Workplace Safety

Individual Differences in Behavior

Alice F Stuhlmacher,Douglas F Cellar

• Publisher: Routledge
• Year: 2012
• ISBN:
  9781136615009
• Edition: 1
• Topic: Business
• Subtopic: Business generale
• Index: Business

Do all you can to minimize dangerous behaviors to benefit communities, employees, and organizations!Safety is a "real world" problem that community psychologists, industrial/organizational psychologists, industrial hygenists, human resources professionals, and corporate insurance groups must deal with on a day-to-day basis. In Workplace Safety: Individual Differences in Behavior you will examine safety behavior and discover practical interventions to help increase the safety awareness of the people in your life. This book takes a look at ways of defining and measuring safety as well as a variety of individual differences like gender, job knowledge, conscientiousness, self-efficacy, risk avoidance, and stress tolerance that are important in creating safety interventions and improving the selection and training of employees.Workplace safety is of prime importance in today's increasingly litigious society. It has been estimated that each year in the United States, there are 100, 000 work-related accident or disease fatalities, 400, 000 workers who become disabled, and 6 million workplace injuries. Of equal importance are driver safety and safety hazards in public spaces such as malls and individual stores. Workplace Safety: Individual Differences in Behavior examines:

• the importance of measurement in understanding worker abilities and defining safety behaviors
• the often-neglected issue of gender differences in safety definitions and research
• the relationship between personality variables, job, knowledge, and accident involvement
• the five-factor personality model for predicting safety behavior
• a model of safety consciousness
• types of safety hazards in public spaces
• monetary costs of accidents in malls and stores
• a practitioner's perspective on individual differences in safety behaviorWorkplace Safety: Individual Differences in Behavior takes an incisive look at these issues with a unique focus on the way individual differences in people impact safety behavior in the real world.

A Comprehensive Method for the Assessment of Industrial Injury Events

Carlla S. Smith

Gary S. Silverman

T. M. Heckert

M. H. Brodke

B. E. Hayes

M. K. Silverman

L. K. Mattimore

Bowling Green University

Summary. Because work-related injuries are infrequent and often poorly documented, injury event operationalizations beyond recorded rates would be beneficial. This study describes a method that uses self-reported and recorded events.

Researchers interviewed workers and obtained recorded events from personnel files to develop the near miss and unreported injury events measures. The self-reported event measures, with other safety variables and demographics, were then administered to two groups of plant workers (N = 115 and N = 120). Results indicated that self-reported events differed from recorded events and are related to other work injury variables (e.g., work hazards, overtime). An expanded safety protocol such as this one may provide additional tools to investigate the injury event process. [Article copies available for a fee from The Haworth Document Delivery Service: 1-800-342-9678. E-mail address: <http://[email protected]> Website: <http://www.HaworthPress.com>

Keywords. Injury events, recorded injury events, safety methods, self-reported injury events, work safety

Because industrial mishaps involving equipment damage and/or personal injury are relatively rare events (Chapanis, 1959), researchers are typically faced with methodological and statistical problems in collecting, analyzing, and interpreting these types of data. For example, the low base rate of industrial injuries per worker often necessitates collecting injury event data over long time periods to increase statistical power and using non-parametric statistical tests to compensate for the highly skewed data. Although statistical options such as non-parametric analyses and normalization formulae exist, these tests often lack sufficient statistical power to find significant effects (especially with small samples sizes) and/or transform the data into a form very different from the original data.

Beyond the straightforward statistical/methodological problems associated with collecting injury events and associated injuries, other, less apparent issues probably have an even greater impact on the accuracy of injury-related data. Under-reporting work conditions related to injuries is quite common. Researchers who studied Occupational Safety and Health Administration (OSHA) logs have found that work-related injuries and illnesses and lost work days are frequently not recorded (Oleinick et al., 1993), and only 75% of those organizations required to keep an OSHA log comply (Seligman, Sieber, Pederson, Sundin, & Frazier, 1988). In some situations, workers may not have reported work injuries to the worker compensation system because they received treatment from company-based or external insurance and disability programs (Fingar, Hopkins, & Nelson, 1992; Murphy, Sorock, Courtney, Webster, & Leamon, 1996).

