Chapter 15 — Training and Development: Considerations in Design

Chapter 15 opens with eight learning goals, numbered 15.1 through 15.8, reproduced verbatim because Cascio and Aguinis use this numbering scheme throughout the book and instructors sometimes reference goal numbers directly.

1. 15.1 Identify key factors that are driving the demand for well-designed and well-executed programs of workplace learning.
2. 15.2 Explain what training and development activities are.
3. 15.3 Illustrate the fundamental requirements of sound training practice.
4. 15.4 Assess training needs and specify training objectives.
5. 15.5 Describe features of the learning environment that facilitate learning and transfer.
6. 15.6 Specify key elements of successful team training.
7. 15.7 Incorporate classic principles of learning into all training designs.
8. 15.8 Integrate key elements that will maximize positive transfer of training to the job.

Change, growth, and development are facts of organizational life: young people today typically change jobs at least seven times by their late 20s, and the workforce's composition keeps shifting — globally there will be many more older than younger people, and in the U.S. the non-Hispanic white population drops below 50% by 2040 (Toossi, 2012).

This carries two implications for employers. First, because products and services can be copied, a workforce's ability to innovate, solve problems, and build relationships becomes an organization's only sustainable advantage, making talent development vital. Second, managing a culturally diverse workforce and harnessing varied workers' motivation is a continuing challenge that ongoing training must address.

Four Trends Driving Demand for Workplace Learning (Cascio, 2017)

• Growing demands for personal and professional development — young adults rank continuous learning as the top feature of a new job; employers want "plug and play" recruits, but only 11% report finding them (Abadzi, 2016; Coy, 2014; Weber, 2014), shifting the skill-building burden onto employer training aligned with operating goals (Nathan, 2016).
• The effects of digital technology on work — cloud/mobile computing, big data/ML, sensors, robotics, and clean-energy technology are transforming global business; employees can now take a course on any subject online, part of a shift toward consumer-centric learning that puts employees, not training departments, in charge (Deloitte, Touche, Tohmatsu, Ltd., 2016).
• Increased training opportunities for nonstandard workers — more workers operate outside traditional full-time employment as free agents/e-lancers, allied-firm employees, or volunteers (Boudreau, Jesuthasan, & Creelman, 2015; Cascio & Boudreau, 2017); a gig-economy "career" accumulates project credits rather than hierarchical positions, but development remains critical to continued employability.
• Greater use of teams — team members must learn to ask for ideas, offer help unprompted, listen, give feedback, and value others' contributions; decades of research define what effective team training looks like (Cannon-Bowers & Bowers, 2011; Salas, Burke, & Cannon-Bowers, 2002; Salas & Cannon-Bowers, 2001).

Companies increasingly treat training expense as a capital cost, on par with plants and equipment — superior learning and growth opportunities give organizations a distinct edge competing for talent (Loughery, 2016; Sparshott, 2017).

Training and development share four properties (Goldstein & Ford, 2001; Kraiger, 2003; Noe, 2017): they are learning experiences; planned by the organization; occur after the individual joins; and further the organization's goals. They are therefore planned programs to improve performance at the individual, group, and/or organizational level — implying measurable, relatively permanent changes in knowledge, skills, attitudes, and/or social behavior.

Learning vs. Performance — a Deliberately Drawn Distinction

The phrase "relatively permanent" distinguishes learning from performance — a temporal distinction. Learning is a relatively permanent change in behavior from practice or experience (not simple maturation): the ability to perform, available over time. Performance is the demonstration of learning — observable behavior from which learning is inferred — and is affected by physical or mental state (fatigue, low motivation, noise, anxiety), which suppresses short-run performance far more than it erases long-term learning.

Done well, training produces sustained changes benefiting individuals, teams, organizations, and society (Aguinis & Kraiger, 2009). Surveys consistently identify four characteristics of companies with the most effective training (Colvin, 2009; Rifkin, 2011):

• Top management is committed — training is part of the corporate culture, as at Google, Disney, GE, and Cisco.
• Training is tied to business strategy and objectives and linked to bottom-line results.
• Organizational environments are "feedback rich" — stressing continuous improvement, risk taking, and learning from both success and failure.
• There is commitment to invest the necessary time and money.

Top-management commitment is not a soft variable: when management-by-objectives is implemented with high commitment, productivity gains run five times higher than with low commitment (Rodgers & Hunter, 1991), and job satisfaction rises about a third of a standard deviation with high commitment versus little or none with low/moderate commitment (Rodgers, Hunter, & Rogers, 1993).

