Traffic Locus of Control Scale

1. Abstract

The Traffic Locus of Control Scale (T-LOC) is a domain-specific psychometric instrument designed to assess drivers’ causal attributions and generalized expectancies regarding the determinants of traffic accidents. Originally conceptualized and validated by traffic psychologists Türker Özkan and Timo Lajunen at the University of Helsinki and Middle East Technical University, the instrument adapts Julian B. Rotter’s social learning theory and locus of control paradigm specifically to the automotive context. Comprising 17 items, the scale evaluates the degree to which an individual attributes accident involvement to internal or external agents. Psychometric investigations consistently reveal a robust multidimensional factorial structure, typically yielding four distinct dimensions: Self (internal attributions reflecting one’s driving skills, risk-taking, and active violations), Other Drivers (external-social attributions directed toward other road users’ misbehaviors), Vehicle and Environment (external-situational attributions concerning vehicular mechanical failure, adverse weather, poor illumination, and dangerous roadway infrastructure), and Fate (external-fatalistic attributions encompassing luck, destiny, and coincidence).

The T-LOC utilizes a 5-point Likert response scale, where respondents evaluate the degree of accident dependency for each specified factor. Psychometric evaluations across diverse national driving populations—including extensive cross-cultural validations in Finland, Turkey, Iran, China, Romania, and Greece—demonstrate satisfactory to strong internal consistency reliability, with subscale Cronbach’s alpha coefficients typically ranging from .65 to .86. The instrument exhibits robust convergent and criterion-related validity through established correlations with the Driver Behaviour Questionnaire (DBQ), self-reported moving violations, historical crash involvement, aberrant driving habits, risk perception, and driver attitudes toward advanced driver assistance systems (ADAS) such as Intelligent Speed Adaptation (ISA) and Adaptive Cruise Control (ACC). The T-LOC serves as an empirical tool for traffic safety researchers, automotive human factors engineers, transport psychologists, and driver education specialists aiming to understand cognitive biases in accident causation, develop individualized behavioral remediation, and advance traffic injury prevention paradigms globally.

2. Keywords

Traffic Locus of Control, T-LOC, Traffic Psychology, Attribution Theory, Driver Behavior, Road Safety, Accident Causation, Fatalism, Driver Risk-Taking, Psychometrics

3. Authors

The Traffic Locus of Control Scale was primarily conceptualized and validated by leading traffic and transportation psychologists based in Finland and Turkey:

• Türker Özkan, Ph.D.: Professor of Psychology and Director of the Safety Research Unit, Department of Psychology, Middle East Technical University (METU), Ankara, Turkey; formerly associated with the Traffic Research Unit, Department of Psychology, University of Helsinki, Finland. Email: [email protected] / [email protected].
• Timo Lajunen, Ph.D.: Professor of Safety and Transportation Psychology, Department of Psychology, Middle East Technical University (METU), Turkey, and Senior Researcher in the Human Factors and Safety Behaviour Group, Department of Psychology, University of Helsinki, Finland; also affiliated with the Department of Psychology, Norwegian University of Science and Technology (NTNU), Trondheim, Norway. Email: [email protected].
• Jyrki Kaistinen, M.Soc.Sc.: Human Factors Researcher and Cognitive Scientist, Human Factors and Safety Behaviour Group, Department of Psychology, University of Helsinki, Finland.

4. Purpose

The primary purpose of the Traffic Locus of Control Scale (T-LOC) is to quantify an individual’s cognitive orientation regarding the perceived control and causal attribution of motor vehicle collisions. While generalized locus of control inventories—such as Rotter’s Internal-External (I-E) Scale or Levenson’s Multidimensional Locus of Control Scale—measure overarching expectancies across broad life scenarios, psychometric literature consistently underscores that domain-specific measures offer dramatically superior predictive accuracy for behavioral outcomes within circumscribed environmental contexts. Driving represents an exceptionally high-risk, dynamic, multi-agent, semi-autonomous activity wherein split-second decisions interact with physical kinetics, engineered environments, and institutional regulations. Consequently, an individual’s generalized belief system often fails to account for their localized behavioral adaptations, hazard perceptions, and risk appraisals behind the wheel.

