Table of Contents:

“Mini” Degree

The “mini” master’s degree is a fast-track format, concentrating what would be one or two years of content into around 40 hours of instruction. Mini degrees are typically positioned as offering more and deeper content than a micro-credential, but less rigorous content than a traditional master’s degree, ’[he material in a “mini” degree will more likely be delivered by the same faculty delivering the material in the traditional “in person” degrees.

Universities may also be willing to customize the material in their analytics and data science courses and teach at your location (or provide customized online instruction). For example, our university offers collaborative corporate partners the opportunity to have select mini degrees (and some traditional degrees) taught onsite. One cadence example is where the class of 20-ish employees meet for four hours, two days a week for 12-18 months to complete the degree.

The primary advantage of collaborating with a university to customize the delivery of on-site degrees is that the faculty member can integrate the organization's proprietary data into the exercises and examples - the data does not leave their site and is only accessed by their employees.

In addition, the “class” can work on “real” organizational projects as part of the course material. Finally, since the courses and faculty office hours are on-site, consistent attendance is higher and students (employees) can work in study groups more conveniently. Faculty would be required to sign non-disclosure agreements to use proprietary data.

Certification

A certification is different from a certificate. Again, we know ... it’s complicated. The goal of a certification is to validate a participant’s competency or skill through a standardized assessment - much like the Bar exam for lawyers or the CPA exam for accountants. Certifications are often managed by an accrediting body and are less likely to be offered by a university.

While there is no official accrediting body for analytics and data science, the

Certified Analytics Professional (CAP) Certification8 offered by the not-for-profit organization INFORMS - Institute for Operations Research and the Management Sciences - has gained recognition as certifying an individual’s baseline knowledge of analytical concepts. The CAP requires a passing standardized exam covering seven topics (see Table 6.1).

Here is a sample question for the CAP exam (taken from the publicly available CAP Handbook):

A clothing company wants to use analytics to decide which customers to send a promotional catalogue in order to attain a targeted response rate.

Which of the following techniques would be the MOST appropriate to use for making this decision?

a. Integer programming

b. Logistic regression

c. Analysis of variance

d. Linear regression

The answer is “b”, logistic regression. This is true because the objective is a predictive model with a binary response variable (the targeted customer will either respond or not respond).

There are two important points about the CAP certification for analytics managers considering this option for their employees. The first point is that preparation for the certification is not standardized; individuals can be self-taught through the materials provided on the CAP website or they can take some form of a preparation course.

Table 6.1 Certified Analytics Professional Certification Exam Topics

Topic

Approximate Percent of Material Covered (%)

Business Problem Framing

(12-18)

Analytics Problem Framing

(14-20)

Data

(18-26)

Methodology Approach

(12-18)

Model Building

(13-19)

Model Deployment

(7-11)

Model Life cycle Management

(4-8)

Some universities will offer the CAP as the "final exam" to their mini-degree, their capstone course, or their certificate program. Organizations should consider opportunities for synergies between a university's continuing education options and the CAP for their employees.

The second is that the CAP does not integrate or assess any programming skills - only conceptual understanding. However, all major software companies have “stack- able” certifications that can be paired with the CAP. The most well-established and comprehensive programming certifications are offered by the SAS Institute9 — who also participated in the development of the CAP.

With such a wide range of options (and providers), why would an organization partner with a university for continuing education in analytics and “upskilling” of the current employees? From our experience, there are three reasons.

  • 1. Reputation and credibility. Specifically, without a formal accreditation agency, there is no standardization or validation of skills in analytics and data science. A university’s accreditation and reputation serve as a proxy in the marketplace. However, as highlighted in previous chapters, where a program is housed (e.g., business school versus college of computing) will influence the content and strengths of the graduates.
  • 2. Well-aligned curriculum. While selecting “ad hoc” continuing education options from across multiple providers may allow individuals and organizations to “surgically” select specific courses to address very specific skills gaps, a wholistic curriculum including stackable badges, micro-credentials, and mini-degrees has an advantage over disparate courses because of the continuity of faculty instructors and the integration of relevant (and frequently customized) examples and applications across the curriculum continuum.
  • 3- Integration with research. Where organizations are also engaged with a university through a data science research lab, there are synergistic opportunities to have the faculty leading the research to help “upskill” the employees to inte- grate/deploy research products. These research faculty can also share research and “thought leadership” in data science more broadly across the organization.
 
Source
< Prev   CONTENTS   Source   Next >