Additional Determinants of Effective Training — the Noe & Colquitt (2002) Model

Training success depends not only on training quality but on the trainee's and program's relationship to the broader organizational context — organizational support and readiness for training can enhance or detract from training's impact (Colquitt, LePine, & Noe, 2000; Sitzmann & Weinhardt, 2015). Figure 15.1 (Noe & Colquitt, 2002) models individual characteristics — trainability, personality, age, attitudes — as influencing motivation, learning, transfer, and job performance.

Motivation to learn is a critical mediator: a force that energizes, directs, and maintains behavior (Steers & Porter, 1975), explaining significant variance in learning outcomes beyond cognitive ability alone (Colquitt et al., 2000; Noe & Colquitt, 2002). Work-environment features — climate, opportunity to perform trained tasks, organizational justice, individual vs. team context — matter at every stage: before training (motivation), during (learning), after (transfer, performance). Some individual traits (trainability, personality) are hard to influence, but others are organizationally shapeable: career attitudes, pretraining self-efficacy, the valence of training, and the work environment itself (Brown & Sitzmann, 2011; Quiñones, 1997; Switzer, Nagy, & Mullins, 2005).

To realize training's potential, resist emphasizing technology and technique and instead define first what is to be learned (Campbell, 1971, 1988) — viewing training as a network of interrelated components embedded in a larger organizational context (Aguinis & Kraiger, 2009; Brown & Sitzmann, 2011; Quiñones, 1995, 1997), per Figure 15.2, the chapter's general systems model.

Program development has three essential phases: needs assessment (planning), training and development (implementation), and evaluation. Needs assessment is the foundation — later phases depend on it, and incomplete assessment produces training out of tune with real needs. The implementation phase designs the environment to achieve specified objectives — "a delicate process that requires a blend of learning principles and media selection, based on the tasks that the trainee is eventually expected to perform" (Goldstein & Ford, 2001, p. 28). If the first two phases are done carefully, evaluation (Chapter 16) is straightforward: establishing success criteria and using experimental/quasi-experimental designs to determine what changed.

Four Types of Training-Evaluation Validity

Figure 15.2 frames possible training goals as four validity questions, each pointing to a different evaluation design:

• Training validity — did trainees learn anything during training?
• Transfer validity — did KSAs learned in training improve job performance?
• Intraorganizational validity — does a new trainee group perform like the original group, in the same organization?
• Interorganizational validity — can a program that works in one organization succeed in another?

These often yield different evaluation models, or different forms of the same model (Kraiger, 2002; Mattson, 2003; Noe, 2017; Wang & Wilcox, 2006). Evaluation should provide continuous closed-loop feedback that reassesses instructional needs for the next round of development.

The chapter specifies six steps for defining what is to be learned and the substantive content of training:

1. Analyze the interaction of training with other HR systems (recruitment, staffing, performance management, incentives).
2. Determine training needs.
3. Specify training objectives.
4. Decompose the learning task into its structural components.
5. Determine an optimal sequencing of the components.
6. Consider alternative ways of learning.

The overall goal is linking training content to desired job behaviors — reflecting the modern trainer's role as focused on improving performance under the organization's competitive strategy, not training for its own sake (Nathan, 2016; Tyler, 2008).

Interactions of Training and Development With Other Systems

Training operates in a complex organizational milieu; ignoring that context often produces programs with no effect, or worse. Even "whom do we train?" illustrates this — organizational boundaries are blurring between customers, suppliers, employees, non-employees, and competitors (Cascio & Boudreau, 2017), so any individual or group needing specific capabilities is a potential candidate for training (Cascio, 2010).

Training often fails to produce results because it isn't aligned with the organization's strategic direction (Nathan, 2016; Noe, 2017). Fixing that requires identifying needed future capabilities, comparing them to current capabilities, and closing the gap — with implications for recruitment, staffing, and incentives too. Three conditions must hold: the individual must be able to learn ("can do"), motivated to learn ("will do"), and supported by those who influence them.

Needs assessment determines whether training is necessary before resources are spent. Kraiger (2003) makes three points: it is treated as essential across virtually all instructional-design models; despite that, many programs skip it in practice — a large-scale meta-analysis found only 6% of studies reported any needs assessment before implementation (Arthur et al., 2003); and it has comparatively little ongoing research or theory relative to other training topics.

Pretraining motivation rises as adults see training as relevant to daily work. A thorough needs assessment — including experienced subject matter experts — demonstrates training's value beforehand, lowers anxiety, and boosts organizational support for transfer (Goldstein & Ford, 2001; Klein, Noe, & Wang, 2006; Noe, 2017).