The T-LOC bridges this operational gap by measuring how motorists explain the etiology of road traffic accidents. By examining whether drivers view collisions as self-determined events resulting from behavioral choices or as uncontrollable occurrences driven by external forces, the scale addresses several theoretical and applied imperatives:

• Exploration of Cognitive Biases and Defensive Attributions: Human drivers universally exhibit an optimism bias, often referred to as the “illusory superiority” or “better-than-average” effect, wherein motorists overestimate their driving competencies while deprecating the skill and safety compliance of peers. The T-LOC operationalizes this phenomenon by isolating attributions assigned to the self versus attributions directed toward other road users.
• Prediction of Aberrant Driving Behaviors: Research reveals that drivers exhibiting an external traffic locus of control—specifically those attributing accidents to fate, adverse roadway conditions, or other drivers—engage in significantly higher frequencies of intentional traffic violations, dangerous overtaking, excessive speeding, and aggressive driving acts, as captured by the Driver Behaviour Questionnaire (DBQ). When motorists believe that accidents are fundamentally determined by fate or environmental adversity, their perceived self-efficacy for collision avoidance declines, undermining their motivation to execute precautionary driving tactics.
• Driver Education, Training, and Rehabilitation: In occupational fleet safety programs and court-mandated driver rehabilitation programs for chronic traffic violators, identifying driver attributional profiles allows educators to tailor remedial cognitive interventions. For instance, drivers scoring high on the Fate or Other Drivers subscales can be guided through attributional retraining therapies that foster an internal sense of agency, personal accountability, and defensive driving responsibility.
• Ergonomics and Human Factors in Vehicle Automation: Modern automotive safety relies heavily on Advanced Driver Assistance Systems (ADAS), such as Adaptive Cruise Control (ACC), Lane Keeping Assist (LKA), and Intelligent Speed Adaptation (ISA). Drivers with strong external loci of control react differently to autonomous interventions than those with strong internal loci. Individuals who ascribe crash causation to internal factors may resist vehicle automated overrides due to perceived loss of autonomy, whereas fatalistic drivers may exhibit dangerous over-reliance (automation complacency) on safety systems.

5. Psychological Construct

The central psychological construct assessed by the T-LOC is domain-specific attributional style and perceived behavioral control regarding traffic safety and vehicular collisions. Grounded in cognitive and social psychology, causal attribution refers to the underlying cognitive mechanism through which individuals explain the causes of behaviors, events, and outcomes. In the context of traffic safety, an accident represents a severe, salient negative outcome. When individuals reflect upon the prospective causes of accidents, their cognitive schemas invoke internalized causal models consisting of four differentiated sub-constructs:

1. Self (Internal Driving Factors)

The Self dimension measures an internal locus of control wherein the respondent acknowledges personal responsibility, behavioral choice, skill limitations, and intentional risk-taking as primary determinants of vehicular accidents. Individuals scoring high on this subscale attribute crash risk to actionable and modifiable behaviors under their direct volition, such as their driving skill deficits, driving under excessive velocity, tailgating (following too closely), and risky overtaking maneuvers. Theoretically, high internal attribution represents an adaptive cognitive appraisal associated with heightened vigilance, proactive hazard scanning, self-regulatory driving habits, and an awareness of one’s physical and cognitive boundaries on the road.

2. Other Drivers (External Social Factors)

The Other Drivers dimension captures an externalized interpersonal attribution. It assesses the extent to which a motorist projects accident causation onto the perceived incompetence, negligence, aggression, and rule violations of other road users. Items load on factors such as other drivers’ skill deficits, risky driving styles, excessive speeding, tailgating, driving under the influence of alcohol, and reckless overtaking. While other road users unquestionably introduce genuine objective hazards, disproportionately high scores on this dimension—unbalanced by self-reflection—indicate cognitive projection, defensive externalization, and the fundamental attribution error, wherein observers attribute others’ behaviors to internal stable character flaws while attributing their own slips to situational constraints.

