Supervision and Management (200-300 words) M8

Journal of Safety Research, Vol. 30, No. 4, pp. 237–249, 1999 Copyright © 1999 National Safety Council and Elsevier Science Ltd

Printed in the USA. All rights reserved 0022–4375/99 $–see front matter

Pergamon

PII S0022-4375(99)00019-5

Critical Success Factors for Behavior-Based Safety: A Study of Twenty Industry-wide Applications

Jason P. DePasquale and E. Scott Geller

One-on-one interviews and focus-group meetings were held at 20 organiza- tions that had implemented a behavior-based safety (BBS) process in order to find reasons for program success/failures. A total of 31 focus groups gave 629 answers to six different questions. A content analysis of these responses uncov- ered critical information for understanding what employees are looking for in a BBS program. A perception survey administered to individual employees (n 5 701) at these organizations measured a variety of variables identified in prior research to influence success in safety efforts. The survey data showed five vari- ables to be significantly predictive of employee involvement in a BBS process: 1) perceptions that BBS training was effective; 2) trust in management abilities; 3) accountability for BBS through performance appraisals; 4) whether or not one had received education in BBS; and 5) tenure with the organization. Also, employees in organizations mandating employee participation in a BBS process (n58 companies) reported significantly higher levels of: (a) involvement; (b) trust in management; (c) trust in coworkers; and (d) satisfaction with BBS train- ing than did employees whose process was completely voluntary (n 5 12 com- panies). In addition, employees in mandatory processes reported significantly greater frequency of giving and receiving positive behavior-based feedback. © 1999 National Safety Council and Elsevier Science Ltd

Keywords: Behavior-based safety, injury reduction, safety culture, employee involvement, training effectiveness

INTRODUCTION jured (U.S. Department of Health and Human Ser- vices [DHHS], 1998). In addition to these sober-

Occupational injuries due to at-risk work behavior ing statistics are those that estimate manufacturing remain a significant problem. On a daily basis, an facilities annually incur an average of 101 lost estimated 16 workers are killed and 36,000 are in- workdays due to injuries per 100 employees

Dr. E. Scott Geller, a professor of Psychology at Virginia Jason P. DePasquale is an Associate Project Director with Tech for three decades and a Senior Partner of Safety Perfor-

International Survey Research (ISR). He received his BA in mance Solutions, has authored more than 250 research articles Psychology from Albany State in 1991, and his MS in Indus- and over 50 book chapters addressing the development and trial/Organizational from Rensselaer Polytechnic Institute & evaluation of behavior-change interventions to improve qual- State University in 1994. Currently he is a Ph.D. candidate in ity of life. He recently published his seventh book on manag- Industrial/Organizational Psychology at Virginia Tech. ing behaviors and attitudes for occupational health and safety.

Winter 1999/Volume 30/Number 4 237

(Leigh, 1995; Miller, 1997; National Institute for Occupational Safety Health [NIOSH], 1998; Na- tional Safety Council [NSC], 1998; U.S. Bureau of Labor Statistics [USBLS], 1997). In sum, ap- proximately 250,000 potential productive years of life are lost each year in the United States because of premature death due to work-related injuries (Baker, Conroy, & Johnston, 1992).

Miller (1997) estimated that every year U.S. employers pay approximately $200 billion in di- rect costs associated with injuries that occur both on and off the job. Occupational injuries account for three-quarters of this total or nearly $155 bil- lion annually. This amounts to over $1400 per work-related injury. The majority of these costs are in the form of insurance premiums for work- ers and their families, as well as workers’ com- pensation for days lost from work (Miller, 1997). And to make matters worse, these estimates are likely gross under-estimations because of unreli- able surveillance techniques and the fact that many injuries are not reported (Baker et al., 1992; The National Committee for Injury Pre- vention and Control [NCIPC], 1989).

In order to combat the ever-present threat of em- ployee injury and associated losses, many organiza- tions are implementing what is referred to as behav- ior-based safety (BBS). This approach to injury control has a number of advantages over more tra- ditional approaches, including: (a) it can be admin- istered by individuals with minimal professional training; (b) it can reach people in the setting where the problems occur (e.g., school, workplace, or community at large); and (c) the leaders in the set- tings can be taught the behavior-change techniques most likely to work under specific circumstances (e.g., Baer, Wolf, & Risley, 1968; Daniels, 1989). Research has also shown BBS to be cost effective, primarily because behavior-change interventions are straightforward and relatively easy to adminis- ter, and their impact can be regularly monitored by indigenous personnel (Daniels, 1989; Geller, 1996; Geller, Lehman, & Kalsher, 1989; Sulzer-Azaroff & De Santamaria, 1980).

Over the years, BBS has been applied fre- quently and successfully in various industrial set- tings. For example, research has demonstrated the cost-effectiveness of: (a) participative educa- tion to increase safety-belt use (Kello, Geller, Rice, & Bryant, 1988); (b) incentives/rewards to increase safety-belt use (Geller, 1984; Geller & Hahn 1984); (c) goal-setting and behavioral feed- back to increase safety-belt use, turn-signal use, and complete intersection stopping (Ludwig &

Geller, 1991, 1997) and reduce driving speed (Van Houten & Nau, 1983); and (d) pledge-card commitment strategies to increase employees’ use of safety belts (Geller, Rudd, Kalsher, Streff, & Lehman, 1987) and other personal protective equipment (Streff, Kalsher, & Geller, 1993).