The Three-Facet Approach, Plus Demographic Analysis (Figure 15.3)

Most needs-assessment methods fall under McGehee and Thayer's (1961) classic three-facet approach: organization analysis (where is training needed), operations analysis (what should the content be), and individual analysis (who needs it, and of what kind). All three should run continuously at three levels — organization (goal-setting managers), operations (managers specifying how goals are met), and individual (managers and workers doing the work).

Latham (1988) adds a fourth facet, demographic analysis — needs analysis by population (e.g., older workers, expatriates, managers at different levels) — to the traditional trichotomy. Figure 15.3 situates all four within the external environment (judicial decisions, civil rights laws, union activity, productivity, accidents, turnover, absenteeism), which supplies the outermost layer of relevant information.

Operations Analysis in Depth — Cognitive Task Analysis and Competency Models

Operations analysis systematically collects information on how work is done, performance standards, task execution, and needed competencies, drawing on managers and subordinates close to the work — who know the jobs best and whose involvement builds commitment, provided they have the experience and confidence to give meaningful data (Ford, Smith, Sego, & Quiñones, 1993).

For complex, high-stakes work (pilots, surgeons, accident-investigation teams), cognitive task analysis (CTA) targets the mental aspects of performance — decision making, problem solving, pattern recognition — that conventional task analysis (which identifies what gets done) cannot observe directly (Brannick, Pearlman, & Sanchez, 2017).

An emerging trend uses competency models to drive training curricula. A competency is a cluster of interrelated knowledge, skills, values, attitudes, or personal characteristics presumed important for job success (Noe, 2008, 2017); once validated, an organization-specific model can drive training design, development plans, 360-degree reviews, staffing, and promotion decisions.

Individual Analysis and Individual Development Plans (IDPs)

Individual analysis assesses how each employee performs against required standards, and — since standards shift in rapidly changing environments — determines whether training, versus new hiring or work redesign, is the right fix. One especially fruitful approach combines behaviorally based performance management with individual development plans (IDPs) from self-analysis. IDPs are a road map for self-development and should include:

• Statements of aims — desired changes in knowledge, skills, attitudes, values, or relationships.
• Definitions — areas of study, reflection, or testing, with activities or questions that support those aims.
• Ideas about priorities — what should be learned first.

Individuals typically build their own IDPs, with assistance, through career-planning workshops, structured exercises, management-by-objectives, or assessment centers. The full needs-assessment process should reveal what workers do, what behaviors matter, what learning is required, and what instructional content will produce it (Blanchard & Thacker, 2013; Goldstein & Ford, 2001) — guiding every subsequent training-method and evaluation choice.

Rapid Prototyping — an Alternative to Traditional Needs Assessment

Needs assessment remains important (Nathan, 2016) but can be ponderous, so some designers use rapid prototyping instead, borrowed from software development (Brown & Sitzmann, 2011; Welbourne, 2011): (1) assess needs and set training objectives; (2) construct and user-test prototypes; (3) implement and refine. It relies on parallel work, minimal commitments, and extensive testing — building and testing training before needs assessment is fully complete, then revising based on results. It suits settings where instruction is fast and cheap to build and revise; it fits poorly where instruction is costly or user testing is infeasible.

Once needs are identified, the fundamental next step is specifying training objectives — what is to be learned (Blanchard & Thacker, 2013; Campbell, 1988) — stated either behaviorally or operationally.

Behavioral Objectives — Mager's Three Components

Behavioral objectives refer to observable, measurable actions. Each should describe (a) the desired behavior, (b) the conditions under which it should occur, and (c) the standard by which it is judged (Mager, 1984). Worked example, for civil engineering students: "In a two-hour test following the last week of training [conditions], the student will be able to list the sequence of steps involved in building an on-ramp to a highway, specifying the standards for completion of each step [behavior]. All steps must be included in the correct order, and the standards for completion must match those in the textbook [criteria]."

Operational (End-Result) Objectives

Objectives may also be operational: "lower production costs" is vague, while "lower the costs of producing Model 600 lawn sprinklers 15% by April 30, by having one operator execute all operations using computer-controlled machinery" is precise and lets you assess its actual contribution to results.

Once objectives are specified, the next task is designing the environment to achieve them. Noe and Colquitt (2002) identify seven features of a learning environment that facilitate learning and transfer:

1. Trainees understand the training program's objectives and expected outcomes.
2. Training content is meaningful — examples, exercises, and terms are relevant.
3. Trainees get cues that aid recall — diagrams, models, key behaviors, advance organizers.
4. Trainees have opportunities to practice.
5. Trainees receive feedback from trainers, observers, video, or the task itself.
6. Trainees can observe and interact with other trainees.
7. The training program is properly coordinated and arranged.