3. Vehicle and Environment (External Situational/Physical Factors)

The Vehicle and Environment dimension captures external, non-social, contextual contingencies. It reflects the extent to which an individual views traffic accidents as the consequence of external physical hazards beyond immediate driver control, specifically sudden mechanical failure of the automobile, hazardous roadway geometry or surfaces, and adverse meteorological conditions such as heavy rainfall, ice, snow, or compromised illumination/darkness. While environmental awareness is essential for safe navigation, overestimating the role of physical conditions in accidents can serve as a cognitive justification for failing to adapt driving behavior to prevailing weather conditions (e.g., failing to reduce speed on wet roads).

4. Fate / Luck (External Fatalistic Factors)

The Fate dimension represents the fatalistic extreme of the external locus of control spectrum. It encapsulates beliefs that motor vehicle crashes are governed by stochastic, preordained, or supernatural forces that render human intervention futile. Items assessing bad luck, destiny, and coincidence comprise this factor. In psychological literature, fatalism is inversely associated with self-protective health behaviors. Within transportation systems, drivers with strong fatalistic tendencies often exhibit low seatbelt compliance, higher acceptance of speeding, and reduced cognitive engagement with defensive driving protocols, operating under the assumption that if an accident is destined to happen, personal precautions cannot alter the outcome.

6. Theoretical Framework

The Traffic Locus of Control Scale integrates foundational models from social learning theory, cognitive attribution theory, and transportation safety paradigms. The instrument’s conceptual architecture is derived from three primary theoretical pillars:

Rotter’s Social Learning Theory and Generalized Expectancies

In Julian Rotter’s (1954, 1966) Social Learning Theory, behavior potential ($BP$) is a function of expectancy ($E$) that a given reinforcement will follow the behavior in a given situation, and the value of that reinforcement ($RV$):

$$BP_{s1, r_a, s} = f(E_{s1, r_a, s} \times RV_{a, s})$$

Rotter posited that generalized expectancies across situations crystallize into a personality dimension designated as Locus of Control (LOC). Individuals with an internal locus perceive reinforcement as contingent upon their personal effort, competence, and deliberate actions. Conversely, individuals with an external locus perceive reinforcements and life events as contingent upon chance, luck, systemic constraints, or powerful others. The T-LOC adapts Rotter’s formulations by asserting that generalized expectancies become domain-differentiated through direct driving experience, driver education, cultural socialization, and exposure to vehicular trauma.

Weiner’s Attribution Theory

Bernard Weiner’s (1985, 1986) cognitive model of attribution posits that individuals evaluate unexpected, negative, or salient events along three orthogonal dimensional axes: Locus of Causality (internal vs. external), Stability (stable over time vs. unstable/transitory), and Controllability (volitionally controllable vs. uncontrollable). The T-LOC mirrors Weiner’s taxonomic structure within the driving environment:

• Internal-Controllable: One’s own risk-taking, driving speed, following distance (Self factor).
• Internal-Unstable/Variable: Situational lapses in driver skill or reaction time (Self factor).
• External-Social-Controllable (by Others): Other drivers’ speeding, alcohol impairment, and overtaking (Other Drivers factor).
• External-Physical-Variable: Weather shifts, sudden vehicular mechanical failure (Vehicle and Environment factor).
• External-Uncontrollable-Stochastic: Pure coincidence, destiny, and fortune (Fate factor).

Multidimensional Locus of Control Models

The T-LOC draws structural inspiration from Hannah Levenson’s (1974, 1981) tri-dimensional reconceptualization of Rotter’s unidimensional scale. Levenson demonstrated that the external pole must be bifurcated into “Powerful Others” and “Chance/Fate,” because people who believe the world is governed by orderly social institutions behave differently than those who believe events are arbitrary. Özkan and Lajunen expanded this multi-agent logic to traffic systems: in road safety, the “Powerful Others” construct translates into Other Drivers (social external agents), while physical-situational variables form a separate empirical cluster (Vehicle and Environment), and existential stochasticity forms the Fate factor.

