This memo summarizes some initial research investigating NU’s recent rating under the Carnegie Classification system as a Highest Research Activity doctoral institution.
This rating represents a jump in research stature given NU received a High Research rating in 2010. Criteria for research activity includes such factors as research and development expenditures, research staff and number of doctoral conferrals. The category of Doctoral Universities includes institutions that awarded at least 20 research/scholarship doctoral degrees during the latest update year. A total of 335 eligible institutions were assigned to the following categories:
- R1: Doctoral Universities – Highest research activity (n=115)
- R2: Doctoral Universities – Higher research activity (n=107)
- R3: Doctoral Universities – Moderate research activity (n=113 )
Several institutions moved in and out of the doctoral granting categories from 2010 to 2015. For example, of the 108 institutions considered among the highest-level research category in 2010, 100 (over 90 percent) remained in that category (R1). Fifteen institutions including NU moved from level 2 to level 1, while eight moved from level 1 to level 2.
Research Rating Methodology
The Carnegie Classification Basic research rating is based on two indices of research activity—aggregate and per-capita—derived from a principal components analysis (PCA) of several correlates of research activity: science and engineering (S&E) R&D expenditures; non-S&E R&D expenditures; S&E research staff (postdoctoral appointees and non-faculty research staff with doctorates); doctoral conferrals by broad disciplinary area (humanities, social sciences, STEM fields, and other fields); and the first three items divided by the number of full-time faculty for the per-capita analysis. To reduce the effect of outliers, all raw data are converted into rank scores to be inputted into the PCA. Based on this analysis, the indicators with the highest loadings were: Aggregate index: Doctorates in STEM (.914), Research Staff (.902) and S&E R&D Expenditures (.900) and Per-capita index: Per-capita S&E R&D Expenditures (.931) and Per-capita Research Staff (.928).
The 2015 classification employs the most recent available data from the source federal agencies — the National Center for Education Statistics and National Center for Science and Engineering Statistics — and a nonprofit agency, The College Board. These data represent the time period of 2013-14. Doctoral degree conferrals by field were based on IPEDS Completions data reporting 2013-14 degree conferrals. Faculty counts were from the IPEDS (HR) staff by occupational category, faculty and tenure status data for Fall 2013. R&D expenditures came from the NSF Survey of Research and Development Expenditures at Universities and Colleges for fiscal year 2014. Research staff data came from the NSF Survey of Graduate Students and Postdoctorates in Science and Engineering for Fall 2013.
The following sections describe NU’s position on these indices, compared to all research universities (n=276), Highest Research Activity institutions (n=115) and a group of comparison group institutions which are typically used as a peer group for NU
1. Doctorates in STEM: Figure 1 displays some statistics for doctorates conferred in STEM fields. Across all institutions reporting data (n=276), the mean number of STEM doctorates was 100 (median=51). Focusing only on the Highest Research activity institutions, the mean number of STEM doctorates rose to 202 (median=152). Northeastern reported a total number of 119 STEM doctorates for this period, about .15 standard deviations above the mean, ranking it 84th among all institutions in this category. This represented a 63% increase over the 2010 number of 73 doctorates, when Northeastern was ranked 95th in this category. In terms of its peer group, NU’s 119 STEM doctorates places it at about .2 standard deviations above the mean of 107 doctorates for this group.
2. Science and Engineering Research Staff: NU reported a total of 141 research staff for 2015, representing a 33% increase over 2010. That number, nevertheless, still placed NU well below the average for all universities, as well as Highest Research Activity and comparison group schools (see Figure 2). For example, in the Highest Research Activity category, the mean number of research staff was 604 with the median equal to 387, indicating that NU was near the bottom of this category.
3. Per Capita S&E Research Staff: NU reported approximately one science and engineering research staff to every five tenured/tenure track faculty member, an increase of 68% over what was reported for 2010 . NU was slightly below the mean for all institutions (see Figure 3), but well above the median to rank 98th in this category. It still lagged far behind most Highest Research Activity schools in this respect (i.e., .4 s.d. below the mean), and trailed most comparison group schools, ranking 11th in this category out of 17 schools.
4. Science and Engineering R&D Expenditures: NU more than doubled its S&E R&D Expenditures from 2010. Despite this jump, it still fell below the average expenditures of other institutions, e.g., .9 standard deviations below the mean of Highest Research Activity schools (see Figure 4). Moreover, NU fell in the bottom half of its comparison group with a ranking of 11 on this measure (out of 17 schools).
5. Per Capita S&E R&D Expenditures: Similar to actual S&E R&D expenditures, NU improved considerably over 2010 in this category, seeing a jump in $89,000 per faculty member (see Figure 5). Relatively speaking, NU ranked 116th out of all 276 schools in this category, compared to a rank of 137 in 2010. NU ranked 12th in this category among its comparison school group peers.