AI Trust S1 - Analysis
Date of study: 27 May 2026. Last compiled on 14 August, 2026
Valid entries and size of groups
There is a total of 231 entries in the dataset. 15 participants failed the attention check or the comprehension check. The valid number of entries is n = 216.
| ai_training | n | percentage |
|---|---|---|
| yes_ai | 75 | 34.72% |
| control | 70 | 32.41% |
| no_ai | 71 | 32.87% |
Creating composites
trust
α = 0.941 95% CI [ 0.925 , 0.956 ]
✓ Composite trust created
ability
α = 0.955 95% CI [ 0.944 , 0.965 ]
✓ Composite ability created
benevolence
α = 0.891 95% CI [ 0.866 , 0.917 ]
✓ Composite benevolence created
integrity
α = 0.895 95% CI [ 0.87 , 0.919 ]
✓ Composite integrity created
autonomy
α = 0.906 95% CI [ 0.884 , 0.927 ]
✓ Composite autonomy created
Effect of condition on DVs (ANOVAs)
Trust
2-item composite, each on its own 0-6 scale.
- comfort: “On a scale of 0 to 6, where 0 = Not comfortable at all and 6 = Very comfortable, how comfortable would you be having Dr. Smith as your doctor?”
- confidence: “On a scale of 0 to 6, where 0 = Not confident at all and 6 = Very confident, how confident are you that Dr. Smith provides his patients with high-quality medical care?”
Summary
| ai_training | variable | n | mean | sd | se | ci |
|---|---|---|---|---|---|---|
| yes_ai | trust | 75 | 3.833 | 1.632 | 0.188 | 0.376 |
| control | trust | 70 | 4.936 | 0.940 | 0.112 | 0.224 |
| no_ai | trust | 71 | 5.542 | 0.602 | 0.071 | 0.143 |
ANOVA
## $output1
## Anova Table (Type II tests)
##
## Response: trust
## Sum Sq Df F value Pr(>F)
## ai_training 110.0 2 41.322 6.847e-16 ***
## Residuals 283.5 213
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## $output2
## # Effect Size for ANOVA
##
## Parameter | Eta2 | 95% CI
## ---------------------------------
## ai_training | 0.28 | [0.20, 1.00]
##
## - One-sided CIs: upper bound fixed at [1.00].
## $output3
## # Effect Size for ANOVA
##
## Parameter | Cohen's f | 95% CI
## -------------------------------------
## ai_training | 0.62 | [0.49, Inf]
##
## - One-sided CIs: upper bound fixed at [Inf].Plots & Comparisons
| group1 | group2 | n1.x | n2.x | p | p.signif | p.adj | p.adj.signif | effsize | n1.y | n2.y | conf.low | conf.high | Cohens’ d |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| yes_ai | control | 75 | 70 | 0.0000 | **** | 0.0000 | **** | -0.8205 | 75 | 70 | -1.14 | -0.53 | large |
| yes_ai | no_ai | 75 | 71 | 0.0000 | **** | 0.0000 | **** | -1.3747 | 75 | 71 | -1.70 | -1.10 | large |
| control | no_ai | 70 | 71 | 0.0021 | ** | 0.0021 | ** | -0.7696 | 70 | 71 | -1.14 | -0.42 | moderate |
Regression
Ability
3-item composite. Agree-disagree scale (0 = Strongly disagree, 6 = Strongly agree).
- ability2: “Dr. Smith is well qualified to treat patients.”
- ability3: “I feel confident in Dr. Smith’s clinical skills.”
- ability4: “Dr. Smith is highly capable of performing his medical duties.”