7. Validity

The construct, criterion, convergent, and discriminant validity of the T-LOC have been evaluated across various cross-national populations and research settings:

Construct and Factorial Validity

Initial validation studies by Özkan and Lajunen (2005) involved substantial cohorts of Finnish and Turkish motorists. Exploratory and confirmatory factor analyses systematically verified the presence of the four theoretical dimensions: Self, Other Drivers, Vehicle and Environment, and Fate. Cross-cultural invariance analyses confirmed that the underlying four-factor construct remained stable across disparate driving environments—from low-fatality, highly institutionalized Nordic traffic environments (Finland) to middle-income, high-fatality Mediterranean environments characterized by distinct driving norms (Turkey).

Convergent Validity

Convergent validity is documented through statistically significant relationships between T-LOC dimensions and established psychometric instruments measuring aberrant driver behaviors, anger, and self-efficacy:

• Driver Behaviour Questionnaire (DBQ): Scores on the T-LOC Self dimension demonstrate strong positive correlations with the DBQ Violations and Aggressive Violations scales. Drivers who acknowledge their own risk-taking and high-speed driving as primary crash hazards paradoxically report higher frequencies of committing active, intentional violations. Conversely, drivers scoring high on the Fate dimension exhibit positive correlations with DBQ Lapses and Errors, suggesting that fatalistic beliefs co-occur with lower cognitive mindfulness and reduced defensive driving self-regulation.
• Driving Skills and Self-Efficacy: The Driver Skill Inventory (DSI) shows that motorists who rate their perceptual-motor skills exceptionally high relative to safety skills score higher on the Other Drivers factor. This reflects cognitive overconfidence, where the driver attributes accident vulnerabilities entirely to the driving shortcomings of peers.

Criterion-Related and Predictive Validity

The T-LOC shows predictive utility regarding actual road safety records, accident history, and technology acceptance:

• Crash Involvement and Moving Violations: In longitudinal and cross-sectional designs, high external fatalistic scores (Fate) are predictive of higher historical road traffic collision rates. In contrast, high internal scores on the Self dimension correlate with higher moving violation citations (specifically radar-recorded speeding offenses), reflecting the direct link between self-acknowledged behavioral risk-taking and active driving styles.
• Attitudes Toward Vehicle Automation: In human-factors evaluations assessing attitudes toward Intelligent Speed Adaptation (ISA) and Adaptive Cruise Control (ACC) (Özkan, Lajunen, & Kaistinen), individuals with an internal locus of control (Self) exhibited heightened initial resistance toward automated speed interventions, viewing them as encroachments upon personal driving agency. In contrast, those scoring high on the Other Drivers and Vehicle/Environment dimensions were more supportive of mandatory ADAS deployment, perceiving vehicle automation as a technological countermeasure against road hazards.

8. Reliability

The psychometric reliability of the Traffic Locus of Control Scale has been demonstrated via internal consistency metrics and temporal stability indices across international cohorts:

Internal Consistency Reliability

Across validation studies involving diverse driver samples, the subscales of the T-LOC consistently exhibit acceptable to strong internal consistency coefficients (Cronbach’s alpha, $\alpha$, and McDonald’s omega, $\omega$):

• Self Factor: Cronbach’s alpha coefficients routinely span between $\alpha = .70$ and $\alpha = .81$. In the foundational Finnish validation sample ($N = 254$), the alpha was observed at .74; in the Turkish validation sample ($N = 286$), the alpha reached .76. Items assessing driving speed (Item 7), tailgating (Item 9), and risk-taking (Item 2) demonstrate high item-total correlations ($r_{it} > .50$).
• Other Drivers Factor: This subscale exhibits strong reliability coefficients, typically ranging between $\alpha = .73$ and $\alpha = .84$. In the Nordic sample, $\alpha$ was .80, whereas in the Middle Eastern sample, $\alpha$ was .81. The items reflecting peer risk-taking (Item 4) and peer speeding (Item 8) show strong internal coherence.
• Vehicle and Environment Factor: Exhibiting the lowest item count (3 items: Items 6, 12, 13), this subscale typically yields alpha coefficients between $\alpha = .64$ and $\alpha = .72$. Given the brevity of a three-item composite, an alpha coefficient within this range meets psychometric criteria for exploratory and applied research.
• Fate Factor: Comprising three items (Items 5, 11, 17: bad luck, fate, coincidence), this subscale displays high internal consistency due to item homogeneity, with Cronbach’s alpha coefficients spanning $\alpha = .75$ to $\alpha = .86$ across European, Middle Eastern, and Asian driving cohorts.

