Thoughts on Variability in Graduation Rates at Engineering Colleges

As universities are celebrating commencement exercises with their graduates this month, it is an appropriate time to talk about graduation rates and congratulate engineering colleges with high graduation rates. 

One might ask: What is the range on graduation rates at U.S. engineering colleges. The answer may surprise you.  Some colleges graduate almost 100% of their students in 5 years; however, for some colleges, the graduation rate is less than 20%.  Read on to find out more about the variation in the 5-year graduation rates at U.S. engineering college and the relationship between the graduation rate and first year retention. The 5-year graduation rate and first year retention rates are both based on the entering freshman cohort of first time, full-time freshmen. 

While the average graduation rate is generally available for all universities (See College Score Card, https://collegescorecard.ed.gov/), the graduation rates for engineering colleges (within universities) typically are not available.  So it is exciting news that the American Society for Engineering Education (ASEE) has established a database in partnership with U.S. engineering colleges that includes retention and graduation rates.    I advocated for such a database at the 2006 ASEE annual conference (Veenstra & Herrin, 2006).  

In February, ASEE published distribution percentiles of the 5-year graduation rates for the 2007 freshman class (first time, full-time students)  with 56, 393 freshmen from 150 engineering colleges(“DataBytes,” 2016).   This equates to 53% of the full-time U.S. engineering freshmen in the same year (refer to Gibbons (2009) indicating a total of 106,110 engineering freshmen in 2007).

5-Year Graduation Rates Shows Variability across Engineering Colleges

 The distribution percentiles for the 5-year graduation rate shown in Figure 1 are listed in the ASEE Connections article (“DataBytes,“ 2016). 

 

Figure 1: Cumulative Distribution of the 5-Year Graduation Rate (Data Source: ASEE Connections Feb. 2016)

Figure 1: Cumulative Distribution of the 5-Year Graduation Rate (Data Source: ASEE Connections Feb. 2016)

Note that Figure 1 shows a 50% range of 34 % to 64% with a median graduation rate of 47% for the 150 engineering colleges.  The variability among the engineering colleges is significant.  The 10% percentile college has a 26% graduation rate compared to 77% for the 90th percentile college, a 51% difference.  The odds of graduating within 5 years from the college with the 77% graduation rate are 3 to 1 compared to the college with the 26 graduation rate.   Furthermore, the minimum 5-year graduation rate is 4% while the highest graduation rate is 97%.    It is worth knowing that generally, with one more year (6-years) that the variability will decrease (see Veenstra & Herrin (2006) for an example of university graduation rates.)  Some students continue in engineering and with more time,  graduate in 6 or 7 years.  The engineering colleges with low graduation rates especially should show higher 6-year graduation rates. 

The 10th percentile for the 5-year graduation rate is 25.5% (“DataBytes,” 2016).   By the definition of the 10th percentile in a sample of 150 colleges, 15 engineering colleges have reported a 5-year graduation rate less than or equal to 25.5%.  Even with the expectation of a higher 6-year graduation rate, this scenario is discouraging.  Students and their families invest heavily in the college the student chooses.  Colleges have a social responsibility to help admitted students achieve their educational goals (Veenstra, 2009).  Are these fifteen colleges demonstrating socially responsibility to their students?   

How do the Distributions of the Graduation Rate and First Year Retention Compare?

The ASEE article also listed the distribution percentiles for the first year retention (“DataBytes,” 2016).  The first year retention is the percent of incoming freshmen to enroll in the engineering college for the fall semester, 2nd year. As a comparison, Figure 2 shows the estimated normal distribution for the first year retention and 5-year graduation rates based on the percentiles.  (Using the Anderson-Darling test, the data fits a normal distribution.) 

Figure 2: Estimated Normal Distributions for 5-Year Graduation Rate and First Year Retention

Figure 2: Estimated Normal Distributions for 5-Year Graduation Rate and First Year Retention

Figure 2 shows two patterns:

1) A large gap in the median graduation rate by engineering college compared to the median first year retention.  The upper bound on the graduation rate at an engineering college is the first year retention. Could the gap be less?

2) The variability is much larger for the graduation rate than the first year retention.  How could the variation decrease and as a result, more students graduate from engineering colleges?

