Measurement items and measurement-model statistics
| Construct | Item | Measurement items | Loading | Alpha | CR | AVE | Measurement basis |
|---|---|---|---|---|---|---|---|
| Interaction quality | IQ1 | The AI system provides clear information during performance evaluations | 0.795 | 0.827 | 0.885 | 0.658 | Pelau et al. (2021), Haugeland et al. (2022) |
| IQ2 | The AI is easy to use for HR tasks | 0.836 | |||||
| IQ3 | The AI system responds quickly to feedback | 0.795 | |||||
| IQ4 | AI makes HR processes easier to manage | 0.818 | |||||
| Human agency | HA1 | I feel in control of the AI-assisted evaluation process | 0.784 | 0.688 | 0.826 | 0.613 | Fanni et al. (2023), Vantrepotte et al. (2022) |
| HA2 | I can provide feedback to the AI system | 0.818 | |||||
| HA3 | I can override the AI decisions if necessary | 0.745 | |||||
| Perceived humanness | PHUM1 | The AI understands employee needs during performance reviews | 0.852 | 0.830 | 0.898 | 0.746 | Lu et al. (2022) |
| PHUM2 | The AI system provides feedback like a human manager | 0.867 | |||||
| PHUM3 | Interaction with the AI feels like dealing with a person | 0.872 | |||||
| Voice | VOICE1 | I can express my opinions during AI-led performance evaluations | 0.836 | 0.728 | 0.846 | 0.647 | Colquitt (2001), Hellwig et al. (2023) |
| VOICE2 | My input is considered in the AI system's decisions | 0.819 | |||||
| VOICE3 | I feel I can influence AI decisions that affect me | 0.755 | |||||
| Fairness | FAIR1 | The AI system treats employees fairly in HR processes | 0.890 | 0.700 | 0.869 | 0.769 | Colquitt (2001), Shin and Park (2019), Shin (2020) |
| FAIR2 | I believe the AI system is unbiased | 0.864 | |||||
| Accountability | ACC1 | The AI system holds me accountable for my performance | 0.805 | 0.717 | 0.841 | 0.638 | Bovens (2010), Shin and Park (2019) |
| ACC2 | I am responsible for the work in the AI system | 0.749 | |||||
| ACC3 | The AI decisions are explainable and justifiable | 0.840 | |||||
| Transparency | TRANS1 | I understand how the AI system makes HR decisions | 0.815 | 0.792 | 0.878 | 0.707 | Shin and Park (2019), Shin (2020, 2021) |
| TRANS2 | The AI system explains its HR decision-making process | 0.854 | |||||
| TRANS3 | The AI provides enough information about its decisions | 0.853 | |||||
| Trust in AI | TRUST1 | I trust the AI to evaluate employees' performance accurately | 0.854 | 0.787 | 0.876 | 0.702 | Lee and See (2004), Shin and Park (2019) |
| TRUST2 | The AI system follows through on its promises | 0.829 | |||||
| TRUST3 | Overall the decision by AI can be trusted | 0.829 | |||||
| Task complexity | TC1 | The HR task using AI involves managing many details | 0.820 | 0.686 | 0.829 | 0.619 | Wood (1986) |
| TC2 | Using AI for HR decisions requires concentration | 0.822 | |||||
| TC3 | AI tasks in HR involve several steps that are difficult to track | 0.712 | |||||
| Perceived uncanniness | UNC1 | The AI system feels unnatural when used for HR decisions | 0.875 | 0.849 | 0.905 | 0.762 | Ho and MacDorman (2010), Sullivan et al. (2022) |
| UNC2 | The AI's human-like qualities make me uncomfortable | 0.909 | |||||
| UNC3 | The AI feels unsettling in HR process | 0.832 |
| Construct | Item | Measurement items | Loading | Alpha | CR | AVE | Measurement basis |
|---|---|---|---|---|---|---|---|
| Interaction quality | IQ1 | The AI system provides clear information during performance evaluations | 0.795 | 0.827 | 0.885 | 0.658 | |
| IQ2 | The AI is easy to use for HR tasks | 0.836 | |||||
| IQ3 | The AI system responds quickly to feedback | 0.795 | |||||
| IQ4 | AI makes HR processes easier to manage | 0.818 | |||||
| Human agency | HA1 | I feel in control of the AI-assisted evaluation process | 0.784 | 0.688 | 0.826 | 0.613 | |
| HA2 | I can provide feedback to the AI system | 0.818 | |||||
| HA3 | I can override the AI decisions if necessary | 0.745 | |||||
| Perceived humanness | PHUM1 | The AI understands employee needs during performance reviews | 0.852 | 0.830 | 0.898 | 0.746 | |
| PHUM2 | The AI system provides feedback like a human manager | 0.867 | |||||
| PHUM3 | Interaction with the AI feels like dealing with a person | 0.872 | |||||
| Voice | VOICE1 | I can express my opinions during AI-led performance evaluations | 0.836 | 0.728 | 0.846 | 0.647 | |
| VOICE2 | My input is considered in the AI system's decisions | 0.819 | |||||
| VOICE3 | I feel I can influence AI decisions that affect me | 0.755 | |||||
| Fairness | FAIR1 | The AI system treats employees fairly in HR processes | 0.890 | 0.700 | 0.869 | 0.769 | |
| FAIR2 | I believe the AI system is unbiased | 0.864 | |||||
| Accountability | ACC1 | The AI system holds me accountable for my performance | 0.805 | 0.717 | 0.841 | 0.638 | |
| ACC2 | I am responsible for the work in the AI system | 0.749 | |||||
| ACC3 | The AI decisions are explainable and justifiable | 0.840 | |||||
| Transparency | TRANS1 | I understand how the AI system makes HR decisions | 0.815 | 0.792 | 0.878 | 0.707 | |
| TRANS2 | The AI system explains its HR decision-making process | 0.854 | |||||
| TRANS3 | The AI provides enough information about its decisions | 0.853 | |||||
| Trust in AI | TRUST1 | I trust the AI to evaluate employees' performance accurately | 0.854 | 0.787 | 0.876 | 0.702 | |
| TRUST2 | The AI system follows through on its promises | 0.829 | |||||
| TRUST3 | Overall the decision by AI can be trusted | 0.829 | |||||
| Task complexity | TC1 | The HR task using AI involves managing many details | 0.820 | 0.686 | 0.829 | 0.619 | |
| TC2 | Using AI for HR decisions requires concentration | 0.822 | |||||
| TC3 | AI tasks in HR involve several steps that are difficult to track | 0.712 | |||||
| Perceived uncanniness | UNC1 | The AI system feels unnatural when used for HR decisions | 0.875 | 0.849 | 0.905 | 0.762 | |
| UNC2 | The AI's human-like qualities make me uncomfortable | 0.909 | |||||
| UNC3 | The AI feels unsettling in HR process | 0.832 |
Note(s): FAIR3 was removed before final estimation because it explicitly incorporated trust wording. All reported analyses use FAIR1 and FAIR2
Sharing content requires targeting cookies to be enabled. Please update your cookie preferences to use this feature.