Limits...
Measuring Laypeople's Trust in Experts in a Digital Age: The Muenster Epistemic Trustworthiness Inventory (METI).

Hendriks F, Kienhues D, Bromme R - PLoS ONE (2015)

Bottom Line: Exploratory (n = 237) and confirmatory factor analyses (n = 345) showed that the Muenster Epistemic Trustworthiness Inventory (METI) is composed of these three factors.A subsequent experimental study (n = 137) showed that all three dimensions of the METI are sensitive to variation in source characteristics.We propose using this inventory to measure assignments of epistemic trustworthiness, that is, all judgments laypeople make when deciding whether to place epistemic trust in-and defer to-an expert in order to solve a scientific informational problem that is beyond their understanding.

View Article: PubMed Central - PubMed

Affiliation: Institute for Psychology, University of Muenster, Muenster, Nordrhein-Westfalen, Germany.

ABSTRACT
Given their lack of background knowledge, laypeople require expert help when dealing with scientific information. To decide whose help is dependable, laypeople must judge an expert's epistemic trustworthiness in terms of competence, adherence to scientific standards, and good intentions. Online, this may be difficult due to the often limited and sometimes unreliable source information available. To measure laypeople's evaluations of experts (encountered online), we constructed an inventory to assess epistemic trustworthiness on the dimensions expertise, integrity, and benevolence. Exploratory (n = 237) and confirmatory factor analyses (n = 345) showed that the Muenster Epistemic Trustworthiness Inventory (METI) is composed of these three factors. A subsequent experimental study (n = 137) showed that all three dimensions of the METI are sensitive to variation in source characteristics. We propose using this inventory to measure assignments of epistemic trustworthiness, that is, all judgments laypeople make when deciding whether to place epistemic trust in-and defer to-an expert in order to solve a scientific informational problem that is beyond their understanding.

No MeSH data available.


Interaction graph: Mean ratings on the scale ‘benevolence’.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4608577&req=5

pone.0139309.g003: Interaction graph: Mean ratings on the scale ‘benevolence’.

Mentions: For ‘benevolence’, all main effects and the interaction were significant with p < .001 (main effect of dimensions of ET: F (1.91, 259.99) = 79.34, p < .001, ηp2 = .37; main effect of manifestation of ET: F (1, 136) = 994.58, p < .001, ηp2 = .88; interaction of dimensions and manifestation of ET: F (2, 272) = 341.55, p < .001, ηp2 = .72. Planned contrasts were conducted to break down the interaction. Comparing the margin in ratings between authors who were described to be of low or high benevolence, there was a significant difference to the margin between authors described to be low or high in expertise (F (1, 136) = 585.57, p < .001, r = .90), and integrity (F (1, 136) = 28.40, p < .001, r = .42). In Fig 3 it can be seen that the margin between means in ratings of ‘benevolence’ is larger, when descriptions were targeting benevolence than when they were targeting expertise or integrity (means and standard deviations can be found in Table 6).


Measuring Laypeople's Trust in Experts in a Digital Age: The Muenster Epistemic Trustworthiness Inventory (METI).

Hendriks F, Kienhues D, Bromme R - PLoS ONE (2015)

Interaction graph: Mean ratings on the scale ‘benevolence’.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4608577&req=5

pone.0139309.g003: Interaction graph: Mean ratings on the scale ‘benevolence’.
Mentions: For ‘benevolence’, all main effects and the interaction were significant with p < .001 (main effect of dimensions of ET: F (1.91, 259.99) = 79.34, p < .001, ηp2 = .37; main effect of manifestation of ET: F (1, 136) = 994.58, p < .001, ηp2 = .88; interaction of dimensions and manifestation of ET: F (2, 272) = 341.55, p < .001, ηp2 = .72. Planned contrasts were conducted to break down the interaction. Comparing the margin in ratings between authors who were described to be of low or high benevolence, there was a significant difference to the margin between authors described to be low or high in expertise (F (1, 136) = 585.57, p < .001, r = .90), and integrity (F (1, 136) = 28.40, p < .001, r = .42). In Fig 3 it can be seen that the margin between means in ratings of ‘benevolence’ is larger, when descriptions were targeting benevolence than when they were targeting expertise or integrity (means and standard deviations can be found in Table 6).

Bottom Line: Exploratory (n = 237) and confirmatory factor analyses (n = 345) showed that the Muenster Epistemic Trustworthiness Inventory (METI) is composed of these three factors.A subsequent experimental study (n = 137) showed that all three dimensions of the METI are sensitive to variation in source characteristics.We propose using this inventory to measure assignments of epistemic trustworthiness, that is, all judgments laypeople make when deciding whether to place epistemic trust in-and defer to-an expert in order to solve a scientific informational problem that is beyond their understanding.

View Article: PubMed Central - PubMed

Affiliation: Institute for Psychology, University of Muenster, Muenster, Nordrhein-Westfalen, Germany.

ABSTRACT
Given their lack of background knowledge, laypeople require expert help when dealing with scientific information. To decide whose help is dependable, laypeople must judge an expert's epistemic trustworthiness in terms of competence, adherence to scientific standards, and good intentions. Online, this may be difficult due to the often limited and sometimes unreliable source information available. To measure laypeople's evaluations of experts (encountered online), we constructed an inventory to assess epistemic trustworthiness on the dimensions expertise, integrity, and benevolence. Exploratory (n = 237) and confirmatory factor analyses (n = 345) showed that the Muenster Epistemic Trustworthiness Inventory (METI) is composed of these three factors. A subsequent experimental study (n = 137) showed that all three dimensions of the METI are sensitive to variation in source characteristics. We propose using this inventory to measure assignments of epistemic trustworthiness, that is, all judgments laypeople make when deciding whether to place epistemic trust in-and defer to-an expert in order to solve a scientific informational problem that is beyond their understanding.

No MeSH data available.