Table 2.
Probit Estimates of the Effect of Race/Ethnicity and Other Characteristics of MD Investigator Applicants on National Institutes of Health R01 Type 1 Research Grant Applications’ Award Probability, Fiscal Years 2000–2006*
| VARIABLES | Marginal effect [standard error] for applications from MDs | ||||
|---|---|---|---|---|---|
| All | At medical schools |
At other institutions |
At medical schools, with HSR |
At medical schools, no HSR |
|
| Race/ethnicity | |||||
| Asian | −0.026† [0.009] |
−0.013 [0.010] |
−0.051‡ [0.015] |
−0.017 [0.016] |
−0.013 [0.014] |
| Black | −0.093‡ [0.022] |
−0.065† [0.025] |
−0.152‡ [0.037] |
−0.069§ [0.034] |
−0.054 [0.037] |
| Hispanic | −0.000 [0.015] |
0.016 [0.018] |
−0.035 [0.025] |
0.001 [0.024] |
0.036 [0.028] |
| Employer NIH funding rank | |||||
| 1–30 | 0.097‡ [0.012] |
0.132‡ [0.031] |
0.103‡ [0.022] |
0.142‡ [0.043] |
0.129† [0.045] |
| 31–100 | 0.074‡ [0.012] |
0.095† [0.033] |
0.116‡ [0.017] |
0.105§ [0.048] |
0.096§ [0.048] |
| Experience with NIH funding | |||||
| Prior NIH grants | 0.118‡ [0.007] |
0.111‡ [0.009] |
0.132‡ [0.014] |
0.106‡ [0.012] |
0.116‡ [0.013] |
| Served on NIH review committee | 0.070‡ [0.007] |
0.076‡ [0.008] |
0.052‡ [0.013] |
0.071‡ [0.011] |
0.082‡ [0.011] |
| Human subjects research (HSR) | −0.041‡ [0.006] |
−0.045‡ [0.007] |
−0.036† [0.011] |
-- | -- |
| Researcher productivity | |||||
| Citations | |||||
| 3rd quartile (24–84 citations) | 0.033† [0.012] |
0.046‡ [0.014] |
0.003 [0.022] |
0.050† [0.019] |
0.044§ [0.020] |
| 4th quartile (> 84 citations) | 0.069‡ [0.015] |
0.079‡ [0.018] |
0.038 [0.026] |
0.051§ [0.023] |
0.104‡ [0.026] |
| Authorship | |||||
| Ratio of last-authored to total publications | 0.065‡ [0.015] |
0.064‡ [0.017] |
0.079† [0.027] |
0.044 [0.026] |
0.085‡ [0.023] |
| Ratio of single-authored to total publications | 0.050† [0.018] |
0.054† [0.021] |
0.050 [0.033] |
0.009 [0.028] |
0.101‡ [0.030] |
| Total applications | 30,759 | 21,944 | 8,815 | 10,876 | 11,055 |
Models include controls for demographic characteristics, education and training, employer characteristics, NIH experience, and research productivity. Estimates are marginal effects that report the change in probability of receiving an R01 award given an infinitesimal change in continuous independent variables. Marginal effects on dummy variables report change in probability of receiving an R01 award given a change in the dummy from 0 to 1. Multiply marginal effects by 100 to obtain percentage points. Robust standard errors clustered on individual applicant are given in brackets. Sources: NIH IMPAC II, National Science Foundation Survey of Earned Doctorates, and Association of American Medical Colleges Faculty Roster. Total applications from MDs with non-missing information on covariates.
P < .01.
P < .001.
P < .05.