Gagné's Three Principles of Coordination

On coordination, Gagné's (1962) classic paper offers three principles: (1) any task can be analyzed into distinct component tasks; (2) these components mediate final performance — their presence enables positive transfer, their absence reduces it to near zero; and (3) sound training design means identifying components, mastering each fully, and sequencing them for optimal mediational effect (p. 88).

Successful final performance depends on first mastering component subtasks — for procedural tasks, there's a more and a less efficient sequence, and mastering each subtask before the whole task is the efficient one. Gagné's ideas came from military skill-learning research, later strongly supported (Gagné, 1967, 1977; Gagné & Briggs, 1979; Gagné & Rohwer, 1969), and apply equally to training aimed at knowledge or attitude change.

These principles are necessary but not sufficient — individual and work-environment characteristics still matter. Tracey, Hinkin, Tannenbaum, and Mathieu (2001) studied 420 hotel managers: job involvement, organizational commitment, and perceived work-environment support predicted pretraining self-efficacy, which predicted pretraining motivation, which predicted posttraining reactions and declarative/procedural knowledge scores.

Technology-Delivered Instruction: Computer-Based Training and Smartphone Learning

Brown (2001) found considerable variability among 78 employees in an intranet course's level of practice and time on task, both predicting knowledge gain — those who skipped material or rushed learned less. Computer-based training matches classroom training on outcomes, but far less is known about smartphone learning (Cascio, 2017; Kraiger, 2014), which differs in three ways: learners aren't tethered to a workstation, they see content through a much smaller frame, and content arrives as smaller chunks assembled over time rather than one sitting (Kraiger, 2014).

Team performance has grown in emphasis — almost 90% of corporations worldwide use teams of some sort (EY, 2017). A team is defined by its common goal; if members hold conflicting goals, the unit's efficiency suffers (Kramer, Thayer, & Salas, 2013; Mathieu, Hollenbeck, van Knippenberg, & Ilgen, 2017) — the chapter's example is a baseball player who always swings for home runs regardless of the game situation.

Individual training can't do the whole job; interactions among members must be addressed directly (De Church & Mesmer-Magnus, 2010). This is what makes team training unique — it always uses simulation or real-life practice and always focuses on interactions among team members, equipment, and procedures (Bass, 1980; Colvin, 2006).

A Systematic Four-Step Approach to Team Training

Researchers (Cannon-Bowers & Bowers, 2011; Salas et al., 2002; Salas, Tannenbaum, Cohen, & Latham, 2013) developed a systematic four-step approach:

1. Conduct a team-training needs analysis — identify member interdependencies and coordination skills, and the cognitive skills/knowledge needed to interact as a team (e.g., roles and responsibilities).
2. Develop objectives for both taskwork and teamwork skills — core teamwork skills include adaptability, shared situational awareness, performance monitoring/feedback, leadership, interpersonal skills, coordination, communication, decision making; attitudinal skills include valuing teamwork, placing team goals above individual ones, mutual trust, shared vision (Cannon-Bowers & Bowers, 2011). Sequence taskwork before teamwork (Salas et al., 2002, 2013).
3. Design exercises around those objectives, with guided practice and feedback: team-coordination training (information exchange, cooperation), cross-training (exposure to teammates' roles), and guided team self-correction (reviewing events, exchanging feedback, planning ahead).
4. Design measures of team effectiveness against those objectives and use results to guide future training — evaluate collective efficacy, shared knowledge structures, team situational awareness, and shared mental models (Kraiger, 2003).

A popular application is crew resource management (CRM) training, using flight simulators to improve team communication and aviation safety — more than 50 studies show positive benefits (Aguinis & Kraiger, 2009), though CRM works better in aviation than in the more recently adopted health care setting (Salas, Wilson, & Burke, 2006).

Beyond objectives, environment, and sequencing, one problem remains: people have their own favorite ways of learning. The chapter's illustration: Susan prefers verbal learning (books), Nancy prefers kinesthetic/hands-on learning (a class), and Nicole prefers trial-and-error experiential learning (just experimenting). Others learn best visually (charts, pictures) or vicariously (watching others).

Simulation Games and Technology-Delivered Instruction (TDI)

TDI's popularity offers a way to tailor learning to individuals (Kolodny, 2016; Kraiger & Jerden, 2007). Simulation games — computer-delivered decision-making exercises in an artificial environment — are intrinsically motivating. Meta-analysis shows meaningful gains versus comparison groups: posttraining self-efficacy 20% higher, factual knowledge 11% higher, skill-based knowledge 14% higher, retention 9% higher (Sitzmann, 2011).