Behavior-based safety starts by defining one or more critical behaviors to target. Then these behaviors are observed and recorded in particular work settings. When a relatively stable baseline measure of the frequency, duration, or rate of be- havior is obtained, an intervention is imple- mented to change the behavior in beneficial di- rections. This intervention might involve removing environmental barriers, modifying a workstation, or adding antecedents or consequences to the sit- uation to alter response probability. The fre- quency, duration, or rate of the target behavior is recorded during and after the intervention and compared to baseline measures of behavior to de- termine intervention impact (Daniels, 1989; Geller, 1996, 1998b).

Facilitating Employee Involvement in BBS

Although BBS methods are consistently effective at increasing the occurrence of safe behaviors, they can only work optimally if used throughout an organization. In other words, if employees do not “buy-in” to BBS principles, participate ac- tively in observation and feedback sessions, and help to implement BBS intervention procedures, research describing the impact of this approach is academic (pun intended). Therefore, a primary objective of the current study was to identify or- ganizational and interpersonal variables that in- hibit or facilitate employee involvement in a BBS process. The research literature offered some guidance for our focus and for formulating hy- potheses.

Providing adequate BBS training. Learning experiences can be a potent source

of stimulation. Training programs are learning experiences designed to produce desired cogni- tive and/or behavior change among participants. An employees’ first exposure to a BBS process often occurs during a training session. As such, employee perceptions regarding the quality and relevance of BBS training received may have great potential for determining the frequency and quality of involvement in a BBS process. How- ever, our literature review found no empirical in- vestigations of the relationship between percep-

Journal of Safety Research 238

tions of a training process and subsequent involvement in a safety program.

Mandatory versus voluntary participation. One obvious way to promote employee par-

ticipation in a BBS process is to make such a process mandatory. In other words, some mini- mal level of participation in a BBS process could be a job requirement. A potential drawback of such a mandatory approach, however, is that em- ployees may feel a diminished sense of personal control and engage in resistance behaviors (Geller, 1998a). For example, monitoring sys- tems that produced constant vigilance of produc- tion behaviors (Grant, Higgins, & Irving, 1988) or management-set standards employees per- ceived as unreasonably high led to the occur- rence of false reporting behaviors to beat the sys- tem (Dose & Klimoski, 1995). Research by Kelman (1958) suggests organizations limiting personal control increase the occurrence of de- fensive strategies, including the motivation to do only what is required and no more.

Other research has found beneficial effects of elevated levels of personal control. For example, organizational structures that provide employees greater discretion were shown to correlate posi- tively with employee satisfaction, perceived quality of work life, and actual job performance (Hackman & Lawler, 1971; Sashkin, 1984). In contrast, a lack of control over organizational outcomes can lead to a state of learned helpless- ness (Seligman, 1975) and lowered effectiveness on the job. In this state, employees are said to possess “trained incapacity” (Dose & Klimoski, 1995).

The role of interpersonal trust. Little research has examined the direct impact

of interpersonal trust on performance outcomes (McAllister, 1995). However, researchers have long recognized trust as a hallmark of favorable relationships (Hart, 1988; Zand, 1972) and orga- nizational performance (Granovetter, 1985). In fact, trust between individuals and groups within an organization is considered a critical ingredient in the long-term stability of the organization and the well-being of its members (Cook & Wall, 1980). In other words, it is generally agreed that high levels of interpersonal trust are beneficial for most organizational functioning.

Yamagishi and Cook (1993) investigated the impact of interpersonal trust on individual partic- ipation in exchange relationships. They found

strong support for the notion that interpersonal trust increases an individuals’s level of participa- tion in a group situation. In fact, several other studies have shown a beneficial impact of inter- personal trust on the frequency and quality of group member interaction as well as participation in decision-making situations (e.g., Samuelson, Messick, Rutte, & Wilke, 1984;). The role of in- terpersonal trust may be particularly critical for successful BBS. An important BBS method in- volves employees in the observation of other em- ployees’ work behaviors. These observations are recorded systematically on a behavioral checklist and turned in for tracking and group feedback. If employees do not trust both the intentions and abilities of the behavioral observers, an observa- tion and feedback process cannot work. In other words, if employees perceive the BBS process to be another method for management to monitor their behavior, or for coworkers to “rat” on each other, participation in the process will likely be minimal. As concluded by Geller (1998c), “Lack of interpersonal trust causes resistance to an ob- servation and feedback process, and interper- sonal trust is what’s missing in a culture deemed unready for behavior-based safety” (p. 14).

Hypotheses

Based on our review of the literature related to the present research, we derived the following hypotheses.

Hypothesis 1. The quantity and quality of BBS training will influence the amount of employee involvement in a BBS process.

Hypothesis 2. Mandated participation in a BBS process will decrease personal involvement and satisfaction with a BBS process.

Hypothesis 3. Trust in management abilities and intentions will facilitate employee involve- ment in a BBS process.