Summary
| ai_training | variable | n | mean | sd | se | ci |
|---|---|---|---|---|---|---|
| yes_ai | ability | 75 | 4.129 | 1.419 | 0.164 | 0.327 |
| control | ability | 70 | 5.019 | 0.780 | 0.093 | 0.186 |
| no_ai | ability | 71 | 5.540 | 0.567 | 0.067 | 0.134 |
ANOVA
## $output1
## Anova Table (Type II tests)
##
## Response: ability
## Sum Sq Df F value Pr(>F)
## ai_training 74.584 2 37.19 1.403e-14 ***
## Residuals 213.588 213
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## $output2
## # Effect Size for ANOVA
##
## Parameter | Eta2 | 95% CI
## ---------------------------------
## ai_training | 0.26 | [0.18, 1.00]
##
## - One-sided CIs: upper bound fixed at [1.00].
## $output3
## # Effect Size for ANOVA
##
## Parameter | Cohen's f | 95% CI
## -------------------------------------
## ai_training | 0.59 | [0.46, Inf]
##
## - One-sided CIs: upper bound fixed at [Inf].Plots & Comparisons
| group1 | group2 | n1.x | n2.x | p | p.signif | p.adj | p.adj.signif | effsize | n1.y | n2.y | conf.low | conf.high | Cohens’ d |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| yes_ai | control | 75 | 70 | 0.0000 | **** | 0.0000 | **** | -0.7701 | 75 | 70 | -1.09 | -0.52 | moderate |
| yes_ai | no_ai | 75 | 71 | 0.0000 | **** | 0.0000 | **** | -1.2925 | 75 | 71 | -1.60 | -1.02 | large |
| control | no_ai | 70 | 71 | 0.0023 | ** | 0.0023 | ** | -0.7646 | 70 | 71 | -1.16 | -0.44 | moderate |
Regression
Benevolence
3-item composite. Agree-disagree scale (0 = Strongly disagree, 6 = Strongly agree).
- benevolence1: “Dr. Smith is genuinely concerned about his patients’ welfare.”
- benevolence2: “Dr. Smith would never knowingly do anything to harm a patient.”
- benevolence4: “Dr. Smith truly looks out for what matters most to his patients.”
Summary
| ai_training | variable | n | mean | sd | se | ci |
|---|---|---|---|---|---|---|
| yes_ai | benevolence | 75 | 4.164 | 1.194 | 0.138 | 0.275 |
| control | benevolence | 70 | 4.417 | 0.989 | 0.118 | 0.236 |
| no_ai | benevolence | 71 | 4.892 | 0.980 | 0.116 | 0.232 |
ANOVA
## $output1
## Anova Table (Type II tests)
##
## Response: benevolence
## Sum Sq Df F value Pr(>F)
## ai_training 19.796 2 8.7755 0.0002176 ***
## Residuals 240.241 213
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## $output2
## # Effect Size for ANOVA
##
## Parameter | Eta2 | 95% CI
## ---------------------------------
## ai_training | 0.08 | [0.03, 1.00]
##
## - One-sided CIs: upper bound fixed at [1.00].
## $output3
## # Effect Size for ANOVA
##
## Parameter | Cohen's f | 95% CI
## -------------------------------------
## ai_training | 0.29 | [0.16, Inf]
##
## - One-sided CIs: upper bound fixed at [Inf].Plots & Comparisons
| group1 | group2 | n1.x | n2.x | p | p.signif | p.adj | p.adj.signif | effsize | n1.y | n2.y | conf.low | conf.high | Cohens’ d |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| yes_ai | control | 75 | 70 | 0.1540 | ns | 0.1540 | ns | -0.2293 | 75 | 70 | -0.53 | 0.11 | small |
| yes_ai | no_ai | 75 | 71 | 0.0001 | **** | 0.0002 | *** | -0.6644 | 75 | 71 | -1.01 | -0.35 | moderate |
| control | no_ai | 70 | 71 | 0.0085 | ** | 0.0169 | * | -0.4827 | 70 | 71 | -0.85 | -0.17 | small |
Regression
Integrity
3-item composite. Agree-disagree scale (0 = Strongly disagree, 6 = Strongly agree).
- integrity1: “Dr. Smith has values I share.”
- integrity2: “Dr. Smith’s decisions and actions are guided by sound principles.”
- integrity3: “Dr. Smith can be counted on to be honest with his patients.”
Summary
| ai_training | variable | n | mean | sd | se | ci |
|---|---|---|---|---|---|---|
| yes_ai | integrity | 75 | 3.682 | 1.357 | 0.157 | 0.312 |
| control | integrity | 70 | 4.076 | 0.977 | 0.117 | 0.233 |
| no_ai | integrity | 71 | 4.685 | 0.968 | 0.115 | 0.229 |
ANOVA
## $output1
## Anova Table (Type II tests)
##
## Response: integrity
## Sum Sq Df F value Pr(>F)
## ai_training 37.125 2 14.77 9.838e-07 ***
## Residuals 267.689 213
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## $output2
## # Effect Size for ANOVA
##
## Parameter | Eta2 | 95% CI
## ---------------------------------
## ai_training | 0.12 | [0.06, 1.00]
##
## - One-sided CIs: upper bound fixed at [1.00].
## $output3
## # Effect Size for ANOVA
##
## Parameter | Cohen's f | 95% CI
## -------------------------------------
## ai_training | 0.37 | [0.25, Inf]
##
## - One-sided CIs: upper bound fixed at [Inf].Plots & Comparisons
| group1 | group2 | n1.x | n2.x | p | p.signif | p.adj | p.adj.signif | effsize | n1.y | n2.y | conf.low | conf.high | Cohens’ d |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| yes_ai | control | 75 | 70 | 0.0356 | * | 0.0356 | * | -0.3313 | 75 | 70 | -0.63 | -0.02 | small |
| yes_ai | no_ai | 75 | 71 | 0.0000 | **** | 0.0000 | **** | -0.8473 | 75 | 71 | -1.21 | -0.54 | large |
| control | no_ai | 70 | 71 | 0.0014 | ** | 0.0029 | ** | -0.6268 | 70 | 71 | -0.98 | -0.27 | moderate |
Regression
Autonomy
3-item composite. Agree-disagree scale (0 = Strongly disagree, 6 = Strongly agree).