Test-Retest Stability

Temporal stability over time has been verified across intervals of 4 to 8 weeks among non-clinical driver cohorts. Pearson test-retest correlation coefficients ($r_{tt}$) indicate substantial longitudinal stability:

• Self: $r_{tt} = .74$ ($p < .001$)
• Other Drivers: $r_{tt} = .78$ ($p < .001$)
• Vehicle and Environment: $r_{tt} = .69$ ($p < .001$)
• Fate: $r_{tt} = .82$ ($p < .001$)

These coefficients demonstrate that while driver attributional styles are sufficiently stable to reflect trait-like generalized expectancies, they retain some cognitive malleability, supporting their potential use as outcome measures in driver retraining programs.

9. Factor Analysis

The latent factorial architecture of the T-LOC has been investigated through both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA):

Exploratory Factor Analysis (EFA)

In initial scale development, Principal Axis Factoring (PAF) and Principal Component Analysis (PCA) were conducted using oblique rotation methods (Promax and Direct Oblimin) to permit inter-factor correlations among the dimensions of causal attribution. Extraction criteria based on eigenvalues ($lambda > 1.0$) and Cattell’s scree plot inspection confirmed a four-factor solution accounting for approximately 52% to 61% of the total cumulative variance across sample iterations.

Confirmatory Factor Analysis (CFA) and Goodness-of-Fit

Subsequent psychometric investigations have employed Structural Equation Modeling (SEM) to evaluate the structural integrity of the 17-item four-factor model against alternative competing models (e.g., a unidimensional model or a two-factor Internal vs. External model). CFA results confirm that the theoretical four-factor model provides a significantly superior fit to empirical covariance matrices across international cohorts:

• Chi-Square to Degrees of Freedom Ratio ($\chi^2/df$): Values regularly range from 1.62 to 2.45, falling below the conservative 3.0 threshold indicative of good structural fit.
• Root Mean Square Error of Approximation (RMSEA): Estimates span between .042 and .061 ($90% \text{ CI } [.035, .068]$), meeting Browne and Cudeck’s criterion for close approximate fit.
• Comparative Fit Index (CFI): Standardized values range from .92 to .96, demonstrating satisfactory fit relative to the null independence model.
• Tucker-Lewis Index (TLI): Coefficients regularly exceed the .90 benchmark, spanning .91 to .95.
• Standardized Root Mean Square Residual (SRMR): Residual indices remain low, typically ranging from .041 to .056.

Measurement invariance tests demonstrate metric and scalar invariance across biological sexes and partial scalar invariance across diverse national cohorts, verifying that the T-LOC can be utilized for comparative international traffic safety investigations.

10. Instrument / Measurement Tool

• Instrument Name: Traffic Locus of Control Scale (T-LOC)
• Primary Authors: Türker Özkan, Timo Lajunen, and Jyrki Kaistinen
• Construct Assessed: Causal attributions and perceived locus of control regarding vehicular accident causation
• Administration Format: Self-administered paper-and-pencil or computerized/online psychometric inventory
• Completion Duration: Approximately 5 to 8 minutes
• Target Population: Licensed active drivers across diverse demographics; applicable to professional drivers, novice motorists, and clinical/traffic offender cohorts
• Total Item Count: 17 explicit declarative statements
• Item Phrasing Stem: Each item begins with the invariant declarative stem: “Whether or not I get into car accident depends mostly on…”
• Subscale Structural Division:
  • Self Factor (5 items): Items 1, 2, 7, 9, 16
  • Other Drivers Factor (6 items): Items 3, 4, 8, 10, 14, 15
  • Vehicle and Environment Factor (3 items): Items 6, 12, 13
  • Fate Factor (3 items): Items 5, 11, 17
• Response Scale Options: 5-point Likert response format scored from 1 to 5:
  • 1 = Strongly disagree / Does not depend at all
  • 2 = Disagree / Depends slightly
  • 3 = Neither agree nor disagree / Depends moderately
  • 4 = Agree / Depends largely
  • 5 = Strongly agree / Depends completely
• Scoring and Computational Rules:
  • All 17 items are positively scored; there are no reverse-coded items.
  • Subscale scores can be derived either through direct linear summation of designated item ratings or by computing the arithmetic mean across dimension items.
  • Mean subscale scores preserve the original 1-to-5 metric: Self (range 1.0–5.0), Other Drivers (range 1.0–5.0), Vehicle and Environment (range 1.0–5.0), and Fate (range 1.0–5.0).
  • Higher subscale scores denote stronger causal attribution to that specific domain of accident determinants. Researchers should avoid calculating an aggregate “total T-LOC score,” as the four dimensions represent distinct, non-continuous constructs rather than a single continuum.