Hypothesis 4. Trust in coworkers abilities and in- tentions will facilitate employees involvement in a BBS process.

METHODS

Subjects and Setting

Participation was solicited through survey re- search conducted in a professional safety journal, as well as through presentations and interactions at professional safety conferences and work-

Winter 1999/Volume 30/Number 4 239

shops. The participants were employees working at 20 different industrial sites. From these 20 or- ganizations, 245 employees (221 males, 24 fe- males) made up 31 different focus groups. About 80% of these participants were male, a male-to- female ratio that paralleled the workforce. All 20 of the sites visited were involved in an employee- driven BBS process that was in place for at least one year. Each BBS process included interper- sonal observation and feedback with a checklist of specific safe and at-risk behaviors.

Measures

The culture survey administered in the current study used a combination of scales found in pre- vious research, plus scales and items designed specifically for this investigation. Scale items were presented in a 7-point Likert format.

Interpersonal trust. Interpersonal trust was measured with the 12-

item scale developed by Cook and Wall (1980). The interpersonal trust scale accesses employee levels of trust with management intentions and ability, and coworker intentions and ability. Cronbach Alpha levels for the interpersonal trust scales have been documented at over 0.80.

Impulsivity. Impulsivity is attributed to people who “act

on the spur of the moment without being aware of any consequences or risk involved” (Clift, Wilkins, & Davidson, 1993, p. 404). Nine items from the Eysenck, Pearson, Easting, and Allsopp (1985) L7 scale were used to assess impulsivity. This shortened scale has been shown to have ac- ceptable inter-item reliability estimates of 0.77 (Clift et al., 1993).

Involvement. The involvement scale was designed specifi-

cally for the current investigation. The tree items that comprise the scale ask participants to rate their level of current and future involvement in the BBS process.

Perceptions of BBS training. The perceptions of BBS training items were

also constructed specifically for this study. These three items ask participants to rate the BBS train- ing they received with regard to relevance, im- portance, and quality.

Single item scales. A number of single-item measures asked re-

spondents whether or not they had received BBS training, whether or not they had been trained to give behavior-based feedback, how often and how much they gave and received positive and negative feedback, and how satisfied they were with the BBS process.

Procedures

During scheduled site visits to all 20 participat- ing organizations, data were collected by two re- searchers via two distinct methods: focus group discussions and perception surveys. One focus group involved only members of the safety steer- ing committee, and the other involved a random selection of hourly employees. While this should have resulted in 40 focus-group sessions, nine of the sites visited did not have an intact safety committee. Therefore, a total of 31 focus groups were performed (20 with hourly employees and 11 with members of the safety committee). Focus groups ranged in size from 4 to 22.

The focus-group sessions lasted approximately 90 minutes, during which one of two researchers asked a series of six questions designed to solicit employees’ opinions regarding the necessary in- gredients for an effective employee-driven BBS process. All reactions to these questions were re- corded simultaneously and independently on data collection sheets by both researchers.

After completing both focus groups, research- ers left a perception survey with each organiza- tion. The contact person was instructed to get as many employees as possible to complete these surveys. When it was not possible to survey all employees, the contact person was asked to ob- tain a representative sample of every work area. When completed, the surveys were mailed to the research team for analysis.

RESULTS

The dimensions measured by the survey as well as their scale alpha levels are displayed in Table 1.

Focus Groups

Two steps were used to analyze information gathered during focus groups. First, researchers looked at the employee responses for all ques- tions as recorded by both focus-group facilita-

Journal of Safety Research 240

Table 1. Variables Measured by the Perception Survey

Number of Items

Comprising Measure Scale Alpha

Position (management vs. front-line) 1 — Impulsivity (Clift, Wilkins, & Davidson, 1986) 9 0.78 Perception safety issues are included in performance appraisals 1 — Trust in management intentions (Cook & Wall, 1980) 3 0.64 Trust in peer intentions (Cook & Wall, 1980) 3 0.80 Satisfaction with the BBS process 1 — Amount of positive feedback given related to BBS process 1 — Amount of positive feedback received related to BBS process 1 — Amount of negative feedback given related to BBS process 1 — Amount of negative feedback received related to BBS process 1 — Whether or not trained to give feedback (Yes or No) 1 — Tenure with the organization 1 — Perceptions of BBS training 3 0.87 Trust in peer abilities (Cook & Wall, 1980) 3 0.63 Trust in management abilities (Cook & Wall, 1980) 3 0.75 Involvement in the BBS process 3 0.78 Whether or not educated in BBS (Yes or No) 1 —

tors. Only those responses that appeared on both of the data sheets were retained for analysis. Sec- ond, for the first five questions, each response was classified as a positive, negative, or neutral statement by two researchers. Neither researcher was aware of how the other was classifying any of the responses. Only if a statement was classi- fied as positive, negative, or neutral by both re- searchers was it retained.

A total of 12 responses were eliminated as a result of this process. For questions 6–8, re- sponses were categorized as a behavior-based factor, a person-based factor, or an environment- based factor. This classification was performed by two subject-matter experts. Consensus was reached on each item. Results of the content analysis for questions 1–8 can be found in the Appendix.