The drawback is cost: complex simulation games run $5–$20 million to build. Traditional online training takes about 220 hours to produce per hour of content, versus 750–1,500 hours for online simulations (Gabriel, 2010; Knowledge@Wharton, 2015; Sitzmann, 2011) — designers offset this through content reuse and by netting development costs against travel-cost savings from replaced classroom training.

As with computer-based instruction generally, e-learning participants vary widely in time spent per module and on practice; those who complete more practice and take more time learn the most (Sitzmann, 2011) — active learners learn the most.

Trainability and Individual Differences

Individual differences in abilities, interests, and personality are central to applied psychology. "Can do" factors (prior achievement, initial skill) and "will do" factors (training expectations) predict training performance (Gordon & Cohen, 1973; Robertson & Downs, 1979, 1989) — general mental ability alone predicts training success across jobs (Colquitt et al., 2000; Ree & Earles, 1991), and so does trainability.

Trainability is a person's ability to acquire skills, knowledge, or behavior to perform a job at a given level, in a given time (Robertson & Downs, 1979) — a combination of ability and motivation. Large-sample meta-analyses (n = 2,542 and 2,772) show work-sample trainability tests validly predict training performance, more strongly than job performance itself (Robertson & Downs, 1989).

Long-term benefit requires efficient learning, retention, and positive transfer, which is why "learning principles" developed over the past century remain training's principal theoretical basis. Which principles matter depends on whether trainees learn skills or facts (Wexley & Latham, 2002). For skills: goal setting, behavior modeling, practice, feedback. For facts: goal setting, meaningfulness of material, practice, feedback.

Goal Setting

"A person who wants to develop herself or himself will do so; a person who wants to be developed rarely is." Goal setting is one of the most effective ways to raise motivation — more than 500 studies show it improves performance across settings (Latham, 2009; Latham & Locke, 2017; Locke & Latham, 2013), founded on the premise that conscious goals regulate behavior (Locke, 1968). Five findings stand out:

1. Goal-setting theory is among the most valid and useful theories in organizational science (Locke & Latham, 2013); it produces about a 10% average productivity gain, best on low-complexity tasks (Schmidt, 2013; Wood, Mento, & Locke, 1987).
2. Goal commitment is necessary for goal setting to work (Locke, Latham, & Erez, 1988), and self-efficacy affects commitment (Frayne & Latham, 1987); once accepted, specific, difficult goals beat easy goals or "do your best" (Klein, Cooper, & Monahan, 2013; Klein, Wesson, Hollenbeck, & Alge, 1999).
3. For complex tasks, participation boosts goal acceptance, especially when a goal is initially seen as unreasonable (Erez, Earley, & Hulin, 1985; Erez & Zidon, 1984); for simple tasks, assigned goals work just as well (Shalley, Oldham, & Porac, 1987).
4. Past goal-setting experience shapes future choices — employees pick harder goals after easy ones, easier goals after hard ones (Locke, Frederick, Buckner, & Bobko, 1984).
5. Providing task guidance and a rationale for the goal further enhances goal setting's effect on performance (Earley, 1985).

Goal setting is not risk-free — excessive risk taking, ignored nongoal dimensions, pressure to cheat, and stress are possible side effects, though controllable (Latham & Locke, 2006; Locke & Latham, 2009; Ordóñez, Schweitzer, Galinsky, & Bazerman, 2009). For trainees specifically: make objectives clear at the outset; set goals challenging but not impossible; and supplement the ultimate goal with subgoals (evaluations, work samples, quizzes) so confidence builds as hurdles clear.

Behavior Modeling

Behavior modeling rests on social-learning theory (Bandura, 1977, 1986, 1991, 2013): we learn by observing others, through attention, retention, reproduction, and motivation. Goldstein and Sorcher's (1974) four-step "applied learning" approach operationalizes it: modeling (watching effective models on video), role-playing (practicing modeled behaviors), social reinforcement (trainer praise/feedback), and transfer of training (using it on the job). In short: observe a model, remember it, do it, use it on the job (Baldwin, 1992) — working through changes in trainees' mental models (Davis & Yi, 2004), at both individual and team level (Marks, Sabella, Burke, & Zacarro, 2002).

For reproducibility goals (e.g., a golf swing), show only correct examples; for generalization goals (e.g., conflict resolution), mix positive and negative examples (Baldwin, 1992). Retention aids — reviewing or rewriting written "learning points," mental rehearsal — enhance modeling (Decker & Nathan, 1985; Mann & Decker, 1984; Hogan, Hakel, & Decker, 1986; Marks et al., 2002), building cognitive "scripts" linking cognition to behavior (Cellar & Wade, 1988). The most effective practice format includes video replay of each rehearsal in a small group with one or two observers (Decker, 1983). The current formula has five components: modeling, retention processes, role playing, social reinforcement, transfer of training (Decker & Nathan, 1985).

