4.10 Team Training: Needs Assessment, Delivery, and Evaluation
Key Takeaways
- The needs assessment sequence is identify skills gaps, develop learning objectives, prepare a training plan, and develop training materials.
- Adult learning theory holds that adults are self-directed, bring experience, are problem-centred, and need immediate relevance.
- Kirkpatrick's four evaluation levels are reaction, learning, behaviour, and results; most organizations stop at reaction.
- Pre-training and post-training testing is the standard method for measuring learning, and it requires the pre-test to be administered before any exposure.
- Skills gap equals required competency minus current competency, and it must be assessed per role, not per person or per department.
Why training appears in a Black Belt blueprint
Two reasons. Inside the project, team members need just-in-time capability in the tools they will use. At the end of the project, the Control phase requires that process owners and staff are trained on the revised methods -- the Body of Knowledge treats that as a sustainment requirement in section VIII.D. A control plan handed to people who have not been trained on it is a document, not a control.
Needs assessment
The Body of Knowledge gives an explicit four-step sequence. Follow it in order.
1. Identify skills gaps
Assess against the role, not the individual, and use a skills matrix:
| Skill | Required level (operator) | Current: A. Chen | Current: R. Patel | Gap |
|---|---|---|---|---|
| Read and react to an X-bar/R chart | 3 (apply unaided) | 2 | 1 | 1 / 2 |
| Perform first-piece verification | 3 | 3 | 2 | 0 / 1 |
| Record reject reason codes | 3 | 3 | 3 | 0 / 0 |
A four-level scale is conventional: 1 aware, 2 can perform with help, 3 can perform unaided, 4 can teach others. Sustainment requires at least one level-4 person per shift, or the capability leaves with a single individual.
Sources of gap evidence: observation at the gemba, error and defect data by cause, audit findings, supervisor assessment, and self-assessment (which reliably overstates and must be triangulated).
2. Develop learning objectives
An objective states what the learner will be able to do, observably, and under what conditions. Use action verbs from Bloom's taxonomy at the appropriate cognitive level -- the same taxonomy ASQ uses to tag the Body of Knowledge.
- Weak: "Understand control charts."
- Strong: "Given an X-bar/R chart with a run of seven points above the centre line, the operator will identify the out-of-control condition and execute the documented reaction plan without assistance."
The objective determines the assessment. If you cannot state how you would test it, the objective is not yet written.
3. Prepare a training plan
The plan specifies audience, objectives, content, delivery mode, duration, schedule, trainer, location, and required materials. Two decisions dominate:
- Sequence: teach in the order the learner will use the skills, not in the order the theory is organized.
- Timing: deliver as close as possible to first use. Training given six weeks before a process change has largely decayed by cutover.
4. Develop training materials
Materials should mirror the job aids people will actually use -- the standard work sheet, the reaction plan, the control chart itself -- rather than exist as separate slide decks. The best training material is the job aid, taught.
Delivery
Adult learning theory
Four principles, and each has a direct implication:
| Principle | Implication for delivery |
|---|---|
| Adults are self-directed | Explain the objective and let learners work; do not lecture at them |
| Adults bring experience | Start from their process and their examples, not generic case studies |
| Adults are problem-centred | Teach a tool by solving a real problem with it, not by defining it first |
| Adults need immediate relevance | State up front how this changes their work tomorrow |
Modes of learning
| Mode | Strengths | Weaknesses | Best for |
|---|---|---|---|
| Classroom | Efficient for groups; discussion | Passive; decays without practice | Concepts and rationale |
| On-the-job / hands-on | Highest transfer; real context | Trainer-intensive; variable quality | Procedures and equipment |
| E-learning | Scalable; consistent; on demand | Low engagement; poor for skills | Awareness, compliance, refreshers |
| Simulation / kata | Safe practice; immediate feedback | Setup cost | High-consequence or rare situations |
| Coaching / mentoring | Individualized; builds depth | Does not scale | Belt development, judgment skills |
Blended delivery is normal: e-learning for the concept, hands-on for the skill, coaching for the judgment.
Soft skills
The Body of Knowledge names soft skills explicitly. Facilitation, listening, questioning, giving feedback, and presenting cannot be taught by lecture; they need practice with observation and feedback, usually through role play with a structured observer checklist.
Evaluation
Kirkpatrick's four levels is the standard framework, and the exam expects you to distinguish them.
| Level | Question | Typical method | Cost | How often used |
|---|---|---|---|---|
| 1. Reaction | Did they find it useful and relevant? | Feedback survey at the end | Low | Almost always |
| 2. Learning | Did knowledge or skill increase? | Pre-test and post-test; skills demonstration | Low-medium | Sometimes |
| 3. Behaviour | Are they doing it differently on the job? | Observation, audit, KBI, 30-60-90 day follow-up | Medium | Rarely |
| 4. Results | Did the business measure move? | Defect rate, cycle time, yield, cost | High | Very rarely |
Level 1 measures satisfaction and predicts almost nothing: learners rate entertaining sessions highly regardless of transfer. Level 3 is the level that matters to a Black Belt, because a control plan depends on behaviour change, not on knowledge. Build the level 3 check into the control plan itself as a key behaviour indicator with a defined audit frequency.
Pre-training and post-training testing
The Body of Knowledge names pre- and post-testing specifically. Three rules make it valid:
- Administer the pre-test before any exposure, including the introductory briefing.
- Use equivalent but not identical items, or the gain measures recall of the test rather than learning.
- Set a defined pass standard tied to the learning objective, and re-train those who do not meet it rather than recording the average and moving on.
Evaluation planning
Decide the evaluation levels, methods, and success criteria while designing the training, not after delivering it. Retro-fitting evaluation is how organizations end up with a stack of level 1 smile sheets and no evidence that anything changed.
A plant trains all operators on a new control plan and reports that 94% rated the session as useful. Which Kirkpatrick level has been evaluated, and what does the result tell the Black Belt about sustainment?
Which sequence does the ASQ Body of Knowledge specify for implementing an effective training curriculum?
A trainer administers a pre-test, delivers the session, then administers the identical test as a post-test and reports a 40-point gain. What is the principal validity problem?