USDA Forest Service employee demographic data for diversity and inclusion analysis, 1995-2017

Metadata:

Data_Quality_Information:

Attribute_Accuracy:

Attribute_Accuracy_Report:

Measurement Quality Objectives (MQOs) were not defined for this study. All known information regarding these data has been provided in this document.

Logical_Consistency_Report:

All data were checked for both completeness and logical consistency prior to publication. Specifically, we checked whether length of service matched separation year minus appointment year. There were negative lengths of service in the raw data and inaccurate lengths of service for many people who served for multiple years; see Process Steps below for more information.

Completeness_Report:

Appointments (new hires) were missing for two fiscal years (where fiscal year [FY] is from October 1 to September 30 of the following calendar year): 2001 and 2010. Separations were missing for fiscal year 2009. See process steps below for more information.

Note: data include values of "Null", "NA", and blank cells. These data are as provided by HRM records.

Lineage:

Source_Information:

Source_Citation:

Citation_Information:

Originator: USDA Forest Service, Human Resources Management

Publication_Date: Unknown

Title:

USDA Forest Service human resources data

Geospatial_Data_Presentation_Form: tabular digital data

Publication_Information:

Publisher: USDA Forest Service, Human Resources Management

Type_of_Source_Media: personal communication

Source_Time_Period_of_Content:

Time_Period_Information:

Range_of_Dates/Times:

Beginning_Date: 19951001

Ending_Date: 20170930

Source_Currentness_Reference:

Publication Date

Source_Citation_Abbreviation:

HRM

Source_Contribution:

Individual-level anonymous records for each employee each year from fiscal year 1995-2017 which include series, grade, duty station, veteran status, and other demographic data.

Process_Step:

Process_Description:

DATA SOURCE

Data were supplied by USDA Forest Service, Human Resources Management (HRM). Data were pulled from the National Finance Center's records and included individual-level anonymous records for each employee each fiscal year (from October 1 to September 30 of the following calendar year) from 1995-2017 because 1995 was the first year digital data were deemed reliable by HRM staff. Data variables included series, grade, duty station, veteran status, and other demographic data. All data were assigned by HRM staff with the exception of race/ethnicity and gender which are self-reported by employees and are able to be changed.


DATA PROCESSING - Additional Variables

Annual data were collapsed into a flat file containing one record per employee per fiscal year. There were several issues with data consistency, so we imputed variables where necessary. These steps are described in the variable descriptions below, but we also provide a higher-level description here.

Additional variables were calculated, such as length of service (FiscalYearsActive) which was obtained by subtracting the first fiscal year in which an employee was active with the Forest Service from the last fiscal year they appear in the data. The variable called “Advance” shows the change in grades over an employee’s career was calculated by subtracting the lowest grade an employee had from their highest grade.

The variable entitled “Min_FY2” was created due to missing data in the Min_FY variable (there are two years where there are no data), we imputed data from appointment year to fill in the missing years. The query was thus: if a person has a Min_FY for one of the years that precedes the missing Min_FY, we will impute this variable with the appointment year.

The variable entitled “Min_FY3” was created because as we completed the transformation steps to create the variable entitled Min_FY2, we observed that the app_year variable was not consistent within employees. In other words, the same employee id was associated with multiple appointment years in several cases. To mitigate this issue, the variable called Min_FY3 was created which uses the minimum value of appointment year for a given employee’s first year of employment.

The variable entitled “Max_FY2” was created due to analogous issues as uncovered with the Min_FY variable. There was one fiscal year with missing data in the Max_FY column. The Max_FY2 variable was created to impute the missing years with information from the Separation Year variable.

The RNOERI_Indigenous categorical variable was created to provide information on employees’ self-reported race/ethnicity classification, while obfuscating the RNO_ERI categories with the smallest sample size per cell by collapsing Native Hawaiian/Pacific Island and American Indian/Alaskan Native into the “Indigenous” category.


DATA PROCESSING - Job Fields

It was important to analyze our data by series in order to investigate if hiring and retention rates, types of diversity, or other issues varied by type of job or job family. Because there are so many series (there are hundreds of them), and because each employee had a series each year, and because some series were common in the Forest Service while others were rare, we needed to find a way to meaningfully reduce the data to use effectively in our analyses. We did several things to achieve this end.

