Duty of Care in Companion Dog Owners Scale

Abstract

The Duty of Care in Companion Dog Owners Scale (DCCDOS), developed by Dr. Carmen R. Glanville, Dr. Lauren M. Hemsworth, Professor Paul H. Hemsworth, and Professor Grahame J. Coleman (2023) at the Animal Welfare Science Centre, Faculty of Veterinary and Agricultural Sciences, University of Melbourne, is a psychometrically validated self-report instrument designed to operationalize and quantify the multidimensional moral, cognitive, and affective drivers of companion dog stewardship. Grounded structurally in Schwartz’s (1977) Norm Activation Model (NAM) and theoretically integrated with the Pet Care Competency (PCC) model (Glanville, Hemsworth, & Coleman, 2020), this 30-item scale captures five distinct, interlinked psychological dimensions: Duty Beliefs (Personal Norms), Problem Awareness, Awareness of Impact, Efficacy, and Ascription of Responsibility. Evaluated within a representative sample of Australian companion dog owners, the instrument demonstrates robust psychometric properties. Exploratory Factor Analysis (EFA) demonstrated strong single-factor or two-factor models across sub-dimensions, explaining up to 59.96% of the variance within domains. Subsequent Structural Equation Modeling (SEM) verified outstanding model fit across the multidimensional system (PCMIN/DF = 1.824, CFI = .947, RMSEA = .043, SRMR = .0458). Internal consistency reliability across the subscales is high, with Cronbach’s alpha coefficients ranging between .73 and .87. Construct and discriminant validity evaluations reveal that while duty of care correlates weakly to moderately with generalized dog–owner relationship quality and emotional pet attachment, it isolates an autonomous psychological mechanism governing perceived obligations and proactive welfare interventions. As companion animal welfare science shifts from preventing overt cruelty toward promoting positive affective states, the DCCDOS provides a vital diagnostic and predictive psychometric tool for researchers, applied ethologists, humane educators, veterinary professionals, and policy architects.

Keywords

Duty of Care, Companion Dog Owners, Norm Activation Model, Pet Care Competency Model, Canine Welfare, Human-Animal Interaction, Ascription of Responsibility, Duty Beliefs, Personal Norms, Interspecies Ethics, Psychometrics, Prosocial Motivation, Pet Attachment.

Authors

The Duty of Care in Companion Dog Owners Scale was designed, psychometrically evaluated, and published by a specialized research team at the Animal Welfare Science Centre within the Faculty of Veterinary and Agricultural Sciences at the University of Melbourne, Victoria, Australia:

• Carmen R. Glanville, Ph.D. (ORCID: 0000-0002-0754-0822) — Animal Welfare Science Centre, Faculty of Veterinary and Agricultural Sciences, University of Melbourne, Parkville, VIC, Australia. (Corresponding author: [email protected]).
• Lauren M. Hemsworth, Ph.D. (ORCID: 0000-0002-7752-8917) — Animal Welfare Science Centre, Faculty of Veterinary and Agricultural Sciences, University of Melbourne, Parkville, VIC, Australia.
• Paul H. Hemsworth, Ph.D. — Professor of Animal Welfare Science, Animal Welfare Science Centre, Faculty of Veterinary and Agricultural Sciences, University of Melbourne, Parkville, VIC, Australia.
• Grahame J. Coleman, Ph.D. — Professor Emeritus of Human-Animal Relationships, Animal Welfare Science Centre, Faculty of Veterinary and Agricultural Sciences, University of Melbourne, Parkville, VIC, Australia.

Purpose

The primary purpose of the Duty of Care in Companion Dog Owners Scale (DCCDOS) is to capture, delineate, and evaluate the specific cognitive, moral, and affective precursors that dictate how human caretakers attend to the physical, psychological, and behavioral needs of domestic canines (Canis lupus familiaris). Although hundreds of millions of households globally share living quarters with companion dogs, canine welfare outcomes vary radically across domestic settings. Suboptimal welfare in companion animals does not typically stem from active, malicious cruelty. Instead, empirical evidence in veterinary behavior and applied anthrozoology confirms that the overwhelming majority of companion animal suffering arises from passive neglect, ignorance of species-specific behavioral biology, deficits in mental enrichment, failure to recognize subclinical physical or emotional distress, and abandonment or relinquishment triggered by unmanaged behavioral problems.

