Weighted Least Squares Regression
Source variable.. FILL93 Dependent variable.. RURGRAD
Source variable.. FILL93 POWER value = -.500
Dependent variable.. RURGRAD
Listwise Deletion of Missing Data
Multiple R .70411 R Square .49577 Adjusted R Square .46851
Standard Error 33.92742
Analysis of Variance:
DF Sum of Squares Mean Square
Regression 18 376871.58 20937.310
Residuals 333 383306.19 1151.070
F = 18.18944 Signif F = .0000
------------------ Variables in the Equation ------------------
|
Variable |
B |
SE B |
Beta |
t |
sig |
Source |
|
Number of rural months |
0.897591 |
0.17216 |
0.22086 |
5.214 |
.0000 |
Survey |
|
% residents black, hispanic |
-0.198845 |
0.04494 |
-0.20625 |
-4.424 |
.0000 |
ResData |
|
Program mission is rural |
11.602491 |
2.81746 |
0.18731 |
4.118 |
.0000 |
Survey |
|
% Rural Population in State |
0.175848 |
0.06033 |
0.13794 |
2.915 |
.0038 |
Census |
|
Program Director is rural contact |
5.714866 |
2.1357 |
0.1201 |
2.676 |
.0078 |
Survey |
|
Partial rural mission |
5.015762 |
2.13954 |
0.10336 |
2.344 |
.0196 |
Survey |
|
Number of OB months |
2.260669 |
0.97968 |
0.102 |
2.308 |
.0216 |
Survey |
|
Number of other non-FP programs |
-1.02811 |
0.46258 |
-0.11026 |
-2.223 |
.0269 |
ResData |
|
% females in program |
-0.127418 |
0.06209 |
-0.08605 |
-2.052 |
.0409 |
ResData |
|
Any fellowship, rural/procedural |
4.62451 |
2.51611 |
0.07483 |
1.838 |
.067 |
Survey |
|
City population at main program site |
-1.22E-06 |
7.14E-07 |
-0.07629 |
-1.713 |
.0876 |
Census |
|
Public hospital or private |
2.894421 |
2.16724 |
0.05805 |
1.336 |
.1826 |
ResData |
|
% of faculty that had been rural docs |
0.048292 |
0.04077 |
0.05588 |
1.184 |
.2371 |
Survey |
|
% of residents who were D.O. |
6.49223 |
6.93487 |
0.03836 |
0.936 |
.3499 |
AMA masterfile |
|
Type 1 program |
-3.773988 |
5.60184 |
-0.02994 |
-0.674 |
.501 |
ResData |
|
Title VII grant funds to program |
1.11E-05 |
2.13E-05 |
0.02101 |
0.52 |
.6035 |
HRSA |
|
Type 2 program |
-0.785059 |
3.12744 |
-0.01721 |
-0.251 |
0.802 |
ResData |
|
Type 3 program |
-0.114516 |
3.36903 |
-0.00199 |
-0.034 |
0.9729 |
ResData |
|
(Constant) |
20.337683 |
5.98834 |
3.396 |
0.0008 |
Survey data had 96 % response rate from programs
ResData - Data from individual programs as published in Residency Directory correlating with same years of survey, 1992 - 1994
Results - Basically no change in the original analysis. Programs with more rural months, fewer minorities, a mission for rural, a location in a more rural state, a program director who was the rural contact, a partial mission for rural, more ob months, fewer other residency programs to compete with, a lower % female residents (barely contributes). Borderline contributions in programs with a rural or procedural fellowships and those with smaller city populations at the main program site.
The following variables did not contribute, including residency programs in public hospitals, the % of faculty that had been rural docs, the % of residents who were D.O., program type, or Title VII grant funds (examined by year, amount, or any at all).
Fam Med Residency Programs and the Graduation of Rural Family Physicians
Medical Schools and Rural Graduation Rates
Leadership Factors in Developing RME
Research in Rural Medical Education