11. Permissions & Fee and Test Year

• Year of Initial Publication: 2005 (formal psychometric publication in Transportation Research Part F: Traffic Psychology and Behaviour).
• Copyright & Ownership: The intellectual property of the instrument is held by the original authors (Türker Özkan and Timo Lajunen) and the respective academic publishing bodies (Elsevier Ltd.).
• Usage Permissions: The T-LOC is considered an open-access psychometric instrument for non-commercial academic research, pedagogical use, and independent scientific investigations. Researchers may reproduce and administer the scale within academic protocols without paying licensing fees, provided formal bibliographic citation is credited to the original authors.
• Commercial and Fleet Licensing: Commercial applications, including deployment within proprietary fleet risk-management systems, enterprise hiring batteries, or for-profit occupational testing, require formal written permissions and licensing agreements from the copyright holders.
• Contact and Acquisition: Inquiries regarding adaptation, cross-cultural translation, or commercial implementation should be directed to the corresponding authors at Middle East Technical University (Department of Psychology, METU, Ankara, Turkey) or the University of Helsinki (Department of Psychology, Helsinki, Finland).

12. References

Below are primary academic references documenting the development, theoretical grounding, and psychometric validation of the Traffic Locus of Control Scale:

• Levenson, H. (1974). Activism and powerful others: Distinctions within the concept of internal-external control. Journal of Personality Assessment, 38(4), 377–383. https://doi.org/10.1080/00223891.1974.10119988
• Levenson, H. (1981). Differentiating among internality, powerful others, and chance. In H. M. Lefcourt (Ed.), Research with the Locus of Control Construct: Assessment Methods (Vol. 1, pp. 15–63). Academic Press. https://doi.org/10.1016/B978-0-12-443201-7.50006-3
• Özkan, T., & Lajunen, T. (2005). Multidimensional Traffic Locus of Control Scale (T-LOC): Factor structure and relationship to risky driving. Personality and Individual Differences, 38(3), 533–545. https://doi.org/10.1016/j.paid.2004.05.007
• Özkan, T., Lajunen, T., & Kaistinen, J. (2006). Traffic locus of control, driving skills, and attitudes towards in-vehicle technologies (ISA & ACC). In Proceedings of the Human Factors and Ergonomics Society Europe Chapter Annual Conference (pp. 211–220). Shaker Publishing.
• Reason, J., Manstead, A., Stradling, S., Baxter, J., & Campbell, K. (1990). Errors and violations on the roads: A real distinction? Ergonomics, 33(10-11), 1315–1332. https://doi.org/10.1080/00140139008925335
• Rotter, J. B. (1954). Social Learning and Clinical Psychology. Prentice-Hall. https://doi.org/10.1037/10788-000
• Rotter, J. B. (1966). Generalized expectancies for internal versus external control of reinforcement. Psychological Monographs: General and Applied, 80(1), 1–28. https://doi.org/10.1037/h0092976
• Weiner, B. (1985). An attributional theory of achievement motivation and emotion. Psychological Review, 92(4), 548–573. https://doi.org/10.1037/0033-295X.92.4.548
• Weiner, B. (1986). An Attributional Theory of Motivation and Emotion. Springer-Verlag. https://doi.org/10.1007/978-1-4612-4948-1

13. Items of the Scale (Questionnaire)