Historically, human–animal interaction (HAI) researchers relied on broad assessments of affective bonding, such as pet attachment inventories (e.g., the Lexington Attachment to Pets Scale) or dog–owner relationship dimensions (e.g., the Monash Dog Owner Relationship Scale). However, empirical studies routinely encounter an ethical and behavioral paradox: strong emotional attachment to an animal does not reliably protect that animal from poor welfare outcomes. An owner may deeply love their dog and consider it an honorary child or family member, yet simultaneously expose the dog to chronic obesity through inappropriate overfeeding, allow social isolation that generates profound separation anxiety, or fail to implement environmental enrichment and veterinary care. Consequently, emotional affinity alone is insufficient to predict or ensure high-standard animal husbandry. There existed a pressing scientific imperative for a psychological measurement tool explicitly calibrated to evaluate the internalized obligation of stewardship—termed the owner’s duty of care.

From an applied perspective, the DCCDOS fulfills several critical research, educational, and clinical functions:

• Targeted Behavioral Interventions: By decomposing duty of care into granular sociocognitive components (such as recognizing that a problem exists versus believing that one has the personal capacity or responsibility to rectify it), the scale enables welfare organizations to design customized public education campaigns. If dog owners score low on Awareness of Impact but high on Duty Beliefs, campaigns do not need to lecture owners on morality; rather, they must deliver concrete educational information showing how basic managerial choices (e.g., exercise, cognitive enrichment) directly influence canine emotional balance.
• Veterinary Behavioral Consultation: Clinical veterinary behaviorists and animal trainers can employ the subscales to diagnose parental and management barriers. When a companion dog presents with destructive separation-related behaviors or fear-based reactivity, an owner who exhibits high Ascription of Responsibility and high Efficacy will respond far more successfully to complex, longitudinal behavior-modification plans than an owner who displaces responsibility entirely onto external environmental factors or dog temperament.
• Policy Development and Shelter Intake Prevention: Animal shelters, humane societies, and local governance bodies can utilize the instrument to benchmark the normative baseline of care within municipal populations, establishing early-warning metrics for high-risk demographics susceptible to surrender, neglect, or welfare compromise during socioeconomic hardship.
• Theoretical Expansion of Prosocial Psychology: The DCCDOS bridges conventional social psychology and anthrozoological ethics, demonstrating that theoretical models originally formulated to explain intra-human altruism and pro-environmental actions—such as recycling or blood donation—can be successfully translated to interspecies relationships characterized by extreme power asymmetries.

Psychological Construct

The construct of “Duty of Care” within the companion animal paradigm represents a synthesized, multidimensional psychological engine that motivates a caretaker to systematically monitor, preserve, and optimize the physiological, psychological, and environmental welfare states of a dependent animal. Under the DCCDOS operationalization, duty of care is not a monolithic, unitary trait, nor is it a simple emotional state. Instead, it constitutes an activated sociocognitive and moral sequence comprising five distinct dimensions:

1. Duty Beliefs (Personal Norms)

This dimension taps into the owner’s internalized, deeply held moral imperatives regarding the rights of animals and their perceived duties toward them. Unlike social norms—which reflect external societal expectations and social sanctions—personal norms reflect an individual’s internalized moral standards. In the context of the DCCDOS, Duty Beliefs encapsulate the conviction that bringing a sentient canine into a domestic household creates an absolute, non-negotiable ethical obligation to safeguard its physical health, psychological flourishing, and emotional happiness. It conceptualizes the dog as a sentient bearer of intrinsic moral claims, comparable in dignity to family members, and views the human–canine bond as an explicit relationship of trust. Owners scoring high on this dimension experience intrinsic cognitive and emotional pride when fulfilling these standards and anticipate internalized moral guilt or cognitive dissonance if they fail to uphold them.

2. Problem Awareness

A cognitive-perceptual construct, Problem Awareness measures the caretaker’s acute sensitivity to discrepancies between their dog’s current state of existence and an optimal state of physical and psychological well-being. Many owners remain trapped in an illusion of adequate welfare because their dog does not display overt, catastrophic clinical disease or screaming distress. High Problem Awareness reflects the vigilant recognition that an animal’s welfare is dynamic and that there is virtually always room for enhancement: the realization that the dog could lead a more stimulating, healthier, or happier life. This dimension captures the crucial first gate in proactive husbandry: an owner cannot be motivated to resolve a deficit that they perceive as non-existent.