Several studies have directly demonstrated the failure of OSHA and/or company-based reporting systems. For example, Weddle’s (1996) investigation of hospital environmental service workers (N = 372) showed that, of the workers who experienced injuries in the previous year (n = 108), 38.9% did not report their injuries. Weddle found that older and longer tenured workers were more likely to have not reported their injuries. The most commonly cited reason for not reporting was that the injury was minor; however, many of these injuries needed medical care (64.4%) or required lost work time (44.1%).

Pransky, Snyder, Dembe, and Himmelstein (1999) used a case study approach in multiple industries to examine the extent of under-reporting work-related injuries and the reasons behind them. Questionnaires and interviews were administered to 110 workers in similar jobs and their management at three industrial facilities. Less than 5% of the workers reported a work-related injury or illness within the previous years, although 50% or more experienced work-related symptoms or persistent problems, and 30% reported lost time or work restrictions because of a work-related injury or illness. Some of the reasons for which workers reported these discrepancies included fear of (management) reprisal, and lack of management responsiveness. Managers reported administrative barriers to accurate reporting, mostly to achieve pre-set safety goals. For these reasons, Pransky et al. (1999) recommended that, because of the widespread inaccuracy of recorded injury data, worker surveys and symptoms reports be used to provide more accurate and timely accounts.

The previous discussion has aptly demonstrated that many organizational reporting systems are replete with omissions and spurious information. This situation is of even greater concern because many safety training programs are based on data obtained from company records. For these reasons, operationalizations of industrial injury events beyond event rates from personnel records (recorded events) would be beneficial to safety researchers and practitioners. The present study presents the development and assessment (validity) of an industrial injury research method for the manufacturing industry that uses near (or near-miss) injury events and unreported injury events in addition to recorded injury events.

This research consisted of two phases. In the first phase, a small group of workers was interviewed with structured interviews to obtain near miss events. These data were used to construct a near-injury events self-report measure. Data on recorded events and resulting injuries were also collected from personnel records and interviews. This information was used to construct an unreported events and injuries self-report measure. In the second phase, the near events and unreported events measures, along with relevant demographic and validity (e.g., job safety and hazards) measures, were administered to two samples of industrial workers. Both the self-report measures and the recorded events data were used to obtain a more realistic and comprehensive picture of the injury event process in the two plants.

Method

Participants

Sample 1 (Plastics Plant): The final sample consisted of 115 shift-workers (48% of the total production workforce) across three permanent (fixed) shifts in a non-unionized plant that manufactures small plastic parts. Company tenure averaged seven years, nine months, and the average age of the workers was 37 years. Shiftworkers reported that they worked an average of 6.7 hours of overtime per week.

Sample 2 (Glass Plant): The final sample consisted of 120 shift-workers (25% of the total production workforce) on a weekly rotating shift cycle in a unionized glass manufacturing plant. Due to personnel cutbacks according to seniority, the production workforce was older, ranging from 43-62 years, with more than half being over 50. Tenure with the company ranged from eight to over 42 years, with 93% of the shiftworkers having over 24 years of tenure.

Multiple factors affected the response rate in the plants. Both plants suffered from poor labor-management relationships. However, the union-management relationship in the glass plant was particularly adversarial, reflecting the lower participation rate. Other factors, however, had an impact, such as scheduling difficulties, lack of communication between management and floor supervisors, and inadequate plant facilities for survey administration.

Phase 1: Injury Event Measures Development

Please note the term “injury event” is used here instead of the common term “accident” because “accident” connotes some level of personal blame, whereas “injury event” is a more generic term. Also, the focus of the present study is on these injury events, not the related injuries or shift differences in either. The injuries themselves (including shift differences in injuries) are the central topic in a related article by Smith et al. (1997).

