Development of safety incident coding systems through improving coding reliability

Olsen, Nikki S.; Williamson, Ann

doi:10.1016/j.apergo.2015.04.015

Cited by 6 publications

(2 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Two trained coders coded 19% of the data for 100% of the 13 instruction codes (including the first-level codes e.g., HO-DI missed and second-level codes e.g., distraction ) and 90% of the 19 teaching approach codes (e.g., encouragement ). Percentage agreement was calculated according to the Index of Concordance ( 35 ): Total number of agreements/(Total number of agreements + Total number of disagreements) ( 36 ). Agreement was calculated at 88.89% indicating near perfect reliability ( 37 ).…”

Section: Methodsmentioning

confidence: 99%

Exploring the Dimensions of Driving Instruction through Naturalistic Observation of Formal Practical Lessons with Learner Drivers

Watson-Brown

Scott‐Parker

Simons‐Morton

et al. 2020

Transportation Research Record: Journal of the Transportation R

View full text Add to dashboard Cite

Higher-order driving skills (HO-DS) are deficient in young drivers who are over-represented in road crash fatalities and serious injuries. Teaching HO-DS has strong theoretical support in reducing crash risk. This study contributes to the dearth of literature regarding on-road driver training that can develop these skills. Higher-order driving instruction (HO-DI) is explored in formal on-road driving lessons (in Queensland, Australia) via naturalistic observation. Fifteen instructors and 96 learner drivers aged 16–19 years were recruited, with 110 lessons observed. An HO-DI coding taxonomy informed by the Goals for Driver Education ( 24) was used for content analysis using an a priori approach, comprising eight HO-DI codes: driving plan (formerly driving route), vehicle control and maneuvering, mastery of traffic situations, surveillance, situational risk (formerly environmental variables), personal risk (formerly knowledge of risk), car function (formerly car knowledge), and distraction. Thirty-nine sub-themes were identified within the codes allowing a deeper understanding of instruction, including missed (unseen) and untaken (seen but not actioned) HO-DI opportunities. The findings inform recommendations for the development of best practice HO-DI. This study has implications for the driver training industry with potential for reducing the crash risk of young novice drivers.

show abstract

Section: Methodsmentioning

confidence: 99%

Exploring the Dimensions of Driving Instruction through Naturalistic Observation of Formal Practical Lessons with Learner Drivers

Watson-Brown

Scott‐Parker

Simons‐Morton

et al. 2020

Transportation Research Record: Journal of the Transportation R

View full text Add to dashboard Cite

show abstract

“…The HFACS framework is underpinned by the SCM mark I, though the layers from SCM mark I have been modified (Reason et al 2006), reducing the levels/layers from four to three in the SCM mark III. Modification of the HFACS framework base on the SCM mark III could result in a reduction in the size of the coding system, which is acknowledged as a key factor to attaining high reliability (Olsen and Williamson, 2015;Olsen and Williamson, 2017). Based on these reasons, a modified HFACS framework, HFACS-GMI, is proposed for the Ghana Mining Industry.…”

Section: Hfacs and Its Derivativesmentioning

confidence: 99%

Human Factor Analysis Framework for Ghana’s Mining Industry

Joe-Asare¹,

Amegbey²,

Stemn³

2020

View full text Add to dashboard Cite

In an attempt to incorporate human factors into technical failures as accident causal factors, researchers have promoted the concept of human factor analysis. Human factor analysis models seek to identify latent conditions within the system that influence the operator’s action to trigger an accident. For an effective application of human factor analysis models, a domain-specific model is recommended. Most existing models are developed with category/subcategory peculiar to a particular domain. This presents challenges and hinders effective application outside the domain developed for. This paper sought to propose a human factor analysis framework for Ghana’s mining industry. A comparative study was carried out between three dominated accident causation models and investigation methods in literature; AcciMap, HFACS, and STAMP. The comparative assessment showed that HFACS is suitable for incident data analysis based on the following reason; ease of learning and use, suitability for multiple incident analysis and statistical quantification of trends and patterns, and high inter and intra-coder reliability. A thorough study was done on HFACS and its derivative. Based on recommendations and research findings on HFACS from literature, Human Factor Analysis, and Classification System – Ghana Mining Industry (HFACS-GMI) was proposed. The HFACS-GMI has 4 tiers, namely; External influence/factor, Organisational factor, Local Workplace/Individual Condition and, Unsafe Act. A partial list of causal factors under each tier was generated to serve as a guide during incident coding and investigation. The HFACS-GMI consists of 18 subcategories and these have been discussed. The HFACS-GMI is specific to the Ghanaian Mines and could potentially help in identifying causal and contributing factors of an accident during an incident investigation and data analysis. Keywords: Human Factor Analysis, Causal Factor, Causation Model, Mining Industry

show abstract

Application of classification principles to improve the reliability of incident classification systems: A test case using HFACS-ADF

Olsen

Williamson

2017

Applied Ergonomics

View full text Add to dashboard Cite

Development of safety incident coding systems through improving coding reliability

Cited by 6 publications

References 32 publications

Exploring the Dimensions of Driving Instruction through Naturalistic Observation of Formal Practical Lessons with Learner Drivers

Exploring the Dimensions of Driving Instruction through Naturalistic Observation of Formal Practical Lessons with Learner Drivers

Human Factor Analysis Framework for Ghana’s Mining Industry

Application of classification principles to improve the reliability of incident classification systems: A test case using HFACS-ADF

Contact Info

Product

Resources

About