Perception Survey Results

Of the 756 perception surveys returned 701 were acceptable for analysis. These surveys were re- turned from 15 of the 20 participating organiza- tions. The employees at four organizations failed to complete and return the surveys, and the sur- veys returned from another organization were completed incorrectly. Table 2 displays the inter- correlations for all variables measured by the perception survey.

Predicting involvement in the behavior-based process.

A forward entry regression analysis was per- formed to determine variables most predictive of involvement in the safety process. Employee in- volvement was determined for each employee by summing their 7-point Likert scale responses to the following three involvement-related items: 1) Up to this point I have been very active in this or- ganization’s behavior-based safety program; 2) I willingly attend and participate in most of the be- havior-based safety related activities and meet- ings that occur in this organization; and 3) It is my intention to remain very active in this organi- zation’s behavior-based safety program.

As shown in Table 3, five variables contrib- uted significantly to predicting self-reported lev- els of involvement, namely: (a) adequacy of the BBS training; (b) trust in management abilities; (c) extent that safety is used in performance ap- praisals; (d) whether or not the employee was ed- ucated in the BBS process; and (e) tenure with organization. These variables accounted for 41% of the variance in amount of self-reported in- volvement in the BBS process.

Voluntary vs. mandatory BBS

Table 4 depicts the results of planned comparison t-tests for the perception survey variables be-

Winter 1999/Volume 30/Number 4 241

Table 2. Intercorrelations for All Dependent Variables

Variable 1 2 3 4 5 6 7 8 9 10

1. Position 1.00 2.07 .17* .15* 2.07 2.16* 2.06 2.05 .24** 2.07 2. Years with

company 1.00 2.02 2.04 2.03 2.04 .06 .11 .09 .09 3. Educated in b-b

safety 1.00 .54** .08 .04 .11 .06 .39** .07 4. Trained to observe 1.00 .09 .12 .09 .02 .46** .07 5. Observations

performed 1.00 .69** .22** .09 2.08 .22** 6. Frequency

observed 1.00 .32** .14 2.08 .35** 7. Frequency positive

feedback received 1.00 .65** .14 .78** 8. Frequency negative

feedback received 1.00 .16* .57** 9. Trained to give

feedback 1.00 .17* 10. Frequency positive

feedback given 1.00 11 12 13 14 15 16 17 18 19 20

11. Frequency negative feedback given 1.00 2.02 .04 .10 2.05 2.02 .10 .10 .09 2.11

12. BBS part of performance evaluations 1.00 2.01 .23** .17* .14* .14* .19** .25** .15

13. Impulsivity 1.00 2.17* 2.20** 2.22** 2.32** 2.17* 2.17* 2.16* 14. Perceptions of BBS

training 1.00 .41** .50** .39** .31** .59** .28** 15. Trust in

management intent 1.00 .67** .39** .37** .42** .18** 16. Trust in

management ability 1.00 .30** .33** .46** .26** 17. Trust in peer intent 1.00 .65** .30** .24** 18. Trust in peer ability 1.00 .26** .28** 19. Involvement in

process 1.00 .28** 20. Process satisfaction 1.00

1 2 3 4 5 6 7 8 9 10

11. Frequency negative feedback given .07 .02 2.08 2.04 .19** .29** .30** .43** .08 .60**

12. BBS part of perfor- mance evaluations 2.01 2.13 .02 2.01 .07 .03 2.06 2.02 2.03 2.04

13. Impulsivity 2.08 2.15* 2.08 2.07 2.04 2.05 2.08 2.02 2.09 2.07 14. Perceptions of BBS

training 2.01 .10 .32** .27** .02 .09 .18* .13 .25** .20** 15. Trust in

management intent .30** 2.16* .28** .32** 2.02 2.10 .08 .03 .38** .09 16. Trust in

management ability .15** 2.10 .31** .31** .00 2.03 .10 2.01 .37** .11 17. Trust in peer intent .03 2.01 .06 .06 2.13 2.05 .09 .00 .05 .14* 18. Trust in peer ability .08 2.02 .09 .10 2.08 2.10 2.04 2.17* 2.04 2.07 19. Involvement in

process .06 .11 .40** .33** .09 .08 .17* .14* .31** .22** 20. Process satisfaction 2.07 2.06 .16* .08 2.03 2.02 .01 2.08 2.13 2.01

* p , .05; ** p , .01.

Journal of Safety Research 242

Table 3. Predicting Employee Involvement in a BBS Process

Std Error of Variable(s) Entered R R2 Adjusted R2 the Estimate

1. Adequacy of BBS training 0.54 0.29 0.29 0.54 2. Trust in management abilities 0.59 0.35 0.34 0.51 3. Extent that BBS is used in performance appraisals 0.61 0.37 0.36 0.50 4. Whether or not educated in BBS 0.63 0.39 0.38 0.50 5. Tenure with organization 0.64 0.41 0.40 0.49

tween employees in a mandatory (n 5 273) and employees in a voluntary (n 5 427) BBS pro- cess. As shown in this table, employees in a man- datory BBS process showed higher scores on several questions related to BBS success than did employees in a voluntary BBS process. More specifically, employees in a mandatory process reported significantly higher rates for giving and receiving positive behavior-based feedback and significantly lower rates for receiving negative behavior-based feedback. In addition, employees in a mandatory BBS process demonstrated sig- nificantly greater levels of trust in management (both abilities and intentions), trust in coworkers (both abilities and intentions), and overall satis- faction with the BBS training received. Obvi- ously, our hypothesis that voluntary participation in BBS would be superior to mandatory partici- pation was not supported.