Some factors that influence students dropping out of engineering are academic preparation, financial aid, program quality and a sense of belonging/engagement.  For example, if a student enrolls in Calculus III and he/she did not understand Calculus I or II, he/she may fail Calculus III and transfer out into a major that requires less math.  Sometimes engineering colleges need to redesign and revise their processes that support student academic success.  More students are capable of success in engineering programs.  

The Higher the First Year Retention, the Higher the Graduation Rate

An estimate of the 5-year graduation rate based on first year retention can be predicted.   Figure 3 plots the percentiles for the 5-year engineering graduation rate versus the percentiles of the first year retention rate from the “DataBytes” article. 

Figure 3: Predicting the 5-Year Graduation Rate from the First Year Retention

Figure 3: Predicting the 5-Year Graduation Rate from the First Year Retention

Except for the minimum and maximum (end) points, the graph shows a strong linear relationship with a tread line of the percentiles:  

                              5-Year Graduation Rate = - 79.22 + 1.676 First- Year Retention

 As an example, if a college had 10% of its students dropping out of engineering in the first year (90% retention), the predicted 5-year graduation rate is 72%.   A less successful college with a loss of 1/3 of its freshman class in the first year could expect only a 33% 5-year graduation rate.  

Using this prediction formula for an individual engineering college assumes that the sample of 150 engineering colleges is representative of all engineering colleges.  The estimate using the percentiles is approximate and provides good information; obviously, if the data were available, it would be preferred to generate the prediction formula from the data values associated with each engineering college rather than the percentiles. 

Summary

ASEE’s database on engineering colleges’ statistics is a major effort in the right direction to provide information on trends in engineering education including retention and graduation rates.  I hope the database will continue to grow and support discussions on continuous improvement in the engineering education community.

Figure 1 statistics support that there is significant variation among engineering colleges.  With the data published by ASEE, an engineering college’s dean and staff can compare their college‘s graduation rate to other colleges.  50% for a 5-year graduation rate is average, while 30% or less is mediocre (80% of the colleges achieved at least 30%) and 70% is superlative (in the top-20%, congratulations!).  

The study of graduation rates is multivariate and complex; the colleges’ cultures, admission policies and student support/engagement activities all contribute to the variation.  As discussed, many colleges are quite successful in achieving high graduation rates.  Figure 3 defines the linear relationship between the 5-year graduation rate and the first year retention.  Expectations for a high graduation rates start with the freshman class. An engineering college with a 90% first year retention of its freshmen can expect a 5-year graduation rate of 72% while an engineering college with a 75% first year retention can expect a 47% graduation rate.    Investing time and resources in understanding factors that contribute to an individual engineering college’s retention and graduation rates can be very beneficial to the engineering college, resulting in higher enrollment.  A student-focused culture with a strong focus on quality and continuous improvement is important; research supports that engaged and satisfied students will tend to be more committed to their education, resulting in degree completion and higher graduation rates for the engineering college. 

Some advice to students (and their parents) in selecting an engineering college: ask the admission office for statistics on the engineering college’s first year retention and graduation rates.  Also ask what the employment rate of recent graduates is: for high performing engineering colleges, it will be between 95% and 100%. Engineering colleges with good transparency will include these statistics on their website. 

References:

DataBytes. (2016, February) In Grose, T. (Ed) ASEE Connections, Washington DC:ASEE. http://createsend.com/t/y-45B6B3EF48CE7A3C#databyte

Gibbons, M.T. (2009). Engineering by the numbers, In 2009 Profiles of Engineering and Engineering Technology Colleges (pp. 11-46). Washington DC: ASEE. https://www.asee.org/papers-and-publications/publications/college-profiles/2009-profile-engineering-statistics.pdf

Veenstra, C.P. & Herrin, G.D. (2006). An analysis of graduation rates at research universities, Paper presented at the 2006 Annual Conference & Exposition, Chicago, Illinois. https://peer.asee.org/an-analysis-of-graduation-rates-at-research-universities

Veenstra, C. (2009). Educator’s world: A strategy for improving freshman college retention, The Journal for Quality and Participation, 31(4), 19-23.  http://asq.org/pub/jqp/past/2009/january/index.html

 

Cindy Veenstra, PhD, ASQ Fellow  

Dr. Veenstra can be contacted at cindy@veenstraconsulting.com