Meta-analysis of 117 behavior-modeling studies (Taylor, Russ-Eft, & Chan, 2005) found the largest effects on declarative and procedural knowledge (effect sizes near 1.0); the overall effect on job behavior was d = 0.27, with variance suggesting moderators worth studying (Aguinis, 2004b). It isn't universally suitable — Gist, Schwoerer, and Rosen (1989) found modeling worked for moderate-to-high self-efficacy trainees on computer software, but a one-on-one tutorial worked better for low-self-efficacy trainees.

Lasting transfer depends on what happens after training. Where trainees were encouraged but not evaluated or sanctioned, there was no long-term change (Russell, Wexley, & Hunter, 1984); where managers directed and enforced use of new skills — even removing supervisors who refused — behavior changed (Latham & Saari, 1979). Three strategies follow (Russell et al., 1984): show supervisors why new behaviors outperform old ones; have trainees mentally rehearse until the behavior fits their self-image, then try it; and follow training with goal setting and reinforcement at work.

Meaningfulness of the Material

Factual material is easier to learn when meaningful — rich in associations for trainees. Three techniques help: give an overview upfront so trainees see how units fit together (Wexley & Latham, 2002); use familiar examples and terms tied to real job improvement; and teach simpler skills before complex ones, since complex skills are built from simpler ones (Gagné, 1977; Gist, 1997) — true for accounting, programming, or nuclear medicine alike.

Practice

Learning requires practice — the active use of training content (Ehrenstein, Walker, Czerwinski, & Feldman, 1997; William, 2013) — with three aspects: active practice, overlearning, and session length.

Active practice: verbalizing or reading is not enough for skill learning — only active practice provides the internal cues that regulate motor performance, discarding inefficient motions as feedback continues. Error-management training (Keith & Frese, 2005) is an alternative to the traditional error-avoidant approach: it encourages errors, then reflection on causes and prevention. Meta-analysis found it superior to both error-avoidant and unguided exploratory training (d = .44), with larger effects on posttransfer measures and novel tasks — suggesting deeper understanding that transfers (Keith & Frese, 2008; Aguinis & Kraiger, 2009).

Overlearning: practicing well past first correct performance until a task becomes "second nature" — the single most effective ingredient in preventing skill/knowledge decay (Abadzi, 2016; Arthur, Bennett, Stanush, & McNelly, 1998; Driskell, Willis, & Copper, 1992), critical for infrequent, high-stress tasks (e.g., CPR) but less so for daily-practiced ones. It likely works by strengthening stimulus-response bonds, enhancing automaticity, and building confidence — but its retention boost dissipates to zero after five to six weeks without refreshers (Driskell et al., 1992).

Session length: practice may be distributed (spaced with rest) or massed (crowded together). For equal total practice, distributed practice generally produces better learning (Goldstein & Ford, 2001; Noe, 2017) — continuous practice is fatiguing and understates true learning, and irrelevant performances learned alongside correct ones forget faster during rest periods. Varying tasks during practice may also speed learning and improve retention (Holladay & Quiñones, 2003; Wymbs, Bastian, & Celnik, 2016). Exception: for difficult conceptual "thought problems," massed sessions early on can outperform spread-out sessions.

Feedback

Feedback — information about one's attempts to improve — corrects mistakes and reinforces learning. It may be intrinsic (from the task) or extrinsic (from others); qualitative, quantitative, informative, or evaluative. It generally improves performance (Ashford & De Stobbeleir, 2013; Ilgen, Fisher, & Taylor, 1979; Martocchio & Webster, 1992; Stajkovic & Luthans, 2003), but managers often misjudge its value — Greller (1980) found supervisors underestimate how much subordinates value feedback from the task itself and coworkers, while overestimating formal rewards and comments from the boss.

Key research findings on feedback:

