The study was designed for the purpose of understanding spatial and
temporal differences in Saami pastoral communities and reindeer herding
practices that could be related to the ecological research on reindeer
abundance, productivity, and ecosystem change. Such an inquiry requires a
variable-oriented approach that balances comparability with an understanding of
the cultural practices of indigenous communities (Ragin 1987). We used mixed
methods research (Bergman 2008, Creswell 2009, Teddlie and Tashakkori 2009) and
combined quantitative and qualitative approaches to collect and analyze data on
six thematic fields that are expected to underlie differences in the Saami
pastoral community: environmental risks, governance, herding practices, economic
organization, symbolic, and social capital. The dominant design feature was
quantification to (i) obtain representativeness of districts and subgroups of
the Saami pastoral communities by multilevel stratified sampling and (ii) to
ensure comparability of numeric variables by including closed-ended items to
explore predefined research questions. There are, however, few previous studies
on Saami pastoral communities, and reliance on our predefined notions of
mechanisms that explain spatial and temporal dissimilarities could be seriously
flawed. For each of the thematic areas, we therefore included open-ended
questions and probes that invited two-way conversations based on an
understanding of reindeer herding practices (Bourdieu 1996).
Multilevel stratified sampling
Our aim was to gain an understanding of spatial and temporal differences in reindeer abundance and productivity. Because this varies both spatially and relative to subgroups such as position, sex, and age, we used multilevel stratified sampling. Participants were selected from an electronic database provided by the Reindeer Husbandry Administration, which included data on reindeer herding units and household members, as well as reindeer abundance and productivity. At the first level, neighboring districts with high contrasts in abundances and productivity over the last 20 years were selected (20 of 34 on common winter grazing land in Finnmark). Because these districts vary according to number of licensed herding units, we differentiated the number of Saami pastoralists interviewed according to size of each district. We interviewed 3 reindeer owners in districts with less than 8 herding units, 4 in districts with 8-14 units, and 5 in districts with more than 14 units. We used the official database on herding units, including leaders and household members, to select participants. To limit our sample to active reindeer herders, we excluded those herding units that had, on average, less than 20 reindeer from 2000 to 2006. From June 2007 to August 2008, we interviewed 77 reindeer herders, who represented 35% of the total number of herding units in the 20 districts.
At the next level, we sampled for heterogeneity to obtain a diversity of opinions about herding practices. In all districts, we included three main categories: the district foreman and two herding unit leaders. The two herding unit leaders were ranked and selected according to the highest and lowest average numbers of reindeer from 2000 to 2006 to investigate differences in opinion between small and large reindeer herd owners. If reindeer herders did not want to participate in the study, we moved to the next herder on the ranked lists. In districts with more than eight units, we assigned reindeer herders to categories according to kinship groups (siida), sex, and age. We targeted the categories that were not yet well represented by the previous selections and randomly selected from these. Reindeer herders who spend most of their time working with the herd are difficult to get in contact. We used 1.5 years to establish contact with herders, which resulted in 84% participation from the first and second priorities on the lists. The amount of time used to establish contact also limited the sample size of the interview inquiry.
The interviews were conducted by two of the authors; Vera Hausner (Human Environmental Science) and Johnny-Leo Jernsletten (Anthropology). First, we established contact with the participants by telephone. It was important that the participants could decide for themselves on the settings of the interviews, which took place in private homes, herding cottages, cafés, offices, and outdoors. The settings may have resulted in some qualitative differences between interviews, but the alternative would have been reduced representativeness of the Saami pastoral community. We could not use a tape recorder, since many Saami pastoralists are skeptical about assurances of anonymity, but both researchers took notes and transcribed them together afterward. The interviews were structured according to thematic frameworks, which allowed for conversational and two-way face-to-face communication. We used a general interview guide approach (Teddlie and Tashakkori 2009), where the wording and the order of questions was adjusted according to the course of each interview. We started with informal conversations in which we probed interests, and closed-ended items were usually introduced at a later stage. One of the interviewers could conduct the interview in the Saami language, but this was only necessary in two cases.
Our standards for ethics were approved by the Norwegian Social Science Data Services and included a letter of information about the study before the start of the interview to secure voluntary participation and a perusal of the final transcript of the interview, if requested.
The data were analyzed using multiple techniques. The interviews were mapped for codes and themes using NVivo 8 (11-15). All interviews were classified as an individual “case” (casebook) with personal characteristics such as sex, age, and type of participant. In addition, all “cases” were classified with geographical variables for the districts: East/West Finnmark, Island/Inland, and High/Low density. The variables had to be standardized according to the closed-ended items by quantitizing the strength of variables using fuzzy logic (Ragin 2000, Teddlie and Tashakkori 2009). Differences and similarities between districts and subgroups was analyzed using qualitative comparative analyses (Grimm and Rihoux 2006), multiple correspondence analysis (Greenacre and Blasius 2006), and homogeneity analyses (Gifi 1990, de Leeuw and Mair 2007).
Bergman, M. M. 2008. Advances in mixed methods research: theories and applications. Sage, Thousand Oaks, California, USA.
Bourdieu, P. 1996. Understanding. Theory Culture & Society 13:17-37.
Creswell, J. W. 2009. Research design: qualitative, quantitative, and mixed methods approaches. Sage, Thousand Oaks, California, USA.
de Leeuw, J., and P. Mair. 2007. Homogeneity analysis in R: the package homals. Department of Statistics Papers, Department of Statistics, UCLA, Los Angeles, California, USA.
Gifi, A. 1990. Nonlinear multivariate analysis. Wiley, Chichester, UK.
Greenacre, M. J., and J. Blasius. 2006. Multiple correspondence analysis and related methods. Chapman & Hall/CRC, Boca Raton, Florida, USA.
Grimm, H., and B. Rihoux. 2006. Innovative comparative methods for policy analysis: beyond the quantitative-qualitative divide. Pages XIV, 344 s. Springer, New York, New York, USA.
Ragin, C. C. 1987. The comparative method: moving beyond qualitative and quantitative strategies. University of California Press, Berkeley, California, USA.
Ragin, C. C. 2000. Fuzzy-set social science. University of Chicago Press, Chicago, Illinois, USA.
Teddlie, C. and A. Tashakkori. 2009. Foundations of mixed methods research: integrating quantitative and qualitative approaches in the social and behavioral sciences. Sage, Thousand Oaks, California, USA.