Methodology: Higher Education-Workforce Pipeline IPEDS CIP SOC crosswalk

Original Article

To Be Completed!

** If you are using Microsoft Internet Explorer or Google Chrom browser, you would not be able to read the formulas in this article. These formulas were written in MathML, a W3C standard, and can be viewed in FireFox.

This methodology is much improved over my previous methodology.

In essence, the number of graduates can potentially be hired for a program with CIP code C is:
$$\text{Potentially\_Hired}\left(C\right)=(G_c+\epsilon )\sum _{S_c}{O}_{s_c}\left[\frac{1}{\sum _{C_{S_c}}(G_{C_{S_c}}+\epsilon )}\right]$$
Where $G_x$ is the number of graduates for CIP x, $O_z$ is the number of job opening for a given SOC z, $S_x$ is the SOC S related to a CIP x, $C_z$ is the CIP C related to an SOC z, and $\epsilon$ is a very small number, which play important role when $G_x$ is zero.

Following are drafts and is to be ignored for Now!
Number of graduates for CIP c:

Number of job open for SOC s:
$O_s$
SOC related to a CIP c:
$S_c$
CIP related to an SOC s:
$C_s$
$G_{S_{C_s}}$
$\sum _{-1}^{\mathrm{+1}}$
$\sqrt{{(x_1-x_2)}^2+{(y_1-y_2)}^2}$

Evaluate Nebraska's Education-Workforce Pipeline - CIP SOC Crosswalk

The methodology employed by this article is a major improvement over the previous model in that suggested increase or decrease of award production will have no side effects on related fields!

With the P20 initiatives bobbling up all over the United States, the idea of education pipeline has extended from education to the workforce supply. The idea of linking higher education production to economic prosperity has found its way into legislatures.

The idea of aligning higher education production with the workforce demand isn't new. A crosswalk system that linking the field of study to appropriate occupation was last updated in March 2012 by National Center for Education Statistics (NCES) and the Bureau of Labor Statistics (BLS). The crosswalk, also known as CIP to SOC crosswalk, had been used in various workforce supply studies. However, because the complexity of the crosswalk, most of the studies limited their use to few focused fields of interest.

While the CIP to SOC crosswalk is a useful tool, limitations must be observed. The guiding principle behind the development summarized it well:

• “A CIP-SOC relationship must indicate a “direct” relationship, that is, programs in the CIP category are preparation directly for entry into and performance in jobs in the SOC category. The programs satisfy requirements for entry and/or prepare individuals to meet licensure or certification requirements to work in the occupation.”

An example can provide some clarity to the principle. In the nursing field, the registered nurse program (CIP 513801 Associate degree) is crosswalked to the registered nurse (RN) occupation (SOC 291141 Registered Nurse) and not the Licensed Practical Nurse (LPN) occupation (SOC 292061) even though the registered nurse can certainly worked in the LPN occupation. When interpret the result of this work, please keep this in mind, the result is based on 'appropriate' mapping and not the 'all possible mapping'. In the case of Registered Nurse, State of Nebraska can decide to produce more Registered Nurses to fill the LPN occupation. But that probably isn't the most efficient approach.

This article presents a general approach that is applied to all field of studies. The approach is applied to the state of Nebraska data set but can be easily applied to other states.

The data employed are the 2012-22 long term occupation projection data from Nebraska's Department of Labor and the 2013 degree awarded data from IPEDS' (Integrated Postsecondary Education Data System) completion survey.

While the methodology possessed some very desirable characteristics, like all researches, the data employed played vital role in the outcome. During the development, the job opening advertise data available from Nebraska Department of Labor's website were examined. The data for occupation like 'bus driver, School or Special Client' seems to be OK. The data for hot occupations like nurse or computer related positions post some challenges. For one, there were a lot of entries posted by staffing companies or recruiters and, by reading through some of the entries, some of them are simply phantom entries that mimic the real positions. For the nursing jobs, multi-leveled job listings are common. For example, a position can be advertized as RN/LPN, which begged the question of which category this position is counted under. With these observations, the result presented is based on Labor's projection data which can easily be replaced with more accurate data if such data is available.

The methodology employed by this article is different from that of our previous articles. This new approach eliminated couple of the mentioned limitations of previous methodology. Noticeably, when increase or reduce the number of graduates for a given field following the suggestion of this article would not interfere with the number suggested by this article for other fields. This is a major improvement over the previous model in which additional analysis is needed to make sure no side effect occurred in other related fields.

