Staffing, Training and Development
This domain follows people into and through the organization: recruitment, selection, socialization, training, and development. Items test definitions, the validity logic behind selection tools, and the sequence of training design.
Recruitment
Recruitment builds a qualified applicant pool. Internal sources (posting, bidding, employee referral, promotion) are cheaper and motivating but risk insularity; external sources (advertising, agencies, campus, job boards, social media) widen the pool. A realistic job preview lowers inflated expectations and typically cuts early turnover. Yield ratios map the funnel: applicants per interview, interviews per offer, offers per hire.
Selection
Tools must be reliable (consistent) and valid (job-related). Validity evidence is content (the test samples job tasks), criterion (scores predict performance; the correlation is the validity coefficient), or construct (the test measures an underlying trait). Structured, situational, and panel interviews beat unstructured conversation. Work samples and job knowledge tests are strong; cognitive ability tests predict well but raise adverse impact risk; assessment centers use multiple exercises and assessors. Under the Uniform Guidelines, a protected group's selection rate below 80% of the highest group's rate signals adverse impact.
Training
Needs assessment has three levels: organizational, task (from job analysis), and person. Design uses goals, modeling, practice, feedback, reinforcement, and overlearning, with attention to transfer of training. Methods run from on-the-job training, apprenticeship, vestibule, and simulation to case study, role play, and e-learning. Kirkpatrick's four levels: reaction, learning, behavior, results.
Development
Development prepares people for future roles-job rotation, mentoring, coaching, committee assignments, sabbaticals, formal courses, and 360-degree feedback-rather than fixing a current performance gap.
Worked example
A distribution center needs 24 selectors. Last year 480 applicants yielded 120 interviews, 40 offers, and 20 hires; 16 of the 20 passed probation. Applicant-to-hire yield = 20 ÷ 480 = 4.2%; probation success = 16 ÷ 20 = 80%. Hires needed = 24 ÷ 0.80 = 30; applicants needed = 30 ÷ 0.042 ≈ 715. The reasoning: chain every loss point instead of using only the applicant-to-hire rate, which understates the recruiting load.
Traps that cost marks
- Confusing reliability with validity.
- Calling a test content-valid when it was validated against performance ratings (that is criterion).
- Skipping needs assessment and jumping to a favorite method.
- Treating reaction sheets as evidence of learning.
- Defending a cognitive test without checking adverse impact.