This is a study of low income family expenditure patterns in urban Canada based on data contained in the National Urban Low Income Family Evaluation study. The structured interview method with open-ended questions as set up in the NULIFE study offered data gathered between January 1 to April 15, 1967. Using this data, the relationships between total family income and expenditures on shelter, food and all other expenditures were examined. This data also shows that the relationship between expenditures on shelter, food, and other basic expenses were related to family sizes of one, three, and five person families within Vancouver, Winnipeg and Halifax. Correlation coefficients yielded two negative correlations about expenditures on shelter and food, inferring a positive correlation about expenditures on all other expenses. Findings showed a relationship between income and expenditure as follows: as income increases, percentage of expenditure on shelter and food decreases. At the same time, there was an increased percentage of income spent on other categories grouped together. Relationships for different family sizes and regions indicated trends between the expenditure on shelter, food, and all other basic expenses. Public housing areas within the Winnipeg and Halifax samples influence shelter costs. The one person family had the highest percentage expenditure on shelter in all three cities, whereas the five person family generally had the highest percentage expenditure on food. The three person family had the highest percentage expenditure on all other basic expenses. The Vancouver sample generally had the highest income. With a slightly lower income than Vancouver, the Winnipeg sample had the lowest percentage expenditure on shelter and the highest on all other basic expenses. Halifax, with the lowest income had the highest percentage expenditure on shelter, generally the highest on food, and the lowest percentage expenditure on all other basic expenses. This project has limited value due to the fact that the computer did not produce data such as age and composition of family units which might have yielded more detailed information. / Arts, Faculty of / Social Work, School of / Graduate