We cleaned the series, and recoded employees whose series type had changed to the new series. For example, in 2000 there was a significant revision to series for classifying human resource staff. We used the current series codes and also used them for any old human resources series code that a particular employee may have had. But if an employee had an earlier series code that was not a human resources series (e.g., if an employee started as a seasonal ranger and worked in that job for five years before being hired in HRM) that code was retained for those particular years.

Another data reduction step we took was to categorize each position as: 1) Professional, 2) Technician/Assistant, 3) Trades, or 4) Trainee. Technician/Assistant, Trades, and Trainee are all designated in the Handbook. The rest of the positions were categorized as Professional.

We manually coded each job into job field categories, which was a difficult task because there were such a large number of job series and sometimes multiple job descriptions. We decided to use the following categories: 1) Resource Sciences & Management, 2) Administration/Business Operations, 3) Facilities/Logistics, or 4) Outreach/Communication/Information Management. For a list of job titles and job field category assigned, see \Supplements\Series_Mapping.pdf.

To assess employees between 1995 and 2017 that were associated with the internship/student trainee program we looked at the employees’ job series. Any employee whose first job series ended in a “X99” was classified as an intern/student trainee.


There are three datasets provided within this data publication, each showing a slightly different view of the data. The first dataset provides employee-fiscal year level data (JOF_trends_data.csv). These data can be used to understand trends over time. The second dataset provides summary employee data (one row for each employee) and employee level metrics such as advancement and length of service (PAR_career_data.csv). These data are useful for doing employee-level analyses. Finally, the third dataset identifies which employees (between 1995 and 2017) began their career with the Forest Service through an internship/student trainee program (Internship_data.csv). The data can be cross-referenced with either the first two datasets to understand trends over time or employee-level trends.


REFERENCES

Westphal, Lynne M; Dockry, Michael J; Kenefic, Laura S; Sachdeva, Sonya S; Rhodeland, Amelia; Locke, Dexter H; Kern, Christel C; Huber-Stearns, Heidi R; Coughlan, Michael R. 2022. USDA Forest Service employee diversity during a period of workforce contraction. Journal of Forestry. 120(4): 434-452. https://doi.org/10.1093/jofore/fvab071 and https://research.fs.usda.gov/treesearch/63871

Sachdeva, Sonya S.; Westphal, Lynne M.; Kenefic, Laura S.; Dockry, Michael J.; Locke, Dexter H.; Fisher, Cherie L. 2023. Despite Workforce Diversity Efforts, Career Metrics Differ for Some Demographic Groups in the USDA Forest Service. Society & Natural Resources. 36(6): 680-695. https://doi.org/10.1080/08941920.2023.2183447 and https://research.fs.usda.gov/treesearch/65956

Dockry, Michael J.; Sachdeva, Sonya S.; Fisher, Cherie L.; Kenefic, Laura S.; Locke, Dexter H.; Westphal, Lynne M. 2022. Student trainee and paid internship programs have positive results but do little to influence long-term employee diversity in the USDA Forest Service. PLOS ONE. 17(11): e0277423. 17 p. https://doi.org/10.1371/journal.pone.0277423 and https://research.fs.usda.gov/treesearch/65578

Process_Date: 2022

Entity_and_Attribute_Information:

Overview_Description:

Entity_and_Attribute_Overview:

Below you will find a list and description of the files included in this data publication.

VARIABLE DESCRIPTION FILE (1)

1. \Data\_variable_descriptions.csv: Comma-separated values (CSV) file containing a list and description of variables found in all data files. (A description of these variables is also provided in the metadata below.)

Columns include:

Filename = name of data file
Variable = name of variable
Description = description of variable


DATA FILES (3)

1. \Data\Internship_data.csv: CSV file containing a list of which employees began their career with the Forest Service through an internship/student trainee program.

Variables are listed below:

count = Row count

hidden_ids = Nominal variable showing obscured identifying number for employee.

Pathways = The Pathways program was implemented between 2012 and 2013. For our data, we assumed any intern beginning after FY 2012 was part of the Pathways program so they were flagged as “Pathways”. Fundamentally, this variable just marks post FY 2012 start date.

EverPerms = This variable indicates whether the employee became a permanent FS employee at some point over the course of their career. A value of “1” indicates that the employee did convert to a permanent position at some point.



2. \Data\JOF_trends_data.csv: CSV file containing both permanent and temporary Forest Service employee-fiscal year level demographic and career data related to Westphal et al. (2022; https://doi.org/10.1093/jofore/fvab071).