3. Awareness of Impact

While Problem Awareness detects the existence of a suboptimal state, Awareness of Impact evaluates the owner’s causal comprehension of the direct link between human management decisions and canine outcomes. It measures an owner’s behavioral literacy: the understanding that chronic behavioral disturbances (e.g., destructive chewing, unremitting barking, indoor elimination, hyperactivity) are frequently the direct manifestation of unmet basic ethological needs, such as insufficient physical exercise or an absence of mental enrichment. Owners with elevated Awareness of Impact understand that management practices actively shape neurobehavioral development, affective states, and overall physiological health.

4. Efficacy

Drawing conceptually from Albert Bandura’s social cognitive paradigm, Efficacy in this instrument measures the caretaker’s perceived behavioral control and agency regarding their practical ability to influence their dog’s lived experience. It captures the owner’s confidence that they possess the capacity, skills, and opportunities to meaningfully elevate their dog’s happiness, health, and general quality of life. Even if an owner exhibits soaring moral duty beliefs and acute awareness of canine suffering, a deficit in perceived efficacy induces helplessness and behavioral paralysis, preventing the actualization of attentive care.

5. Ascription of Responsibility

This dimension assesses the locus of moral accountability that the caretaker accepts for their dog’s multidimensional mental, affective, and behavioral states. Grounded in attribution theory, Ascription of Responsibility assesses whether the owner views themselves as personally answerable if their dog experiences loneliness, anxiety, fear, depressive states, boredom, or behavioral dysregulation. Crucially, it contrasts an internal locus of custodial responsibility with externalization (e.g., blaming the dog’s innate personality, breed temperament, genetic determinism, or situational inconvenience). The instrument explicitly tests the resilience of this responsibility when external conditions deteriorate (e.g., financial strain, time scarcity), capturing whether the owner treats duty of care as contingent or unconditional.

Theoretical Framework

The construction and validation of the DCCDOS are anchored in two sophisticated theoretical models drawn from social psychology, moral philosophy, and applied ethology:

Schwartz’s Norm Activation Model (NAM)

The primary architectural scaffolding of the DCCDOS is derived from Shalom H. Schwartz’s seminal Norm Activation Model (1977). Originally developed to elucidate altruistic, prosocial behaviors within human interpersonal contexts—such as organ donation, volunteerism, and environmental conservation—the NAM posits that moral behaviors are directly triggered by the activation of Personal Norms (feelings of personal moral obligation). According to Schwartz, personal norms do not remain chronically active in a psychological vacuum; rather, they are systematically activated by specific cognitive activators structured in a defined causal sequence:

1. Awareness of Consequences (AC): The actor must perceive that another sentient entity faces potential harm, adversity, or unmet needs (operationalized in the DCCDOS as Problem Awareness and Awareness of Impact).
2. Ascription of Responsibility (AR): The actor must attribute the moral responsibility for mitigating that harm or fulfilling those needs to themselves, rather than deflecting it onto third parties or fate.
3. Outcome Efficacy: The actor must believe that targeted actions within their behavioral repertoire have the tangible capacity to alleviate the identified adversity.

When these cognitive gatekeepers are successfully cleared, internalized personal norms (Duty Beliefs) are powerfully activated, generating immediate psychological impulses to execute prosocial behaviors. If an individual attempts to evade this activated obligation, they experience internalized moral distress, guilt, and diminished self-concept. Conversely, if any gatekeeper is blocked—such as through defense mechanisms, denial of consequences, or abnegation of responsibility—norm activation is neutralized, and behavioral neglect ensues. Glanville and colleagues (2023) demonstrated that domestic canine stewardship represents an ideal ecological fit for Schwartz’s model: because a pet dog is entirely captive and dependent upon its human household for every biological, environmental, and emotional necessity, human caretaking behaviors fundamentally constitute a form of asymmetrical interspecies prosociality.

