At a session entitled "Training Methods and Techniques in Occupational Safety and Health" presented at a convention of the American Society of Safety Engineers, Robert Nolan of the USA Casualty Co. put many of his own prescriptions into practice, eliciting enthusiastic participation from the audience.

Almost all attendees at this session actually did do some training on at least an occasional basis. But more than half had never had any formal education in how to do it. How did they learn? Typically, by trial and error, observation, and with the help of mentors.

Asked how much time was spent on the safety training of a new employee, the audience responded with a wide range of numbers, from 30 minutes to 102 hours, most clustering around the 15-25 hour mark. (Of course this is at least partly a reflection of the hazards in a given industry.)

Asked when this safety training took place, "right away on day one" was a frequent answer but perhaps not the most appropriate time, since the HR department will be delivering a whole "barrage of information on hours of work, benefits, payday, location of rest rooms, etc.," in which the safety rules in the handbook might wen be buried. But within the first week or month would be better than the second, much better than after 90 days, and so on.

New employees (new to the operation, not just the company) need safety information promptly, because studies have shown that "there is up to 20 times more chance of an accident within the first 180 days than in the entire rest of their career." If, as is often the case, the newcomer has to wait until the next round of regularly scheduled training on a particular topic, any worker who arrives just after that is going to be at risk until the next cycle.

Another widely shared drawback was the "very limited communication model" employed in much training. How do the majority of inexperienced trainers (and far too many of those who have been at it for a long time) try to achieve their stated goals of changing behavior, complying with regulations, and to help employees "understand the why as well as the what" of safe work practices goals? They talk-they tell trainees what they should do, or avoid doing, and why.

What's wrong with that picture? For one thing, the trainer will never know whether the vital lesson has been learned-until an accident or injury proves that it was not learned by at least one worker. So communication must be a two-way street: after the "sender" has delivered information to the "receiver," there must be feedback to indicate whether the correct message was received. If not, then the sender can "make adjustments" (perhaps use different examples, or different techniques) and try again.

If the only technique used has been talking, a new approach will almost certainly be beneficial. All of the senses can have some role to play in learning, and research on adult learning has suggested how large each role is:

• Taste (use less salt in the recipe next time; I taste saccharine-the respirator doesn't fit properly) accounts for I percent of learning.
• Touch (such items are hot, cold, hard, soft, sharp, greasy, bumpy, etc.) for 1.5 percent.
• Smell (the scents of smoke, banana oil, rotten eggs are among those that can send safety warnings) accounts for 3.5 percent.
• Heating (the instructor's words, an unusual grinding or tapping noise from the machine, a timer buzz or alarm bell) is the channel for 11 percent of messages-and thus learning. It is the most commonly used training medium and the basis for "much of our ineffective training."

But all of these together account for only 17 percent, while 83 percent of learning comes via the sense of sight. Material that can be seen and read, visual aids of all sorts, are therefore the most effective training tools. So why aren't films, videos, overheads, and the like made a part of virtually every training session? Cost is the usual answer, although many visuals can be bought, rented, or even developed in-house relatively inexpensively.

Adult learning theory also tells us that men and women learn more effectively based on:

• Multiple senses (especially not hearing alone)
• Desire (wanting to learn, for any of a number of reasons)
• Immediate benefit (I'm going to have to use this new skill starting tomorrow)
• Active involvement (in the training - e.g., asking questions, hands-on activity)
• Experience relatedness (this gibes with what I already know or have seen or done)
• Informality (e.g., circular arrangement instead of rows of hard chairs with instructor at a lectern; ability to move around during session; meeting on the shop floor at the machine being discussed)
• Assessment (having the sense that useful knowledge has actually been gained)
• Realism (all the illustrations of correct lifting seem to show one-cubic-foot boxes, whereas what I have to lift are eight-foot 2x4s, or bedridden patients).