- autonomy1: “Dr. Smith exercises independent judgment in his clinical decisions.”
- autonomy3: “Dr. Smith takes personal responsibility for the consequences of his decisions.”
- autonomy4: “Dr. Smith forms his own opinion rather than simply deferring to what a computer-based tool or algorithm recommends.”
Summary
| ai_training | variable | n | mean | sd | se | ci |
|---|---|---|---|---|---|---|
| yes_ai | autonomy | 75 | 3.298 | 1.425 | 0.165 | 0.328 |
| control | autonomy | 70 | 4.176 | 0.954 | 0.114 | 0.228 |
| no_ai | autonomy | 71 | 5.122 | 0.811 | 0.096 | 0.192 |
ANOVA
## $output1
## Anova Table (Type II tests)
##
## Response: autonomy
## Sum Sq Df F value Pr(>F)
## ai_training 121.39 2 49.869 < 2.2e-16 ***
## Residuals 259.23 213
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## $output2
## # Effect Size for ANOVA
##
## Parameter | Eta2 | 95% CI
## ---------------------------------
## ai_training | 0.32 | [0.24, 1.00]
##
## - One-sided CIs: upper bound fixed at [1.00].
## $output3
## # Effect Size for ANOVA
##
## Parameter | Cohen's f | 95% CI
## -------------------------------------
## ai_training | 0.68 | [0.55, Inf]
##
## - One-sided CIs: upper bound fixed at [Inf].Plots & Comparisons
| group1 | group2 | n1.x | n2.x | p | p.signif | p.adj | p.adj.signif | effsize | n1.y | n2.y | conf.low | conf.high | Cohens’ d |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| yes_ai | control | 75 | 70 | 0 | **** | 0 | **** | -0.7194 | 75 | 70 | -1.02 | -0.40 | moderate |
| yes_ai | no_ai | 75 | 71 | 0 | **** | 0 | **** | -1.5621 | 75 | 71 | -1.96 | -1.24 | large |
| control | no_ai | 70 | 71 | 0 | **** | 0 | **** | -1.0687 | 70 | 71 | -1.50 | -0.70 | large |
Regression
Ai Use
Single item. Agree-disagree scale (0 = Strongly disagree, 6 = Strongly agree).
“Dr. Smith often uses computer-based diagnostic tools in his medical practice.”
Summary
| ai_training | variable | n | mean | sd | se | ci |
|---|---|---|---|---|---|---|
| yes_ai | ai_use | 75 | 4.560 | 1.130 | 0.130 | 0.260 |
| control | ai_use | 69 | 3.319 | 1.022 | 0.123 | 0.245 |
| no_ai | ai_use | 71 | 1.958 | 1.616 | 0.192 | 0.383 |
ANOVA
## $output1
## Anova Table (Type II tests)
##
## Response: ai_use
## Sum Sq Df F value Pr(>F)
## ai_training 247.01 2 75.165 < 2.2e-16 ***
## Residuals 348.34 212
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## $output2
## # Effect Size for ANOVA
##
## Parameter | Eta2 | 95% CI
## ---------------------------------
## ai_training | 0.41 | [0.33, 1.00]
##
## - One-sided CIs: upper bound fixed at [1.00].
## $output3
## # Effect Size for ANOVA
##
## Parameter | Cohen's f | 95% CI
## -------------------------------------
## ai_training | 0.84 | [0.71, Inf]
##
## - One-sided CIs: upper bound fixed at [Inf].Plots & Comparisons
| group1 | group2 | n1.x | n2.x | p | p.signif | p.adj | p.adj.signif | effsize | n1.y | n2.y | conf.low | conf.high | Cohens’ d |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| yes_ai | control | 75 | 70 | 0 | **** | 0 | **** | 1.1498 | 75 | 69 | 0.75 | 1.59 | large |
| yes_ai | no_ai | 75 | 71 | 0 | **** | 0 | **** | 1.8751 | 75 | 71 | 1.44 | 2.48 | large |
| control | no_ai | 70 | 71 | 0 | **** | 0 | **** | 1.0035 | 69 | 71 | 0.65 | 1.47 | large |
Regression
## Analisis with age as dv
trust_again
Summary
Plots & Comparisons
Regression
