Statistical Analysis, Participant Observation and Key Informant Interviews Summary

44 participate in proposed projects and perceived value of proposed projects were each calculated in the same way by adding each response and dividing by the number of responses to create summary measures for each of the four variables of perception of recovery activities.

3.5.5 Background Information

Background information gathered from subjects included village of residence, address, household size, sex, age, religion, years of formal education, and years living in current village. Subjects were also asked to rank by relative significance, sources of livelihood for each household.

3.6 Statistical Analysis, Participant Observation and Key Informant Interviews

The variables used for statistical analysis are attitudes toward the occupation of fishing, level of fatalistic thinking, perception of the future, spending choices if given 9,000 baht, spending choices if given 110,000 baht, damage to productive and household materials from the tsunami, willingness to participate, and perceived value of the projects. Initially, a canonical correlation is performed as an exploratory technique for two or more sets of variables to assess overall correlation between the sets of variables Harlow 2005: 180. Four standard multiple regressions were performed to follow up on the canonical correlation, each using one of the four variables of ‘Perception of Recovery Activities’ as a dependent variable with each of the other three ‘Perception of Recovery Activities’ 45 variables used as independent variables Gotwals, Dunn and Wayment 2003. Data from participant observation and key informant interviews provide additional information to support and clarify the findings from statistical analysis Pollnac and Crawford 2000.

3.7 Summary

This chapter describes the location of the study and how this investigation may assist the Post-Tsunami Sustainable Coastal Livelihood Program to effectively achieve its goals. The methods for data gathering and analysis are then explored. The questions from the survey used to create each variable to be analyzed are also discussed along with how the responses were coded and calculated. 46 CHAPTER IV ANALYSIS AND RESULTS This chapter will address the research question and hypotheses proposed in Chapter Two. First, this chapter will summarize relevant information from household surveys to provide an overview of the sampled population and their occupations. Measured variables, as discussed in Chapter Three, will be analyzed using canonical correlation, two follow-up multiple regressions and two multiple regressions using only those respondents that evaluated at least one current project. The results of the statistical analysis will be further explained with pertinent data from participant observation and key informant interviews.

4.1 Description of Overall Demographics