At the heart of this new methodology is the assumption of equal chance of employment, which means that all graduates that can be walked to an occupation, they all have the same chance of getting hired. If supported by real-life data, this factor can be modified to provide better result.

The top 10 education programs that could be targeted to produce more graduates are:

CIP	Title	Dgr Level	Job Open	UnFilled
490205	Truck and Bus Driver/Commercial Vehicle Operator and Instructor.	LessAssct	1,024	(953)
513999	Practical Nursing, Vocational Nursing and Nursing Assistants, Other.	LessAssct	468	(456)
513801	Registered Nursing/Registered Nurse.	Associate	702	(292)
520301	Accounting.	Bachelor	467	(156)
110101	Computer and Information Sciences, General.	Bachelor	259	(113)
520801	Finance, General.	Bachelor	300	(55)
460302	Electrician.	LessAssct	154	(96)
120401	Cosmetology/Cosmetologist, General.	LessAssct	126	(84)
110701	Computer Science.	Bachelor	121	(75)
480501	Machine Tool Technology/Machinist.	LessAssct	144	(65)
460303	Lineworker.	LessAssct	71	(56)

The top 10 eduction programs that could consider reducing graduates are:

CIP	Title	Dgr Level	Job Open	Over Supplied
540101	History, General.	Bachelor	22	163
513801	Registered Nursing/Registered Nurse.	Master	11	180
230101	English Language and Literature, General.	Bachelor	27	198
510912	Physician Assistant.	Master	43	245
450101	Social Sciences, General.	Bachelor	41	255
130301	Curriculum and Instruction.	Master	13	398
131202	Elementary Education and Teaching.	Bachelor	276	404
260101	Biology/Biological Sciences, General.	Bachelor	70	436
520201	Business Administration and Management, General.	Bachelor	368	588
520101	Business/Commerce, General.	Bachelor	386	618

college workforce supply/demand - methodology

Interact with the author through EduStats - We value your input.

In conclusion, we have proofed that the methodology we used does allow us to identify both the definitely over supply and under supply academic programs. To put this in plain English, it means that if we labeled an academic program over supply, the program is definitely over supply - no matter how you simulate the hiring based on the crosswalks. The same is true for the under supply academic programs. Our designation, however, does not suggest to the policy maker to increase the number of graduates of all under supply academic programs to the amount of shortages since some of the programs are related.

This article is a technical note that described the research methodology employed in the analysis of my previous article titled 'workforce supply/demand higher education - Nebraska'.

Updated June 8, 2015: An improved approach provides better directions for program management agency or policy maker.

The whole idea behind the analysis is what we called the worst case scenario analysis, which is commonly used in simplify a complicate problem so that some guidance for further analysis can be devised. A common result of such analysis is the lower bound or upper bound of a variable of interest.

The problem of the supply and demand interaction between higher education and workforce is a complex one. It is not mathematically challenge but nevertheless a complex one. The goal is aimed to understand how college/higher education graduates are fed into the workforce.

The ground work to the problem was laid years ago. Researcher and workforce development workers, after years of study, have documented the field and the level of knowledge needed for each occupation. In the same time, crosswalk tables were created that linked each academic program to the related occupation. In the crosswalk framework, the occupation is classified by the so called 'Standard Occupational Code' (SOC) and the academic program is classified by the 'Classification of Instructional Program' (CIP) code and the degree level awarded.

The complexity of the problem rooted at the fact that the crosswalks between the academic program and the occupation is not a single one to one map. As we can all image that graduates from one academic program can be fed into more than one occupation. The reverse of that is alos true: An occupation can accept graduates from more than one academic programs. It is this complexity that have limited most analysis to a smaller scale. For example, a Texas supply and demand study only focused on few big categories and the 'The Occupational Supply Demand System' website only provides tools for navigating between academic programs and occupations.