Assuming that actual learning does take place during a training session, how much of it is retained, and for how long? If it's all gone by tomorrow, it won't contribute much to your trainee's health or safety next week. Research has shown that over the short-term (72 hours) adults tend to remember:

• 10 percent of what they read
• 20 percent of what they hear
• 30 percent of what they see
• 50 percent of what they see and hear
• 70 percent of what they say - preferably in their own words
• 90 percent of what they say as they do.

This obviously presents a clear pattern for construction of effective training. And it suggests that frequent refreshers may be necessary to achieve retention of knowledge over the long term.

The primary cause of ineffective training, according to the experts, is failure to plan adequately. It is essential to devote the necessary time and effort to understanding the subject matter, determining clear objectives, and developing an instructional plan specifically to fit the topic and objectives. This requires "significant commitment at management and supervisory levels."

Unfortunately, however, much training is done "on the fly" and so "may fulfill the letter of the statutes, but certainly does not meet the spirit of the standards."

Instructional objectives. Clear and specific objectives (1) help the trainer determine what is to be done, and how, (2) enable clear statement of the purpose of the training (share this with trainees), and (3) serve as a "road map" for the actual instruction.

What constitutes a helpful set of objectives? They should:

• Specify the desired behavioral results
• Identify the skills and knowledge that will support those behaviors
• Designate the learning experiences needed to acquire and retain the skills, the knowledge, and therefore the desired behaviors.

In stating an objective, choose action words like "perform" and "demonstrate" rather than vaguer, harder-to-verify ones like "know" or "understand." Desired behaviors should also be as precisely measurable as possible-laying X bricks per hour within a variation of no more than O.xcm from true; typing at a rate of Y words per minute with no errors.

Presentation methods. These can be selected once the objectives have been established. Although the lecture may not be the most desirable form of presentation, it is "the only way to do some of it, because it delivers the most information in the shortest time." But this should be supplemented with other techniques such as the workshop or roundtable discussion, role-playing, or demonstration by the trainer and, when possible, the trainees.

On-the-job training may be the most appropriate way to train someone in how to perform a particular operation, but it has at least one drawback: the senior worker who knows the job cold and inside-out also knows "all the shortcuts," which may mean that certain bad habits are being learned along with the routine. So some regular "how's it going? Show me how you do it" checks by the safety trainer or supervisor may be in order.

Again, remember that some kind of feedback mechanism must be included to assure good communication. What it is designed to discover is:

• Was the instruction effective? (Did the trainees learn? Were the objectives met?)
• How can the process be improved?

One way of getting answers to the first question is by conducting tests-oral, written, performance, or a combination. The written test, although easiest (and quickest) to administer to a group of any size, can be hard for some workers and, if too easy (multiple choice, true-false, checklist), may not give an accurate picture of the trainee's understanding-or any picture of how he or she will actually behave. Asking carefully structured questions and stimulating group discussion can be a very effective and informative approach.

Getting a statement of what was learned in the trainees' own words can yield useful information, although it is more time-consuming than a written test that all trainees can work on simultaneously. It's unlikely that performance testing, which most often requires a one-on-one situation, will be part of the training "class." But follow-up during the week or two after training can accomplish the purpose.

Because training is a process, not a product, and effective training requires more than a "quick-fix injection," follow-up must be done by the safety person or supervisor and can enhance and reinforce the training. A good way to manage this, described as "safety coaching," includes:

• Spending significant time on the work floor so as to observe all employees regularly
• Correcting negative behavior on the spot (exemplified by the company president who always carries extra pairs of safety eyeglasses so that if he finds an employee working without them he can could urge "please wear this pair of minefor the rest of the day - I want you to go home with both eyes intact").
• Providing positive reinforcement (catch them doing good-and praise them for it).

All of these techniques are just as readily practiced by members of the safety committee as by a supervisor or professional trainer.

In closing his presentation, die conference speaker suggested that a suitable slogan for trainers would be: "If the learner hasn't learned, the instructor hasn't taught." This may be a little difficult to accept, at times, but it's important to remember that OSHA seems to have much the same opinion. That's why stacks of records of training delivered won't protect against citations if workers interviewed by a compliance officer can't demonstrate the knowledge that the regulations require them to have.