Variables are listed below:

hidden_ids = Nominal variable showing obscured identifying number for employee.

FY = Variable showing fiscal year of the record (fiscal year = period beginning October 1 and ending September 30 of the following year).

grade_group = Categorical variable showing classification of employee’s grade in a given FY into 3 groups. Possible values include 1 to 8, 9 to 12, 13 to 15, and 0 is used to denote volunteers.

job_field = Categorical variable showing classification of occupational series into reduced number of categories. Possible values include Admin, Comms, Facilities, Forest resources or Null. Admin=Administration, Comms=Communications, Facilities, Forest resources. (Note: blanks and Null values are directly from raw HRM data)

separation_type = Categorical variable showing classification of SEP_NATURE_OF_ACTION (original variable from the HRM data) into a reduced number of categories. Possible values include: NA (indicating the record was still active with the agency as of the time data were pulled), Voluntary Retirement, Voluntary, D&D (Death & Disability), Reduction, Involuntary, and Term.

Gender = Categorical variable showing each record’s gender, and gender only refers to male or female for the purpose of this dataset.

sep_year = Variable showing separation year according to HRM records.

app_year = Variable showing year of appointment according to HRM records.

FY_active = Numeric derived variable showing the number of years (or rows) associated with a hidden_ids: the number of fiscal years that the employee is active in the dataset.

Min_FY = Variable showing the first fiscal year that the employee appears in the data.

Max_FY = Variable showing the last fiscal year that the employee appears in the data.

Min_FY2 = Variable showing a revised minimum fiscal year for employees. Due to missing data in the Min_FY variable (there are two years where there are no data), we imputed data from appointment year to fill in the missing years. The query was thus: if a person has a Min_FY for one of the years that precedes the missing Min_FY, we will impute this variable with the appointment year.

Max_FY2 = Variable showing a revised maximum fiscal year for employees. Analogous to the Min_FY issue, there was one year that we had missing data in the Max_FY variable. This variable was created to impute the missing years with information from the Separation Year variable.

Min_FY3 = Updated variable to show first year of employment in all analyses. As we completed the transformation steps to create the variable entitled Min_FY2, we observed that the app_year variable was not consistent within employees. In other words, the same employee id was associated with multiple appointment years in several cases. To mitigate this issue, the variable called Min_FY3 was created which uses the minimum value of appointment year for a given employee’s first year of employment.

Grade = Categorical variable with two-digit code that refers to the employee's pay grade for the record.

deputy_area_nfs = Categorical variable that divides deputy areas into either National Forest System (NFS) or Non-NFS, which includes all other deputy areas: Research & Development, Business Operations, Office of the Chief, and State and Private Forestry.

RNOERI_Indigenous = Categorical variable that provides information on employees’ self-reported race/ethnicity classification. Possible values include White, Afr Amer (African American), Hispanic, Asian, Indigenous (including Native Hawaiian/Pacific Islander and American Indian/Alaskan Native), Two or More, and Unknown.



3. \Data\PAR_career_data.csv: CSV file containing Forest Service employee demographic and career data, for those employees who at some point in their career became permanent or were never volunteers, related to Sachdeva et al. (2023).

Variables are listed below:

hidden_ids = Nominal variable showing obscured identifying number for employee.

grade_group = Categorical variable showing classification of employee’s last grade into 3 groups. Possible values include 1 to 8, 9 to 12, 13 to 15, and 0 is used to denote volunteers.

job_field = Categorical variable showing classification of occupational series into reduced number of categories. Possible values include Admin, Comms, Facilities, Forest resources or Null. Admin=Administration, Comms=Communications, Facilities, Forest resources.

separation_type = Categorical variable showing classification of SEP_NATURE_OF_ACTION (original variable from the HRM data) into reduced number of categories. Possible values include: NA (indicating the record was still active with the agency as of the time data was pulled)le, Voluntary, D&D (Death & Disability), Reduction, Involuntary, Term.

Gender = Categorical variable showing employee’s gender in the last record, and gender only refers to male or female for the purpose of this dataset.

Max_FY2 = Variable indicating latest year of employment for employee.

Min_FY3 = Variable indicating first year of employment for employee.

appt_type = Categorical variable showing employees’ appointment type for the last record. Possible values include Permanent, Temporary, Term.