The Pet Care Competency (PCC) Model

To ground the abstract sociological principles of the NAM within the practical realities of companion animal husbandry, the authors integrated the Pet Care Competency (PCC) model (Glanville, Hemsworth, & Coleman, 2020). The PCC model posits that high-level animal welfare is the direct output of human caretaking competency, which operates through four interactive pillars: Knowledge, Skills, Resources, and Motivation. While conventional veterinary science focused heavily on technical knowledge (e.g., vaccine schedules, nutritional requirements), the PCC model recognized that knowledge remains inert unless energized by potent motivational forces. The DCCDOS provides the psychometric operationalization of the Motivation and Attitudinal pillars within the PCC framework, isolating how moral duty interfaces with perceived competence to drive everyday managerial fidelity.

Validity

The psychometric validation of the DCCDOS was conducted across a stratified, cross-sectional cohort of companion dog owners residing in Australia. The researchers subjected the instrument to rigorous empirical evaluations of construct validity, convergent validity, and structural integrity.

Construct and Convergent Validity

To evaluate construct validity, Glanville et al. (2023) examined bivariate correlations between the five DCCDOS subscales and established anthrozoological psychometric instruments, specifically:

• The Monash Dog Owner Relationship Scale (MDORS): Capturing Dog–Owner Interaction, Emotional Closeness, and Perceived Costs of dog ownership.
• The Lexington Attachment to Pets Scale (LAPS): Measuring generalized affective attachment.

As predicted by the theoretical model, the observed correlations between the duty of care subscales and generalized dog–owner relationship dimensions were generally weak to moderate. For example, emotional closeness on the MDORS shared significant, positive, but moderate correlations with Duty Beliefs, demonstrating that while affectionate attachment and moral obligation co-exist, they are distinct psychological constructs. An owner can feel emotionally close to a dog without necessarily possessing an acute recognition of its behavioral welfare needs or feeling personally responsible for its psychiatric states.

Correlations with pet attachment scores (LAPS) demonstrated a broader gradient from weak to strong across the sub-dimensions. Stronger associations were identified between attachment and personal norms (Duty Beliefs), indicating that owners who view their dogs as integral family members are more inclined to internalize moral duties. However, the correlation between attachment and Problem Awareness remained modest. This critical finding validated the discriminant divergence of the construct: highly attached owners frequently romanticize their pets’ lives and overlook subtle behavioral welfare deficiencies, highlighting that high attachment does not automatically confer diagnostic objectivity or elevated duty of care.

Structural Equation Modeling (SEM)

To evaluate the structural validity of the integrated model, the authors subjected the full item set to comprehensive Structural Equation Modeling (SEM) to test the hypothesized directional pathways derived from Schwartz’s Norm Activation Model. The measurement and structural models exhibited exceptional fit to the empirical data across multiple rigorous indices:

• PCMIN/DF (Chi-square / Degrees of Freedom ratio) = 1.824 (well below the traditional conservative benchmark of < 3.0, indicating outstanding parsimonious fit).
• Comparative Fit Index (CFI) = .947 (surpassing the standard > .90 threshold for robust model fit).
• Root Mean Square Error of Approximation (RMSEA) = .043 (with a 90% confidence interval falling well below the .05 cutoff, indicating minimal approximation error in the population).
• Standardized Root Mean Square Residual (SRMR) = .0458 (comfortably below the .08 threshold, confirming negligible residual variance).

These multivariate findings confirmed that the five subscales function not merely as independent inventories, but as an interconnected sociocognitive network wherein cognitive awareness and efficacy predict personal moral norms through the mediating bridge of ascribed responsibility.

Reliability

The reliability of the Duty of Care in Companion Dog Owners Scale was rigorously established using classical test theory metrics, focusing on internal consistency reliability across each distinct subscale. Across empirical validation trials, the internal consistency was found to be uniformly solid to excellent:

• Duty Beliefs (Personal Norms): Cronbach’s α = .84, demonstrating excellent homogeneity among the items capturing moral obligation, animal rights, and familial duty.
• Problem Awareness: Cronbach’s α = .78, indicating good internal consistency among items assessing the recognition of room for improvement in canine welfare, happiness, and health.
• Awareness of Impact: Cronbach’s α = .73, displaying acceptable to good reliability across items tapping into the causal connection between human management choices (enrichment, exercise) and behavioral well-being.
• Efficacy: Cronbach’s α = .87, reflecting high internal consistency among the items assessing perceived behavioral control and agency.
• Ascription of Responsibility: Cronbach’s α = .86, demonstrating robust reliability across the battery of items assessing perceived responsibility for emotional, cognitive, and behavioral canine states.

