v 
30 
The last data of the study were the responses to the interview question about changes in 
the concept of ownership over the past one or two generations. The individual responses 
appear in Appendix E and a summary in Table 15. The majority of the Cree answered that the 
concept had changed, while the majority of English-Canadians answered that it had not. The 
explanations, however, focused on material conditions and way of life, rather than the concept 
of owning per se. The most frequent explanation for change by the Cree was a new affluence 
compared to the material hardships of earlier years. However, this affluence has been seen by 
some to have caused problems, including crime, less care for people and things, and a general 
change in the way of life, particularly the loss of Cree language and the loss of “bush life”. 
English-Canadians, too, noted affluence and the resulting decrease in attachment to and care 
for property. Some related this to the ease of acquiring goods on credit. Other 
English-Canadians commented on an increase in legal complications in owning and a decrease 
in the owners’ command of their property as a result of government controls and powers of 
expropriation. 
Discussion 
it is possible to draw several methodological lessons from this study. First, it is indeed 
useful to employ a multimethod design in cross-cultural field research. For example, even 
though constrained free-recall had been used successfully in other cross-cultural research 
(e.g. Thomas & Bolton, 1979) and had been successful with four respondents in the pilot work 
for this research, it was not a successful field research task and did not produce useful 
quantitative data. Fortunately, alternative measures were available. 
Second, a redundancy of measures also allows cross-method replication of findings, 
which reduces challenges to validity due to methodological artifacts. Rapoport and Fillenbaum 
(1972) examined several card sort tasks and several techniques of multivariate analyses and 
concluded that it was necessary in psycholinguistics to use multiple methods and to test for the 
randomness of the data, particularly on semantic fields that are terra incognita. 
Third, it seems important to differentiate research tasks requiring high degrees of subject 
compliance from those more amenable to surveying the general population. Most 
psychological research methods have been developed on and for university students as 
subjects (Higbee, Millard & Folkman, 1982). Students apparently are very compliant to research