Grade = Categorical variable with two-digit code that refers to the employee's pay grade for the last record.

Advance = Numeric variable showing the number of grades increased over employee’s tenure.

Initial.Grade = Categorical variable showing employee’s pay grade for their first record.

FiscalYearsActive = Numeric derived variable showing the number of years (or rows) associated with an hidden_ID: the number of fiscal years that the employee is active in the dataset.

BIPOC = Categorical binary variable that divides employee’s self-reported race information into either White or BIPOC (collapses across Black, Hispanic, Native Hawaiian/Pacific Islander, Asian, American Indian/Alaskan Native and Two or more) .

deputy_area_nfs = Categorical variable that divides deputy areas into either National Forest System (NFS) or Non-NFS, which includes all other deputy areas: Research & Development, Business Operations, Office of the Chief, and State and Private Forestry.

RNOERI_Indigenous = Categorical variable that provides information on employees’ self-reported race/ethnicity classification. Possible values include White, Afr Amer (African American), Hispanic, Asian, Indigenous (including Native Hawaiian/Pacific Islander and American Indian/Alaskan Native), and Two or More.


SUPPLEMENTAL FILES (1)

1. \Supplements\Series_Mapping.pdf: Portable Document Format file containing a list of occupational series numbers, job titles, and general job field categories that were assigned for this study.

Columns are listed below:

OCC_SERIES = Occupational series number assigned to job title by HRM.

OCC_SERIES CLEANED * = Cleaned occupational series number, which was done when an employees' series type had changed to a new series.

Prof, Tech/Asst, Trainee, or Trades = Categorical variable showing classification of occupational series into a number of general categories. Possible values include Professional, Technician/Assistant, Trainee, or Trades.

ResSci, Admin/Bisops, Facilities, Outreach Comms = Categorical variable showing classification of occupational series into reduced number of categories (referred to as job_field in data files). Possible values include Admin, Comms, Facilities, Forest resources or Null. Admin=Administration, Comms=Communications, Facilities, Forest resources.

Title = Title of the job.

Entity_and_Attribute_Detail_Citation:

Westphal, Lynne M; Dockry, Michael J; Kenefic, Laura S; Sachdeva, Sonya S; Rhodeland, Amelia; Locke, Dexter H; Kern, Christel C; Huber-Stearns, Heidi R; Coughlan, Michael R. 2022. USDA Forest Service employee diversity during a period of workforce contraction. Journal of Forestry. 120(4): 434-452. https://doi.org/10.1093/jofore/fvab071 and https://research.fs.usda.gov/treesearch/63871

Sachdeva, Sonya S.; Westphal, Lynne M.; Kenefic, Laura S.; Dockry, Michael J.; Locke, Dexter H.; Fisher, Cherie L. 2023. Despite Workforce Diversity Efforts, Career Metrics Differ for Some Demographic Groups in the USDA Forest Service. Society & Natural Resources. 36(6): 680-695. https://doi.org/10.1080/08941920.2023.2183447 and https://research.fs.usda.gov/treesearch/65956

Dockry, Michael J.; Sachdeva, Sonya S.; Fisher, Cherie L.; Kenefic, Laura S.; Locke, Dexter H.; Westphal, Lynne M. 2022. Student trainee and paid internship programs have positive results but do little to influence long-term employee diversity in the USDA Forest Service. PLOS ONE. 17(11): e0277423. 17 p. https://doi.org/10.1371/journal.pone.0277423 and https://research.fs.usda.gov/treesearch/65578

Metadata_Reference_Information:

Metadata_Date: 20241216

Metadata_Contact:

Contact_Information:

Contact_Organization_Primary:

Contact_Organization: USDA Forest Service, Northern Research Station

Contact_Person: Sonya Sachdeva

Contact_Position: Research Social Scientist

Contact_Address:

Address_Type: mailing and physical

Address: 1033 University Pl. Ste. 360

City: Evanston

State_or_Province: IL

Postal_Code: 60201

Country: USA

Contact_Voice_Telephone: 224-999-4004

Contact_Electronic_Mail_Address: sonya.s.sachdeva@usda.gov

Contact Instructions: This contact information was current as of original publication date. For current information see Contact Us page on: https://doi.org/10.2737/RDS.

Metadata_Standard_Name: FGDC Biological Data Profile of the Content Standard for Digital Geospatial Metadata

Metadata_Standard_Version: FGDC-STD-001.1-1999