Table 4. Voluntary versus Mandatory BBS Processes

DISCUSSION

The tremendous improvements in safety and per- formance of companies who have implemented BBS processes has given this approach to safety management credibility and status (Geller, 1999). Unfortunately, little objective research has been performed to elucidate the organizational factors that can facilitate successful implementation of a BBS approach. Instead, many organizations are left to muddle through the unending case studies provided by safety consultants. Although such literature may be enlightening and informative to some degree, it does not adequately inform read- ers of the underlying behavioral (participation), interpersonal (trust), and organizational (man- agement support) processes that lead to success- ful BBS implementation. The current empirical investigation is a first step in understanding what

Voluntary Mandatory Variable BBS Process BBS Process

1. Average frequency of observation made per employee per month 6.5 7.6 2. Average frequency of time observed per employee per month 4.0 6.6* 3. Average frequency positive feedback received per employee per month 1.8 3.6** 4. Average frequency negative feedback received per employee per month 0.6 0.6 5. Average frequency positive feedback given per employee per month 3.1 5.3* 6. Average frequency negative feedback given per employee per month 1.2 1.5 7. Degree of involvement 3.4 3.8* 8. Amount of satisfaction 3.1 3.3 9. Trust in management abilities 3.8 4.2*

10. Trust in management intentions 3.7 4.1* 11. Trust in peer abilities 3.4 3.7* 12. Trust in peer intentions 3.7 4.0* 13. Impulsivity 2.5 2.5 14. BBS is part of performance evaluations 3.3 3.4 15. Satisfaction with BBS Training 3.5 3.8*

* Significant at p , .05; ** significant at p , .01.

Winter 1999/Volume 30/Number 4 243

factors are critical for successful implementation of BBS.

A primary method used in virtually every or- ganization that has reduced injuries with BBS is observation and feedback. One of the first prob- lems encountered by an organization is deciding if employee participation in the observation and feedback process should be mandatory or volun- tary. The obvious concerns associated with mak- ing such a process mandatory is that it will lead to negative reactions from the employees due to reduced levels of choice or perceived control (Geller, 1998a,b). On the other hand, a voluntary observation and feedback process could lead to minimal employee participation.

The current research enabled us to compare organizations incorporating a voluntary observa- tion and feedback process with organizations opting for mandatory interpersonal observation and feedback. Contrary to our research hypothe- sis, voluntary observation and feedback pro- cesses were not viewed any more favorably than were mandatory observation and feedback pro- cesses. In fact, the results for positive regard to- ward an observation and feedback process seemed to favor the mandatory approach. In ad- dition, findings summarized in Table 4 indicate organizations taking the mandatory approach may experience benefits in a number of areas over organizations following an entirely volun- tary approach.

At first these results seem to defy logic or common sense. However, upon closer examina- tion of the mandatory approaches, such findings are not too surprising. First, although employees in a mandatory process had no choice in whether or not they participated, they usually had a good deal of choice regarding when and how they con- ducted observation and feedback sessions. For example, all of the organizations using a manda- tory approach gave employees choice regarding how many times a month they should perform an observation. And all observations occurred with permission from the person being observed. Some of these organizations (37%) rotated ob- servers on a monthly basis. Thus, it seems these organizations were able to circumvent reductions in employee perceptions of personal control by building employee choice into the BBS process.

Besides reporting higher levels of employee involvement, mandatory processes also demon- strated significantly higher levels of trust in man- agement and trust in coworkers, a key ingredient for BBS success (as reported in most focus

groups). Plus, the use of positive BBS feedback was reported more frequently for mandatory than voluntary programs. It’s likely such results are due to the fact that mandatory programs lead to more overall involvement. And as involvement in the observation and feedback process in- creased, more employees increased their trust in the process and realized it was designed to bene- fit them. Moreover, they probably found out that participation did not lead to negative conse- quences such as being “ratted-out” by a coworker or receiving a reprimand from a supervisor.

More experience with interpersonal observa- tion and feedback results in more trust in the in- tentions and abilities of coworkers, and then to more involvement. While these results should not be interpreted as a blanket endorsement for making BBS mandatory in all organizations, it does indicate that mandatory processes are not necessarily detrimental to employee satisfaction, trust, or perceptions of personal control. If insti- tuted properly, mandatory processes can actually facilitate employee perceptions of personal con- trol as well as increase a number of relationship- based variables beneficial to BBS success. Thus, it seems beneficial for management to set an ex- pectation that everyone participate in the obser- vation and feedback process they’ve been trained to implement. But management should then give plenty of leeway regarding how the process oc- curs, and offer ongoing support for the proce- dures customized by the line workers. Beyond the observation and feedback process, the current investigation offers insight regarding a number of additional issues relevant to BBS. These re- sults are particularly important because they come from the perspective of line workers or op- erators. For instance, our data support the notion that employees generally understand manage- ment and supervisors are key determinants of the success or failure of a BBS process. While there was disagreement regarding the extent to which management and supervisors should be involved, everyone acknowledged that these individuals needed to demonstrate visible support of an inter- personal observation and feedback process.

