A. A Note on Programmatic Wording
For the sake of consistency, this report will use the term "counselor" to refer to workforce development professionals who offer one-on-one case management, guidance, career counseling, and referrals to participants. Some agencies very deliberately choose terms such as "employment specialist", "coach", or "navigator" to speak to the philosophy behind their services. Our uniform use of the term "counselor" is not intended to discount the distinction between these terms; rather, we hope it can stand in as a general term that encompasses the wide range of activities and approaches taken by professionals in the field.
Additionally, this report will use the term "participant" to refer to people served by workforce development programs. Some agencies use terms like "client" or "customer".
B. Qualitative Methodology
Program manager interviews
We requested interviews with a targeted list of 25 program managers, chosen for their regional and organizational diversity. Twenty managers (80%) agreed to participate. We interviewed managers at eight of the state's 16 workforce service areas and at ten independent workforce agencies around the state which have contracts with DEED. In addition, we spoke with one manager within the Workforce Center System and one manager in Adult Basic Education.
Eleven program managers we interviewed are located in the Minneapolis-St. Paul metro area; three are located in southern Minnesota, one is located in central Minnesota, and five are located in northern Minnesota. We recorded, transcribed, and qualitatively coded all interviews, which ranged in length from one to two hours. We removed filler words (such as 'um' and 'you know') from quotes for readability.
We designed the survey with an informal focus group of state staff, including former counselors. We drew our survey population mainly from staff assigned as the primary case manager in DEED's participant tracking system Workforce One for participants in the Dislocated Worker, WIA Adult, or Adult Workforce Development programs, or the workforce development programs funded by a direct legislative appropriation (Twin Cities RISE! and the Opportunity Industrialization Centers).
Of all 353 counselors across the state invited to participate, 137 (39%) completed the 15-minute telephone survey, which was conducted by our survey vendor Issues and Answers. The regional breakdown of responses is: 45% Minneapolis-St. Paul metro area, 26% northern Minnesota, 20% southern Minnesota, and 9% central Minnesota. Issues and Answers transcribed all responses, and we qualitatively coded open-ended responses.
C. About the Data
The figures in this report draw from DEED's administrative data on all Dislocated Worker and Adult program participants who completed their program enrollment between 2007 and 2013. We include so many years in our analysis to have a sufficient number of participants in each of the race and class groups to show program activities and employment outcomes. Program data are entered by counselors into the case management system Workforce One and data on employee earnings and work hours are entered by employers into UI's Wage Detail.
Throughout this report, Dislocated Worker refers to both the federally- and state-funded program, as well as to the federal Trade Adjustment Assistance program integrated into Dislocated Worker. Adult programs refer to the federal WIA Adult program and the state Adult Workforce Development competitive grant program.
Wage Detail covers an estimated 97% of all Minnesota employment. It does not include work in other states, self-employment, employment purely on a commission basis, or informal employment. See 'About QCEW' for more information on this data.
D. Data Definitions
Definition of class
We define two classes based on whether the participant did or did not earn enough to meet their family's cost of living prior to program participation. Our use of the labels 'financially secure' and 'financially struggling' for these two classes is intended to be intuitive shorthand. DEED's Labor Market Information Office calculates the cost of a basic needs standard of living based on county of residence, number of children, whether the person is single or partnered, and whether the person is over age 50.
Because we do not know the earnings of the partner, we assume that partnered participants must earn 66% of the family's cost of living (single participants must earn 100% of the family's cost of living). Why 66% and not 50%? The average Minnesotan family consists of one full-time and one part-time worker, and we assume that the participant in a workforce development program is the main earner.
While the concept of class draws on factors other than income (commonly including educational background, occupation, household wealth, and even parents' income), we chose this definition for its workability and meaningfulness. Further research could develop a more nuanced definition of class that includes other factors captured in our administrative data.
Definition of race
Program case managers throughout the state ask Dislocated Worker and Adult program participants to self-identify with one or more of the following racial groups: American Indian or Alaska Native, Asian, Black or African American, Pacific Islander, and White. Participants are asked separately whether they identify as Hispanic or Latino. This mirrors the US Census Bureau practice of terming an Hispanic origin an ethnicity rather than a race. In this report, we will use the term "race" to refer to both race and ethnicity.
DEED does not collect more detailed race and ethnicity information than this, and for the period of this study did not collect immigration status. Therefore, each racial group contains a meaningful amount of diversity that we are not capturing.
Because of small population sizes, this report combines the Asian and Pacific Islander racial groups into one (very few participants identify as Pacific Islander) and does not show statistics on participants identifying with two or more races. In combining race and ethnicity into one demographic, we report all participants identifying as Hispanic or Latino as such, making the other racial groups non-Hispanic or Latino.
Definition of Employment
We define employment status in reference to the total hours worked in the reference year, across all four quarters and all employers. The year prior to participation means the third through sixth quarters prior to the quarter of the enrollment date. We look at least six months prior to participation to better capture the 'status quo' earnings and to avoid the loss of employment that drives many participants into the program. The year after exit means the first through fourth quarters after the quarter of the participant's program exit date.
'Full-time' means employment with yearly work hours equivalent to at least 35 hours per week for 52 weeks. Two or more part-time or temporary jobs may equal full-time work under this definition. 'Part-time' means any employment not meeting this full-time threshold, while 'no work' means the participant had no work hours in the reference year.
We exclude errors in the work hours data where work hours are recorded as zero but wages earned are greater than zero. This data cleaning accounts for the different sample sizes associated with each figure (noted in parentheses below each label).
E. About the Wage Boxplots
The boxplots in Figures 9 through 16 show not just the median earnings but also the wider distribution of earnings.
The green box shows the 25th through 75th percentiles. Half earn an amount within this range. The dark line in the middle of the green box shows the 50th percentile (or median). Half earn more than the median, and half earn less.
The dark horizontal lines extending from the green boxes shows the distribution of earnings within 1.5 times the difference between the 75th and 25th percentiles. Outliers beyond this range are not shown in these graphs.
Earnings are inflation-adjusted to 2014 dollars.
F. Methodology Behind Occupations in Demand
DEED's Labor Market Information Office calculates occupations currently in demand, for each planning region in Minnesota, based on job vacancy rates, total employment rates, and layoff rates. These occupations are classified according to the Standard Occupational Classification (SOC) system, the same system that counselors use to classify the occupation associated with their participant's training program. Although a single training program could be helpful in gaining employment in multiple occupations, counselors would choose the occupation desired by their participant.
Coding occupations is not always intuitive, and keyword searches are not always helpful. While an accountant would be classified in the Auditors and Accountants occupation, a forklift driver is classified in the Industrial Truck and Tractor Operators occupation. A previous analysis found that only about 64% of occupational codes entered into Workforce One are accurate.
For this reason, these figures are likely under-reporting the rate the training in an OID. Data on occupations in demand are available starting in 2008, so these figures only show participants enrolled in or after 2008.