Selected conventional migration correlates and the exploration of internal net migration in New Zealand, 1966-1971 : a thesis presented in partial fulfilment of the requirements for the degree of Master of Arts in Geography at Massey University.
Migration research in New Zealand with national data sets is limited. It is generally assumed, however, that the most common trends of population redistribution such as the movement north, in particular to Auckland, the movement from rural areas to urban areas and the increasing concentration of the population in cities are responses to economic variation in the national space economy. A number of general hypotheses are selected to examine the age and sex characteristics of migrants. The research hypotheses, which focus on relationships between net migration and the largely untested correlates in New Zealand of income, employment, unemployment and population, examine the validity of the assumption that internal migration in New Zealand is a response to spatial variation. In addition to the testing of the hypotheses an examination is made of spatial variation within the selected parameters. The research hypotheses are tested at three levels of data aggregation - regions, counties and urban areas. The migration data, which are generated by residual estimates using both vital statistics and life survivorship techniques indicate that the majority of migrants are the young adults with females being marginally more migratory than males. It is the examination of the selected conventional migration correlates that the most important, and in many cases unexpected, results emerge. It is found that aggregate migrant behaviour in New Zealand cannot be predicted from the selected migration correlates. At the regional level the Central Auckland data affect the nature of the entire relationship with large values for both dependent and independent variables. With the omission of this data correlations between variables approach zero. A number of data problems are apparent, however, which may be of importance in explaining the lack of relationships. On the other hand, it is shown that there is minimal spatial variation within the parameters so that regional migration may be the result of noneconomic space preferences rather than economic and demographic variation. At the county level and urban area levels some relationships emerge which are good. There are again some doubts about these relationships as they may reflect a degree of autocorrelation; the higher levels of migration to larger centres of population being simply a function of the population size of these areas. It is concluded that net migration in New Zealand cannot be explained by previously accepted although largely untested economic and demographic correlates.