When negative reactions about BBS were given, significant attention was given to percep- tions of the BBS steering committee. In such cases, the steering committee was viewed as be- ing out of touch with what is really needed to im- prove safety in the organization. It was also com- mon to find that a large number of employees perceived safety committee members as simply

Journal of Safety Research 244

ingratiating individuals looking to get ahead. This emphasizes the need to select the right peo- ple to serve on the BBS steering committee, and to rotate membership at regular intervals.

The steering committee is responsible for driving an observation and feedback process, and negative perceptions of its members cannot be beneficial. Further research should investigate the type of involvement required by BBS steer- ing committees from the beginning stages of BBS implementation to eventual institutionaliza- tion. For example, active and frequent involve- ment of a BBS steering committee is clearly needed to get a BBS process started, but it might be advantageous to fade out certain kinds of committee activity as BBS principles and proce- dures take hold throughout an organization. In other words, the role of the BBS steering com- mittee needs to change as BBS processes mature. It is not clear, however, what roles should remain and how others should change.

Three of the questions asked of employees during focus groups were particularly informa- tive regarding ways to implement and maintain a successful BBS process. Specifically, by asking employees: (a) how they would improve their BBS process; (b) what were the biggest obstacles to BBS success; and (c) what are the key ingredi- ents to a BBS process, we learned useful per- spectives from the line workers—the employees most directly responsible for BBS and the orga- nization’s safety record. Specifically, our content analysis of employees’ answers to these ques- tions revealed the following noteworthy issues.

First, although it may be impressive to collect a lot of behavioral data from behavioral observa- tions, it is probably in an organization’s best in- terest to start “small” with a simple behavioral checklist. Allow employees time to get used to an observation and feedback process before add- ing too much complexity. Second, it is impera- tive that trainers make clear BBS cannot be a ve- hicle for blaming individuals, or for delivering criticism of any kind. Third, interpersonal trust is essential, from the shop floor to the management boardroom. Trust was mentioned frequently not only as one of the critical factors for success, but also as one of the greatest barriers when it’s not there.

As with interpersonal trust, employees ex- pressed repeatedly the need for continued sup- port from management. While monetary re- sources are essential, they are not sufficient. Visible recognition that management appreciates

the BBS methods is critical. One way to do this is to provide necessary time for employees to per- form behavioral observations and analyze results. And if the outcome from a BBS process indicates a need for certain changes in environmental con- ditions or training procedures, management needs to follow through quickly and effectively.

Other critical success factors for BBS success were revealed through the multiple regression analysis. Most important in getting employee in- volvement was BBS training. These sessions usually represent employees’ first exposure to BBS, and if they are not made relevant to the em- ployees’ work setting and circumstances, em- ployees could actually leave a training session discouraged, confused, and unlikely to get ac- tively involved. Research is needed to define the parameters of a BBS training session which facil- itate subsequent employee participation. Individ- ual characteristics of the trainer are probably rel- evant, as well as the structure and format of the training itself.

Our regression analysis showed that a number of other variables can also help to facilitate em- ployee involvement in a BBS process. As ex- pected from the focus group results, trust in man- agement added significant predictive power to our regression equation. However, we found trust in management abilities and not intentions to be the crucial factor. Perhaps employees believe management is well intentioned when they start the safety process, but if management does not follow through with appropriate support, trust in ability is suspect. So trust in ability means man- agement knows how to facilitate and support BBS. This explanation is consistent with the ma- jority opinion from focus groups that visible management support is essential for a successful BBS process.

In conclusion, we offer a framework for under- standing why some BBS processes succeed and other fail. Analogous to the way we analyzed the focus group discussions, we propose the BBS Safety Success Triad illustrated in Figure 1. The three sides of this triangle are dynamic and interac- tive. When one changes the others are influenced.

The BBS Safety Success Triad summarizes the discussion from 31 focus groups, as well as the analysis of 701 perception surveys (from 15 different organizations). On the person side of the triangle is interpersonal trust; on the environ- ment side of the triangle is management support; and on the behavior side of the triangle is em- ployee participation/involvement. Training is in

Winter 1999/Volume 30/Number 4 245

FIGURE 1 The behavior-based safety success triad.

the middle of the triangle because of its critical role in facilitating all three sides of the Triad. During training all employees receive the princi- ples, procedures, and tools of a BBS process, and management receives the rationale and the method for supporting BBS. Proper training con- vinces participants the process works and the people can implement it. This is necessary for self-efficacy (Bandura, 1997), a person state deemed necessary for any constructive long-term change among individuals in a work culture.

REFERENCES

Baer, D.M., Wolf, M.M., & Risley, T. (1968). Some current dimensions of applied behavior analysis. Journal of Ap- plied Behavior Analysis, 1, 91–97.

Baker, S.P., Conroy, C., & Johnston, J.J. (1992). Occupa- tional injury prevention. Journal of Safety Research, 23(2), 129–133.