Item-total correlations across all retained items substantially exceeded the accepted psychometric minimum of .30, with most items exceeding .50. Inter-item correlations within subscales remained within the recommended range of .20 to .60, confirming high conceptual coherence without redundant item duplication.

Factor Analysis

The factorial architecture of the DCCDOS was established through extensive Exploratory Factor Analysis (EFA) followed by confirmatory structural validation.

Exploratory Factor Analysis (EFA)

During preliminary scale construction, the researchers formulated an initial pool of items distributed across six theoretical domains. Iterative EFA using principal axis factoring and oblique rotation (reflecting the theoretical expectation that sociocognitive predictors are intercorrelated) revealed that four core dimensional structures accounted for the overwhelming share of variance, leading to the refinement and consolidation of the final 30 items across five subscales:

• Personal Norms (Duty Beliefs): EFA extracted a clean, dominant single-factor solution that accounted for 35.80% of the total variance across those items. Factor loadings ranged from .42 to .78, confirming that statements concerning moral obligations, living up to canine trust, and animal rights load reliably on a singular latent construct.
• Ascription of Responsibility: EFA confirmed an unambiguous single-factor structure accounting for 47.44% of the variance. Items evaluating felt responsibility for internal affective states (fear, depression, loneliness, anxiety, mental state) and externalized behaviors (aggression, destructiveness) loaded heavily on this dimension (.55 to .81).
• Efficacy: The three items evaluating owner agency (perceived capacity to make the dog happier, healthier, and lead a better life) converged onto a single powerful factor accounting for 59.96% of the total variance, with factor loadings exceeding .72.
• Awareness of Need (Problem Awareness & Awareness of Impact): Factor analysis of the broader awareness domain yielded a distinct 10-item, two-factor solution accounting for 43.58% of the cumulative variance:
• Factor 1 (Awareness of Impact, 6 items): Focused on human behavioral management, enrichment, and exercise influencing behavioral outcomes.
• Factor 2 (Problem Awareness, 4 items): Focused on the realization that the dog’s life, health, and welfare could be elevated beyond its current state.

Measurement Invariance and Structural Specification

Following EFA, the five-factor structural model was verified via SEM. Factor loadings within the structural equation model confirmed that all items served as statistically significant (p < .001) indicators of their designated latent constructs. Crucially, the psychometric analyses established that while these five dimensions are functionally interrelated within the Norm Activation sequence, they do not collapse into a single higher-order global factor. Consequently, treating the scale as a singular composite score is psychometrically invalid; each subscale reflects an autonomous operational dimension.

Instrument / Measurement Tool

• Test Name: Duty of Care in Companion Dog Owners Scale (DCCDOS)
• Authors: Carmen R. Glanville, Lauren M. Hemsworth, Paul H. Hemsworth, and Grahame J. Coleman
• Year of Publication: 2023
• Instrument Nature: Standardized self-report inventory / psychometric questionnaire
• Construct Assessed: Moral, cognitive, and affective drivers of companion dog care and welfare stewardship
• Target Population: Adult companion dog owners (18 years of age and older)
• Administration Format: Paper-and-pencil, computer-assisted personal interviewing (CAPI), or secure online survey platforms
• Administration Time: Approximately 8 to 12 minutes
• Total Number of Items: 30 items
• Subscale Breakdown:
  • Duty Beliefs (Personal Norms) [DB]: 7 items (DB1 through DB7)
  • Problem Awareness [PA]: 4 items (PA1 through PA4)
  • Awareness of Impact [AoI]: 6 items (AoI1 through AoI6)
  • Efficacy [E]: 3 items (E1 through E3)
  • Ascription of Responsibility [AR]: 10 items (AR1 through AR10)
• Response Scales and Scoring System:
  • Standard 7-Point Likert Agreement Scale: Applied to 21 items, including DB1–DB7, PA1–PA4, AoI1–AoI6, E1–E3, and item AR1.
    • 1 = Strongly disagree
    • 2 = Disagree
    • 3 = Somewhat disagree
    • 4 = Neither agree nor disagree
    • 5 = Somewhat agree
    • 6 = Agree
    • 7 = Strongly agree
  • 5-Point Responsibility Scale: Applied exclusively to items AR2 through AR10 (which ask: “To what extent would you/do you feel responsible for the following…”).
    • 1 = Not at all responsible
    • 2 = Partly responsible
    • 3 = Somewhat responsible
    • 4 = Very responsible
    • 5 = Completely responsible
• Scoring and Aggregation Protocol:
  • CRITICAL SCORING DIRECTIVE: The DCCDOS consists of 5 distinct subscales representing specific sociocognitive constructs within a structural model. Items within each individual subscale can be summed (or averaged) to produce a composite score for that specific theoretical element.
  • DO NOT COMPUTE A GLOBAL TOTAL SCORE: Subscale scores must not be combined into a single grand total. Summing all 30 items across subscales produces an uninterpretable metric that violates the underlying factor structure and obscures the individual activation gates of the Norm Activation Model.
  • Reverse Coding: Item DB7 (“We don’t have any particular obligations or duties to our dogs”) and item AR1 (“During challenging times… it is reasonable to lower our expectations for dog care and welfare”) are phrased in the counter-normative direction and must be reverse-scored prior to calculating subscale indices.