An additional complexity is also exist that the education classification system used to classify the occupation is not directly compatible with that used to classify the academic program either. For our study, since we are only interested in college educated graduates, all jobs classified with less than college degree requirement are discarded based on the idea that, for most cases, it wouldn't worth the investment for a college graduates to take that kind of jobs. In order to address the incompatibility between the two education classification system, a new education classification is improvised which allows the creation of a one to one map, in the mathematical sense, from both the academic and the occupational system to the new classification. The mapping is outlined below:

Academic	Ours	Occupational
Less Than 1 Year Awards	Less than 2 year certificates	Postsecondary vocational training
Between 1 and 2 Years Awards	Less than 2 year certificates	Postsecondary vocational training
Associates Degrees	Associate (+Less than 4 year)	Associate degree
Between 2 and 4 Years Awards	Associate (+Less than 4 year)	Associate degree
Bachelors Degrees	Bachelor (+Certificates)	Bachelor's or higher degree, plus work experience
Bachelors Degrees	Bachelor (+Certificates)	Bachelor's degree
Post-Bachelors Certificates	Bachelor (+Certificates)	Bachelor's degree
Post-Bachelors Certificates	Bachelor (+Certificates)	Bachelor's or higher degree, plus work experience
Masters Degrees	Master (+Certificates)	Master's degree
Post-Masters Certificates	Master (+Certificates)	Master's degree
Doctorate Degrees	Doctor	Doctoral degree
First Professional Degrees	First Professional (+Certificates)	First professional degree
Post-First Professional Certificates	First Professional (+Certificates)	First professional degree
Doctor's degree - research/scholarship	Doctor	Doctoral degree
Doctor's degree - professional practice	First Professional (+Certificates)	First professional degree
Doctor's degree - Other	Doctor	Doctoral degree

In theory, with enough computing power, we can simulate all possible scenarios and draw conclusions from the all possible assumptions. However, that kind of approach could easily bury the intuitive common sense and lost the researcher in the forest of data.

Since our goal is to identify the definitely over-supply and the definitely under-supply academic programs, we chose to use the worse case scenario analysis.

Since the process of establishing the lower bound for over-supply is much straightforward, we will describe it first.

By definition, an academic program is over-supply if there are fewer jobs appropriate for the program than the number of graduates from that academic program. By assuming that all appropriate jobs openings for an academic program are available to graduates from that academic program, we can calculate the number of graduates that could not find a job opening by subtracting the number of job openings from the number of graduates. If the result of the calculation is a postive number, we know the number of graduates would not be able to find the appropriate jobs. In reality, some of the appropriate job openings could be filled with graduates from other academic program and, hence, reduce the number of openings available to our focus academic program. However, the program we identified as over supply will still be over supplying its graduates, just to a bigger amount. The result we arrived is, therefore, a lower bound and the academic program we identified is, therefore, a definitely over supply program.

The process of producing the lower bound for the under-supply academic program is a bit more complicated. The idea begins with that if an academic program produced fewer graduates than what the industry can absorb, then that academic program is a under-supply program. The number of shortage in supply or the number of job openings to fill can be calculated by subtracting the number of graduates from the number of those job openings. As a first attempt, we could proceed the calculation using job openings from all appropriate occupations for a given academic program. However, in reality, some of the appropriate job openings could be filled with graduates from other academic programs. The number we arrived previous is, therefore, an over estimate of the shortage problem. The shortage may not even exist if all those appropriate jobs can be filled with graduates from other academic programs.

To resolve this problem, we begin our first step by identifying all the rival academic programs of our focus academic program. By definition, the rival academic programs are programs that could supply graduates to any of the appropriate job opening of our focus academic program. Once we identified all the rival academic program, we can calculated the total rival graduates by adding all the graduates from these rival academic programs. We, now, recalculate the shortage or the number of job opening to fill by subtracting both the number of graduates of our focus program and the rival graduates from the appropriate job openings of our focus program.In reality, not all rival graduates can fill those appropriate job openings. In that case, the number of job openings to be filled will be larger. The result we arrived is, therefore, an absolute minimum of the number of job openings need to be filled. We, therefore, termed that academic program a definitely under-supply program.

In conclusion, we have proofed that the methodology we used does allow us to identify both the definitely over supply and under supply academic programs. To put this in plain English, it means that if we labeled an academic program over supply, the program is definitely over supply - no matter how you simulate the hiring based on the crosswalks. The same is true for the under supply academic programs. Our designation, however, does not suggest to the policy maker to increase the number of graduates of all under supply academic programs to the amount of shortages since some of the programs are related. Increase the graduates in one academic program may reduce the number of job openings of a rival academic program and move that program off the under supply list.