• Feedback often results from performers proactively seeking it themselves (Herold & Parsons, 1985), especially when they suspect a problem challenging their self-image as competent (Larson, 1989).
• Managers attributing poor performance to low effort use two-way, problem-solving feedback; those attributing it to low ability use one-way "tell-and-sell" — only the problem-solving approach changes behavior (Dugan, 1989).
• More feedback isn't always better — biweekly feedback matched weekly feedback's effect on factory safety behavior (Chhokar & Wallin, 1984). Specificity should vary: more specific feedback helps good-performance learning but can hurt poor-performance learning (Goodman & Wood, 2004).
• Immediate feedback isn't ideal for everyone — withholding it from experienced learners sharpens critical thinking and improves retention; give immediate feedback to novices, less frequent to experts (Brown & Ford, 2002; Schmidt & Bjork, 1992).
• Only feedback attributing performance to controllable causes, explaining why, and specifying improvement steps is useful (Jacoby, Mazursky, Troutman, & Kuss, 1984; Martocchio & Dulebohn, 1994).
• Feedback should lead with positive information before negative to be accepted as accurate (Stone, Gueutal, & McIntosh, 1984); on multiple dimensions, let employees choose which to receive first (Ilgen & Moore, 1987).
• Feedback improves performance beyond training and goal setting alone — present all three as a package (Chhokar & Wallin, 1984).
• Feedback affects group as well as individual performance (Cannon-Bowers & Bowers, 2011): one fast-food store cut food costs 15% and raised profits 193% over a year using performance feedback (Florin-Thuma & Boudreau, 1987); an Air Force base study found group feedback raised productivity 50% over baseline, goal setting 75%, incentives 76%, with no comparable control-group rise (Pritchard, Jones, Roth, Stuebing, & Ekeberg, 1988).
• The trainee's immediate supervisor gives the most powerful feedback of all (Pidd, 2004) — without supervisor reinforcement, training becomes "encapsulated" (Haire, 1964) and transfer is minimal or negative.

Training's usefulness largely depends on transfer of training — applying learned behaviors to the actual job. Transfer may be positive, negative, or neutral; the chapter calls it probably the single most important consideration in training and development. Negative transfer costs twice: the training itself was wasted, and performance is now actively worse (Ford, 2017; Brown & Sitzmann, 2011; Burke & Hutchins, 2008).

A meta-analysis of 107 management-training evaluations found transfer effects vary substantially by rating source (Taylor, Russ-Eft, & Taylor, 2009): self-ratings alone tend toward overly optimistic transfer estimates, subordinate ratings alone toward overly pessimistic ones. Multiple rating sources — supervisors, peers, subordinates, self — give the most realistic picture.

Machin's (2002) Checklist for Maximizing Positive Transfer

Because transfer environments are largely unique to each application (Holton, Chen, & Naquin, 2003), the chapter lists concrete design steps for before, during, and after training:

• Ensure peer and supervisor support for transfer, moderated by how much trainees identify with those groups (Kontoghiorghes, 2004; Pidd, 2004).
• Maximize similarity between the training and work situations; use interactive activities to encourage participation.
• Give trainees broad experience with the tasks so they can handle real-world variation — adaptive expertise (Baldwin, Ford, & Blume, 2009).
• Ensure trainees understand underlying principles, especially for jobs applying principles to problems (engineers, analysts).
• Link training content directly to work content ("what you learn today, you'll use tomorrow"); action learning — working real business problems — is an excellent vehicle for this (Levitz, 2010).
• For team training, transfer is maximized with open information access, diverse membership, and sufficient team size — these three elements explained over half the variance in team-effectiveness ratings in one study (Magjuka & Baldwin, 1991).
• Ensure what's learned is actually used and rewarded — supervisors and peers are the gatekeepers (Ford, Quiñones, Sego, & Sorra, 1992; Pidd, 2004); without their support, don't expect much transfer (Chiaburu & Marinova, 2005; Gaudine & Saks, 2004; Huint & Saks, 2003).

Trainee attitudes also affect transfer (Kraiger, 2014; Noe, 1986, 2008; Switzer et al., 2005): transfer is higher when trainees are confident in the new skills, aware of situations to use them, believe performance will improve, and see the KSAs as genuinely useful — helping them generalize learning from one context to their regular job duties.

Cascio and Aguinis close every chapter with an "Evidence-Based Implications for Practice" list — a distilled set of practitioner takeaways. For Chapter 15, it functions as the chapter's own executive summary.

• Resist the temptation to emphasize technology and techniques in training; instead, take the time to do a thorough needs assessment that reveals what is to be learned at the individual or team level and what the substantive content of training and development should be.
• Recognize that organizational boundaries are blurring, such that the border between employees, nonstandard workers, customers, suppliers, and even competitors is becoming fuzzier — any individual or group needing specific capabilities is a potential training candidate.
• Create an optimal environment for learning — clear objectives, meaningful material, practice and feedback opportunities, and organizational support for the training's content.
• Incorporate learning principles — goal setting, behavior modeling, meaningfulness, practice, feedback — into every training design.
• The most fundamental objective of well-designed training is positive transfer back to the job; use multiple rating sources (supervisors, peers, subordinates, self-ratings) for a realistic transfer assessment.

Chapter 15 ends with ten discussion questions. Below, each is paired with a concise model answer grounded directly in the chapter's content.

1. Your boss asks you to identify key characteristics of organizations and individuals that are related to effective training. What would you say?

