Limitations and Implications

Several limitations of this study must be noted. Even though I analyzed 14 variables that are indicative of stress, there are other variables that were not included but could have been useful (for example, a variable for crowding in households). To get a robust indicator, the variables related to the index should be exhaustively identified and included into the analysis.

Knowing about the variables that affect the index under construction is very important. For example, in this study we know that stress is strongly related to economic conditions. If the variables related to income are excluded from the analysis, the result index looks as shown in the map below and it correlates highly with Minor Population. Care and prudence should therefore be exercised when selecting the variables.

Also the practical usefulness of determining this type of index may be limited because census data is only available once every five years and by the time it is released; about two years have elapsed since the survey was done. Given the dynamic nature of neighborhoods, the index should be constructed as soon as the data is collected.
The index should also be periodically tested to see if it is relevant.

A helpful aspect of this analysis is that if you have the data for the variable that corresponds to the first principle component factor, it would give you a good approximation for the stress index.

^ Top