Permissions & Fee and Test Year

The Duty of Care in Companion Dog Owners Scale was developed in 2023 by researchers at the Animal Welfare Science Centre, University of Melbourne, and published under open-access terms in PLoS ONE.

• Licensing Framework: The scale is licensed under the Creative Commons Attribution 4.0 International License (CC BY 4.0).
• Usage Rights: Under this open-access license, researchers, academic institutions, veterinary professionals, animal sheltering organizations, and educators are permitted to freely copy, distribute, adapt, and administer the instrument for research, pedagogical, and non-commercial purposes without payment of royalty fees.
• Attribution Requirement: Any formal implementation, empirical paper, conference presentation, or digital integration of the scale must cite the original publication (Glanville et al., 2023).
• Commercial Applications: Commercial utilization, proprietary software integration, or fee-charging consultancy deployments require direct consultation with the corresponding author and the University of Melbourne.

References

• Bandura, A. (1997). Self-efficacy: The exercise of control. W. H. Freeman and Company.
• Glanville, C. R., Hemsworth, L. M., & Coleman, G. J. (2020). Pet care competency: Expanding our understanding of the human-animal bond and companion animal welfare. Animals, 10(11), Article 2111. https://doi.org/10.3390/ani10112111
• Glanville, C. R., Hemsworth, L. M., Hemsworth, P. H., & Coleman, G. J. (2023). Duty of care in companion dog owners: Preliminary scale development and empirical exploration. PLoS ONE, 18(5), Article e0285278. https://doi.org/10.1371/journal.pone.0285278
• Johnson, T. P., Garrity, T. F., & Stallones, L. (1992). Psychometric evaluation of the Lexington Attachment to Pets Scale (LAPS). Anthrozoös, 5(3), 160–175. https://doi.org/10.2752/089279392787011395
• Mellor, D. J., Beausoleil, N. J., Littlewood, K. E., McLean, A. N., McGreevy, P. D., Dyson, S., & Damien, C. (2020). The 2020 Five Domains Model: Including human–animal interactions in assessments of animal welfare. Animals, 10(10), Article 1870. https://doi.org/10.3390/ani10101870
• Dwyer, F., Bennett, P. C., & Coleman, G. J. (2006). Development of the Monash Dog Owner Relationship Scale (MDORS). Anthrozoös, 19(3), 243–256. https://doi.org/10.2752/089279306785415592
• Schwartz, S. H. (1977). Normative influences on altruism. In L. Berkowitz (Ed.), Advances in Experimental Social Psychology (Vol. 10, pp. 221–279). Academic Press. https://doi.org/10.1016/S0065-2601(08)60358-5
• Steg, L., & de Groot, J. I. (2010). Explaining prosocial behavior: Testing the value-belief-norm theory of environmentalism and the norm activation model. Rationality and Society, 22(1), 89–110. https://doi.org/10.1177/1043463109353947

Items of the Scale