and often are not known from excavated site contexts.Although there is very little quantitative data for these classes upon which to base statistics, there is good information on their spatial distribution and maps of their variability will be discussed. Data Collection Data were collected for this book during a series of field visits to the Royal British Columbia Museum and the University ofVictoria inVictoria,BC;the Laboratory ofArchaeology at the University of British Columbia inVancouver, BC; and the Burke Museum in Seattle, WA. Unpublished reports were read and photocopied from the Culture Resource Library, Archaeology Branch,Ministry ofTourism,Sport and the Arts,Province of British Columbia inVictoria; the Burke Museum Library; and the Laboratory of Archaeology Library at the University of British Columbia.Additional resources were gleaned from published works, unpublished theses and dissertations, and personal communications. Artifacts (Full Sample) Artifact classification formed the bulk of the field research for this book. Following Burley (1979,1980),Clark (2000),and Matson et al.(1980),a fifty-one-type artifact taxonomy was employed (seeTable 6.2).This typology was formulated by Burley (1979) based on previous work by Mitchell (1971b) and Matson (1974).The artifact classes do not completely mirror the diagnostic trait lists of Locarno Beach and Marpole presented in Chapter 3. However, the Burley (1979) typology is in widespread usage (Clark 2000; Matson et al. 1980) and has shown statistical power in differentiating Locarno Beach and Marpole assemblages (Clark 2000). 98 DATA AND ANALYSIS [3.235.227.36] Project MUSE (2024-03-28 23:08 GMT) 99 TERENCE N. CLARK 100 DATA AND ANALYSIS 101 TERENCE N. CLARK Artifacts were analyzed and coded according to Table 6.2.Artifacts that fell outside the classificatory scheme were excluded from analysis. Tallies for each site component were created.Assemblages were standardized to eliminate bias of very large or small assemblages. Standardization was proportional and was achieved by simply dividing the count of each artifact type by the sample size for that assemblage.This created a proportional score for each type, the sum of which for each assemblage equalled one.This step ensured that each site component had equal weighting in all statistical procedures.This thesis uses sixty-four archaeological site components with 14 148 artifacts (seeTable 6.1).Assemblage size ranged from twenty-one (Richardson I) to 798 (Pitt River II), with an average of 221 artifacts. 102 DATA AND ANALYSIS The results of artifact classification are discussed below and proportional data are presented in Appendix 1. The pair-wise distance matrix for these assemblages is shown below inTable 6.3.Variation in artifact assemblage ranged from a low of 0.1943 (Fossil Bay I to Watmough Bay I) to a high of 1.5039 (Montague Harbour II to Decatur Island West), with an average of 0.9936. Artifacts (Small Sample) To allow direct comparison of artifacts and fauna,an additional set of statistics were run on a subset of both artifacts and fauna for fourteen site components where both data classes were present (seeTable 6.1).The composition of the artifact assemblages is present in Appendix 1. Shown below is the summary distance matrix of the small artifact sample. The average distance score between artifact components is 0.9434.The most similar components were Musqueam Northeast and Tsable River Bridge I (0.4707), while the least similar were Shoemaker Bay II and Decatur IslandWest (1.4749). Fauna (Full Sample) Faunal data have been compiled from previous published and unpublished sources. All taxa were expressed in number...