Bandura, A. (1997). Self-efficacy: The exercise of control. New York: W.H. Freeman and Company.

Clift, S.M., Wilkins, J.C., & Davidson, E.A.F. (1993). Im- pulsiveness, venturesomeness and sexual risk-taking among heterosexual gum clinic attenders. Personality and Individual Differences, 15(4), 403–410.

Cook, J., & Wall, T. (1980). New work attitude measures of trust, organizational commitment and personal need non-fulfill- ment. Journal of Occupational Psychology, 53, 39–52.

Daniels, A.C. (1989). Performance management. Tucker, GA: Performance Management Publications.

Dose, J.J., & Klimoski, R.J. (1995). Doing the right thing in the workplace: Responsibility in the face of accountability. Employee Responsibilities and Rights Journal, 8(1), 35–56.

Eysenck, S.B.J., Pearson, P.R., Easting, G., & Allsopp, J.F. (1985). Age norms for impulsiveness, venturesomeness

and empathy in adults. Personality and Individual Differ- ences, 6, 613–619.

Geller, E.S. (1984). Motivating safety belt use with incen- tives: A critical review of the past and a look to the fu- ture. In Advances in belt restraint systems: Design, per- formance and usage (p. 127–142). Warrendale, PA: Society of Automotive Engineers.

Geller, E.S. (1996). The psychology of safety: How to im- prove behaviors and attitudes on the job. Radnor, PA: Chilton Book Company.

Geller, E.S. (1998a). Beyond safety accountability: How to increase personal responsibility. Neenah, WI: J.J. Keller & Associates, Inc.

Geller, E.S. (1998b). Understanding behavior-based safety: Step-by-step methods to improve your workplace. 2nd. Ed. Neenah, WI: J.J. Keller & Associates, Inc.

Geller, E.S. (1998c). What’s the level of trust in your work- place? Industrial Safety & Hygiene News, 32(3), 14–15.

Geller, E.S. (1999). Behavior-based safety: Confusion, con- troversy, and clarification. Occupational Health & Safety, 68(1), 40–50.

Geller, E.S., & Hahn, H.A. (1984). Promoting safety belt use at industrial sites: An effective program for blue-col- lar employees. Professional Psychology: Research and Practice, 15(4), 553–564.

Geller, E.S., Lehman, G.R., & Kalsher, M.J. (1989). Behav- ior analysis training for occupational safety. Newport, VA: Make-A-Difference, Inc.

Geller, E.S., Rudd, J.R., Kalsher, M.J., Streff, F.M., & Leh- man, G.R. (1987). Employee-based programs to motivate safety belt use: A review of short and long-term effects. Journal of Safety Research, 18, 1–17.

Granovetter, M. (1985). Economic action and social struc- ture: The problem of embeddedness. American Journal of Sociology, 91, 481–510.

Grant, R.A., Higgins, C.A., & Irving, R.H. (1988). Comput- erized performance monitors: Are they costing you cus- tomers? Sloan Management Review, Spring, 39–45.

Hackman, R.J., & Lawler, E.E. (1971). Employee reactions to job characteristics. Journal of Applied Psychology, 55(3), 259–286.

Journal of Safety Research 246

Hart, K.M. (1988). A requisite for employee trust: Leader- ship. Psychology: A Journal of Human Behavior, 25, 1–7.

Kello, J.E., Geller, E.S., Rice, J.C., & Bryant, S.L. (1988). Motivating auto safety belt wearing in industrial settings: From awareness to behavior change. Journal of Organi- zational Behavior Management, 9(2), 7–21.

Kelman, H.C. (1958). Compliance, identification, and inter- nalization: Three processes of attitude change. Conflict Resolution, 2, 51–60.

Leigh, J. (1995). Causes of death in the workplace. West- port, CN: Quorum Books.

Ludwig, T.D., & Geller, E.S. (1991). Improving the driving practices of pizza deliverers: Response generalization and moderating effects of driving history. Journal of Ap- plied Behavior Analysis, 24, 31–34.

Ludwig, T.D., & Geller, E.S. (1997). Assigned versus par- ticipative goal setting and response generalization: Man- aging injury control among professional pizza deliverers. Journal of Applied Psychology, 82(2), 253–261.

McAllister, D.J. (1995). Affect- and cognition-based trust as foundations for interpersonal cooperation in organiza- tions. Academy of Management Journal, 38(1), 24–59.

Miller, T.R. (1997). Estimating the costs of injury to U.S. employers. Journal of Safety Research, 28(1), 1–13.

National Committee for Injury Prevention and Control. (1989). Injury prevention: Meeting the challenge. New York: Oxford University Press.

National Institute for Occupational Safety & Health (1998). Traumatic occupational injury research needs and prior- ities: A report by the NORA traumatic injury team. Spe- cial Research Report, Cincinnati, OH: DHHS (NIOSH) Publication No. (PHS).

National Safety Council. (1998). Accident facts. Itasca, IL: National Safety Council.

Samuelson, C.D., Messick, D.M., Rutte, C.G., & Wilke, H. (1984). Individual and structural solutions to resource di- lemmas in two cultures. Journal of Personality and So- cial Psychology, 37, 139–155.

