Outlier Institutions
This type of analysis helps to provide an effective tool for identifying which institutions perform well above and well below statewide averages. It is a useful tool in identifying which institutions may need to improve their rates of female applications and originations and in which sectors they might need to concentrate.
APPENDIX I, Table 21 contains a chart that graphically organizes the individual institutions and graphs each institution’s performance by census tract. The basis of the graph was female-to-male ratios of application rates. The statewide average ratio in this category is .25. An analysis of the outliers in this area showed that most of the outliers were institutions that had significant numbers of applicants for the middle and upper income census tracts, but zero for the low and moderate-income census tracts. Among these underachievers, thirteen were small institutions having under 200 applications overall. Seven were larger institutions. This observation might suggest that smaller institutions lack the volume of activity to attract sufficient numbers of low-income applicants or that the smaller institutions have very few branches, which may not be near a low-income census tract.There were fewer institutions that had high-ratios compared to the state average ratio, and almost all were in low-income areas. There were two small institutions and seven large institutions over a .80 ratio. However, for these institutions, the other census tract ratios were close to the .25 average. For these institutions that were over the .80 ratio, both small and large institutions had very few applications in the low-income tract area. This explains the high-ratio. If only one female and one male apply, a 1:1 ratio results and skews the data. Again the low activity in the low-income census tracts explains much of the outlier behavior in the data.
APPENDIX I, Table 22 contains a chart that graphically organizes the individual institutions and displays each institution’s performance by census tract. The basis of the graph was female-to-male ratios of origination rates. The statewide average ratio in this category is .98. An analysis of the outliers in this area showed that most of the outliers were institutions that had significant numbers of originations for the middle and upper income census tracts, but zero for the low-income census tracts. Among these underachievers, twenty-two were small institutions having under 200 originations overall. Eleven were larger institutions. In combination with the observations of application activity, this observation supports the rationale that smaller institutions may lack the volume of activity to attract sufficient numbers of low-income applicants or that the smaller institutions have very few branches, which may not be near a low-income census tract. Additionally, among those applicants from low-income census tracts, very few are being originated. This results in higher amounts of ratios equaling zero.There were fewer institutions that had high-ratios compared to the state average ratio, and almost all were in low-and moderate-income areas. There were two small institutions and four large institutions over a 1.3 ratio. For these institutions that were over the 1.3 ratio, both small and large institutions had few applications in the low and moderate-income tract areas. This explains the high-ratio. If very few females and males apply, a higher ratio results and skews the data. Again the low activity in the lower income census tracts explains much of the outlier behavior in the data.
|
|
|
|
|
|