'Do managers accept artificial intelligence? Study 1' (AsPredicted #107,387)
Author(s) Miriam Gieselmann (IWM Tübingen) - m.gieselmann+prereg@mailbox.org Kai Sassenberg (Leibniz-Institut für Wissensmedien, Tübingen) - ksa@leibniz-psychology.org
Pre-registered on 2022/09/19 23:55 (PT)
1) Have any data been collected for this study already? No, no data have been collected for this study yet.
2) What's the main question being asked or hypothesis being tested in this study? H1: Usage of AI in companies is more acceptable in the area of finances (vs the area of human resources).
RQ1: Is the effect of application area on acceptance (described in H1) moderated by task stage?
3) Describe the key dependent variable(s) specifying how they will be measured. Acceptance:
- Benefit (1 item)
- Potential to handle future challenges (1 item)
- Potential to heighten success (1 item)
- Risk of negative consequences (1 item)
- Willingness to invest (1 item)
4) How many and which conditions will participants be assigned to? We will implement a two-factorial mixed-measures design. Participants will be randomly assigned to one potential application area of AI in the business context (between-factor: finances vs human resources).
Each participant will then be exposed to four consecutive task stages of information processing reflecting increasing autonomy of the AI (within-factor: 1. Monitoring, 2. Generating options, 3. Selection, 4. Implementation). The order of the tasks will not be randomized.
5) Specify exactly which analyses you will conduct to examine the main question/hypothesis. In case that the five measurements of acceptance (see point 3) form a scale with α > .7, we will average them into one index of acceptance and submit it to the analysis reported below. Otherwise, we will conduct the following analysis separately for each of the single-item measures of acceptance.
We will calculate a multi-level model regressing acceptance on application area and task stage (assuming a random intercept for participants).
Application area (finances vs human resources) will be included as level-2 predictor (fixed effect). Task stage (monitoring, generating options, selection, implementation) will be included as level-1 predictor (fixed effect) by using contrast coding ([c1] 0.5, -0.5, 0, 0; [c2] 0, 0, 0.5. -0.5, [c3] 0.25, 0.25, -0.25, -0.25). We will further include the interaction between application area and task stage.
Evidence for H1 will be provided by a main effect of application area, whereas evidence for RQ1 will be provided by interaction effects between application area and task stage.
6) Describe exactly how outliers will be defined and handled, and your precise rule(s) for excluding observations. Inclusion requirements for participants:
- Experience being in a management position
- Fluent in German (language the study is conducted in)
- Passing two attention checks and stating two have answered all questions attentively
7) How many observations will be collected or what will determine sample size? No need to justify decision, but be precise about exactly how the number will be determined. As participants matching our inclusion criteria are difficult to recruit, we aim to sample N = 150 valid cases. To account for data exclusion based on the above-described criteria, we will collect data from N = 175 participants (i.e., oversampling of appr. 15%). If no additional submissions come in within 72 hours while we have not reached our desired sample size, we will stop the first wave of data collection and resend the study invitation to suitable potential participants who did not participate in the study by then. After another 72 hours without new submissions, we will stop data collection.
8) Anything else you would like to pre-register? (e.g., secondary analyses, variables collected for exploratory purposes, unusual analyses planned?) We will further assess perceived competencies of existing AI tools (24 items) and innovative behavior (8 Items) as well as for each task stage usefulness (4 items), perceived autonomy, controllability, and task complexity (1 item each).