Sashkin, M. (1984). Participative management is an ethical imperative. Organizational Dynamics, 18(1), 5–22.

Seligman, M.E.P. (1975). Helplessness: On depression de- velopment and death. San Francisco: Freedman.

Streff, F.M., Kalsher, M.J., & Geller, E.S. (1993). Develop- ing efficient workplace safety programs: Observations of response covariation. Journal of Organizational Behav- ior Management, 13(2), 3–15.

Sulzer-Azaroff, B., & De Santamaria, M.C. (1980). Indus- trial safety hazard reduction through performance feed- back. Journal of Applied Behavior Analysis, 13, 287– 295.

United States Bureau of Labor Statistics. (1997). Safety and health statistics. Occupational safety and health home page. url:www.bls.gov

United States Department of Health and Human Services (1998). Traumatic occupational injury research needs and priorities: A report by the NORA traumatic injury team. Atlanta, GA: National Institute for Occupational Safety and Health, Centers for Disease Control and Pre- vention.

Van Houten, R., & Nau, P.A. (1983). Feedback interventions and driving speed: A parametric and comparative analysis. Journal of Applied Behavior Analysis, 16, 253–281.

Yamagishi, T., & Cook, K.S. (1993). Generalized exchange and social dilemmas. Social Psychology Quarterly, 56(4), 235–248.

Zand, D.E. (1972). Trust and managerial problem solving. Administrative Science Quarterly, 17, 229–239.

Winter 1999/Volume 30/Number 4 247

Appendix Focus Group Questions and Results

Focus Group Question 1: How do you feel about the observation and feedback process used in your behavior-based safety process? (104 total responses—56 positive, 37 negative, 11 neutral) Most frequent positive comments Comments Focus groups Organizations –It increases awareness of safe and at-risk behaviors. 17 15 14 –It facilitates positive attitudes. 8 6 5 –It increases accountability. 3 3 2 Most frequent negative comments Comments Focus groups Organizations –It is perceived that ulterior motives are driving the process (e.g., “It is done to blame the workers.”). 15 9 8 –It is inconvenient. 9 8 6 –Participating employees are only interested in the overtime. 2 2 2

Focus Group Question 2: Is participation in the observation and feedback process mandatory or voluntary?

Organizations with a voluntary process 5 12 Correlation between type of process and Organizations with a mandatory process 5 8 positive regard 5 20.304

Focus Group Question 3: What role should management play in your behavior-based safety process? (50 total responses—23 positive, 15 negative, 12 neutral)

Most frequent positive comment Comments Focus groups Organizations –They should be involved, the more the better 6 6 6 Most frequent negative comment Comments Focus groups Organizations –The less they are involved the better 8 8 8

Focus Group Question 4: Do supervisors influence success of the behavior-based safety process?

–12 organizations had a position analogous to a front-line supervisor –11 of those organizations (22 focus groups) commented that a behavior-based safety process will not succeed unless the front line supervisors support it

Focus Group Question 5: What are your perceptions of the behavior-based safety steering committee? (87 total responses—33 positive, 36 negative, 18 neutral)

Most frequent positive comments Comments Focus groups Organizations –I think they’re doing a good job 9 9 8 –They are well intentioned and serious 7 7 7 Most frequent negative comments Comments Focus groups Organizations –They are “out of touch”—spend too much time in meetings 8 7 6 –They are “safety rats” 4 4 3 –The meetings are causing morale and production problems 4 3 3

Focus Group Question 6: How would you improve your behavior-based safety process?

Environment-based (n 5 72) Comments Focus groups Organizations –Use more and better incentives 11 11 10 –Provide adequate training 11 9 6 –Increase quality of machines 9 7 6 –Simplify the checklist 6 6 6 Behavior-based (n 5 23) –Increase employee participation 8 8 7 –Increase use of positive feedback 2 2 2 –Fire at-risk workers 2 2 2 Person-based (n 5 8) –Have a positive attitude 8 8 6

(Continued)

Journal of Safety Research 248

Appendix Continued

Focus Group Question 7: What are the biggest obstacles your behavior-based safety process has faced?

Environment-based (n 5 67) Comments Focus groups Organizations –Lack of management support 22 20 18 –Previous failed safety efforts 10 10 10 –Extra paperwork is aversive 6 6 6 Person-based (n 5 47) –Lack of trust 12 12 12 –Discomfort with being observed 8 8 8 –Lack of buy-in to the process 5 5 5 Behavior-based (n 5 19) –Lack of participation 5 4 4 –Poor communication 3 3 3 –Negative feedback used too frequently 3 2 2

Focus Group Question Number 8: What are the key ingredients for success in a behavior-based safety process?

Person-based (n 5 72) Comments Focus groups Organizations –Trust 13 13 13 –Positive attitudes 5 5 5 –An open mind 5 5 4 Environment-based (n 5 53) –Management support 18 18 18 –Proper training 5 5 5 –The process must be voluntary 4 4 4 Behavior-based (n 5 42) –Employee participation 11 11 11 –Teamwork 6 6 6 –Open communication 5 5 4

Winter 1999/Volume 30/Number 4 249