Organizationally: top-management commitment as part of the culture, training tied to strategy and bottom-line results, feedback-rich environments, and real investment of time and money. Individually: trainability (ability plus motivation), pretraining self-efficacy, positive job attitudes, and favorable expected outcomes (valence). The Noe and Colquitt (2002) model shows these interact at every stage — before training (motivation), during (learning), and after (transfer and performance).

2. Transfer of training is important. What would you do to maximize it?

Follow Machin's (2002) checklist: secure peer and supervisor support; make training resemble the actual work situation; build adaptive expertise through broad task experience; ensure understanding of underlying principles; link content explicitly to work; and for teams, ensure open information access and diverse membership. Most importantly, make sure what's learned is used and rewarded on the job — without supervisor and peer reinforcement, training becomes encapsulated and transfer stalls.

3. Outline a needs-assessment process to identify training needs for supermarket checkers.

Start with organization analysis to confirm checker performance problems are behavior-changeable through training (not staffing or scheduling) and to gauge management support for transfer. Add demographic analysis if a subgroup (new hires, part-timers) has distinct needs. Run operations analysis by observing checkout work directly and consulting supervisors and experienced checkers to define competencies and standards for speed, accuracy, and customer interaction. Finish with individual analysis, comparing each checker's performance against those standards.

4. What should individual development plans include?

IDPs should include statements of aims (desired changes in knowledge, skills, attitudes, values, or relationships), definitions (study/reflection areas with supporting activities or questions), and priorities (what to learn first). They are typically self-built, with assistance, through career-planning workshops, structured exercises, management-by-objectives, or assessment centers — a personal road map for self-development.

5. What would an optimal environment for training and learning look like?

Per Noe and Colquitt's (2002) seven features: trainees understand objectives and outcomes; content is meaningful and relevant; recall cues (diagrams, advance organizers) are provided; real practice opportunities exist; feedback comes from trainers, observers, or the task; trainees can interact with each other; and the program is properly sequenced, following Gagné's principle of mastering component subtasks before the whole task.

6. Describe the components of an integrated approach to the design of team-based training.

The four-step approach starts with a team-training needs analysis identifying interdependencies and required cognitive/coordination skills. Next, objectives are set for taskwork and teamwork skills, with taskwork sequenced first. Third, exercises are built around those objectives using team-coordination training, cross-training, and guided team self-correction, with practice and feedback throughout. Finally, measures of team effectiveness (collective efficacy, shared mental models, team situational awareness) evaluate and refine future team training.

7. How might behavior modeling be useful in team-based training?

Teams can watch videos of effective team interactions, role-play the observed coordination and communication behaviors, receive social reinforcement, and transfer those behaviors to real team tasks (Marks et al., 2002). Because effective teamwork requires generalizable interpersonal skills rather than one fixed motor sequence, team-based modeling should mix positive and negative examples so members learn underlying principles, not just a script.

8. How do behavioral baselines help researchers to assess behavioral transitions in training?

A behavioral baseline captures each individual's starting state before training, using their own prior history as their control — letting researchers measure the transition caused by training itself rather than pre-existing differences. Bass, Cascio, McPherson, and Tragash (1976) used exactly this method with more than 2,000 subjects in race-relations training, establishing attitude baselines via questionnaire before training.

9. Top management asks you to present a briefing on the potential effects of goal setting and feedback. What would you say?

Goal setting reliably improves performance — about 10% in productivity on average — when goals are specific, difficult but attainable, and accepted. Feedback delivered alongside goal setting produces effects well beyond training alone: field studies show feedback, goal setting, and incentives compounding to raise productivity 50 to 76% over baseline. The caveat: feedback must attribute performance to controllable causes, explain the “why,” lead with positive information, and — most powerfully — come from the trainee's immediate supervisor, since unsupported training becomes “encapsulated” and fails to transfer.

10. Your boss just returned from a conference where she saw a demonstration of "rapid prototyping" for safety training of warehouse employees. She asks you to use that approach to design a training program in customer service. Outline the steps you would take.

Rapid prototyping runs in three overlapping phases rather than one linear needs-assessment cycle: first, quickly assess needs and set objectives; second, build a working prototype and test it immediately with real trainees, gathering direct feedback; third, implement the refined version, continuing to adjust based on ongoing testing. This fits customer service well if content is fast and cheap to build and revise — a full, expensive simulation program would instead favor a slower, more thorough needs assessment.

Every bolded or explicitly defined term in Chapter 15, in one line each, in the order the chapter introduces them.

A single table capturing the chapter's systems model, needs-assessment facets, learning principles, and transfer checklist — everything you need to answer a cold-call question about Chapter 15 without re-reading it.