Near-injury events. Near events are defined here as “a situation which could potentially result in injury or damage” (Chapanis, 1959). Approximately 25 dimensions that assess unsafe behaviors were used as core dimensions in the present study (see Tarrants, 1980). In addition to the core dimensions, near events data were gathered through the use of the critical incident technique (Flanagan, 1954). Using the critical incident technique, an interviewer asked a stratified sample of participant observers (workers) to recall and describe unsafe behaviors they had made or had observed in their work environment. A structured interview similar to the one proposed by Tarrants (1980) was used to gather these data.

The data reported in the critical incidents interviews were used to develop the near events measure for the next phase. Respondents were asked to indicate the number of different types of near events they had experienced during the past year, the shift they were on when each event occurred, and their best estimate of the likelihood that future injury might result from the near event. The one-year time span was adopted to minimize distortion of these subjective estimates from historical events. Open-ended questions about near events not covered in the questionnaire were also included. An example of selected items from the near events measure from one organization is provided in Table 1.

Table 1. Selected Items from Near Injury Events Questionnaire

Below is a list of almost or near accidents that have been reported by workers at plants like _____. A near accident is something that could have caused an injury but did not. Describe any near accidents that have happened to you. For each near accident, please tell us your best estimate of how often this near accident has happened to you in the past 12 months on each shift (1 = day, 2 = afternoon, and 3 = midnight shifts). Also, tell us how likely it would be that an injury could result from this near accident (using the list of letters below).

	INJURY LIKELIHOOD
	A = no chance of injury
	B = little chance of injury
	C = fair chance of injury
	D = great chance of injury
NEAR ACCIDENT
I almost had an accident when I adjusted a machine while its parts were still moving.
number of times on shift 1 _____ number of times on shift 2 _____ number of times on shift 3 _____	Injury likelihood. _____ (use letters above)
I almost had an accident when I operated a machine without a safety guard.
number of times on shift 1 _____ number of times on shift 2 _____ number of times on shift 3 _____	Injury likelihood. _____ (use letters above)
I almost had an accident when I slipped on something on the floor.
number of times on shift 1 _____ number of times on shift 2 _____ number of times on shift 3 _____	Injury likelihood. _____ (use letters above)
I almost had an accident when I participated in horseplay in the plant.
number of times on shift 1 _____ number of times on shift 2 _____ number of times on shift 3 _____	Injury likelihood. _____ (use letters above)
I almost had an accident when I lost control of the tool or equipment that I was using.
number of times on shift 1 _____ number of times on shift 2 _____ number of times on shift 3 _____	Injury likelihood. _____ (use letters above)

Recorded injury events. The incidence rate of recorded injury events was assessed. Our definition of injury events covered the spectrum from major events with serious personal injury (and/or extensive equipment/facility damage) to minor events with slight or no personal injury (and/or slight or no equipment/facility damage). Historical data were obtained from personnel records for the previous five years; this degree of aggregation is consistent with prior injury research (e.g., Levin, Oler, & Whiteside, 1985; Wojtczak-Jaroszowa, & Jarosz, 1987). The number of recorded injury events (total and by year) was calculated for each worker who participated in the study. Injury event rate was calculated by assessing the number of events within a given time period and by shift. The BLS statistics were not used here because they are unreliable for small workforces (i.e., exposures below 200,000 man hours for the BLS formu...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright
• Contents
• The Role of Individual Differences in Understanding and Predicting Workplace Safety
• A Comprehensive Method for the Assessment of Industrial Injury Events
• Gender Issues in the Measurement of Physical and Psychological Safety
• Predicting Motor Vehicle Crash Involvement from a Personality Measure and a Driving Knowledge Test
• The Five-Factor Model and Safety in the Workplace: Investigating the Relationships Between Personality and Accident Involvement
• Creating a Safer Working Environment Through Psychological Assessment: A Review of a Measure of Safety Consciousness
• Slips and Falls in Stores and Malls: Implications for Community-Based Injury Prevention
• Individual Differences in Safe Behavior: A Safety Practitioner’s Viewpoint
• Index