Ranking Community Colleges

1
Ranking Community Colleges

• Kevin Carey
• SHEEO / NCES Data Conference
• April 16, 2008

2
The Community College Survey of Student Engagement

• Very similar to NSSE, mostly identical questions.
  All results public.
• Over 600 institutions over the last five years,
  more than half all CCs.
• Results tabulated into five benchmarks --
  Active and Collaborative Learning, Student
  Effort, Academic Challenge, Student-Faculty
  Interaction, Support for Learners.

3
CCSSE Validation Study

• Released December 2006 Kay M. McClenney and C.
  Nathan Marti, "Exploring Relationships Between
  Student Engagement and Student Outcomes in
  Community Colleges Report on Validation
  Research,"
• Based on three separate studies of relationship
  between CCSSE benchmarks (and other constructs)
  and measures like GPA, progression, and
  attainment.
• Findings overall consistent, significant link
  between CCSSE and outcomes, but variable between
  measures and outcomes.

4
Community College Rankings

• Public data, large number of participating CCs,
  validation study, create possibility for CC
  rankings based on valid, outcomes-linked measures
  of academic quality.
• Addition of GRS graduation rates. (Bailey,
  Crosta, Jenkins 2007) found that while GRS
  graduation rate data for CCs flawed measure of
  absolute success, has utility for comparative
  (rankings) purposes.

5
Empirical Weighting

• Most rankings weightings are either arbitrary
  (U.S. News) or equal (Washington Monthly).
  Significant source of valid criticism.
• Validation study creates possibility to move
  toward empirical weighting.
• Simple method Based onTable 33 of the Validation
  study, "Bivariate Correlations between Outcomes
  Measures and CCSSE Constructs," Summed the number
  of time each of the five benchmarks showed a
  statistically significant correlation, giving
  extra weight based on the level of statistical
  significance--3 points for plt.001, 2 points for
  plt.01, and 1 point for plt.05.

6
Weighting Results

• 15 percent for GRS graduation rates (arbitrary)
• Split up remaining 85 percent as followsActive
  and Collaborative Learning .289Student Effort
  .119Academic Challenge .187Student-Faculty
  Interaction .136Support for Learners .119
• Consistent with validation study conclusions
  Active and Collaborative Learning was perhaps
  the most consistent predictor of student success
  across studies and across measures...The only
  other benchmark that exhibited this consistent
  pattern of positive correlations was Academic
  Challenge." (page 85)

7
The Results

8
Implications

• Good teaching doesnt necessarily involve vast
  amounts of money per-student spending at Top 30
  virtually the same as average for all CCs, far
  less than 4-yrs.
• CCs can excel even in difficult environments
  high-poverty, part-time.
• Size matters
• Excellence in higher education doesnt take
  centuries. Most CCs relatively young, some
  younger. South Texas College (1990s), Cascadia
  (2000s).

9
Cascadia Community College

• 2 Overall, 1 on most important indicator
  Active and Collaborative Learning (two separate
  years)
• Opened its doors less than seven years ago.
• Large differences from typical CCSE results
  particularly AC, group projects, etc.
• Purposeful designed with these ideas and
  principles in mind.

10
Quality Matters Most for Students Who Arrive with
the Least