You need to sign in or sign up before continuing.
Back to History

Fields Changed

Registration

Field Before After
Trial Start Date November 26, 2025 January 12, 2026
Last Published November 19, 2025 01:42 PM January 02, 2026 05:48 AM
Intervention (Public) randomized vignette-based survey experiment randomized vignette-based survey experiment.
Intervention Start Date November 26, 2025 January 12, 2026
Intervention End Date July 31, 2026 September 30, 2026
Primary Outcomes (End Points) Our primary outcomes of interest are stated intention to apply, word-of-mouth intentions, and interest in collecting more information about the occupation. Our primary outcomes of interest are stated intention to apply, word-of-mouth intentions, the interest in collecting more information about the occupation and the choice-based outcome.
Primary Outcomes (Explanation) To analyze the effect participants will be asked how likely they would be to apply for the position (intention to apply) and how likely they would be to recommend the apprenticeship to a friend (word-of-mouth intention). Both outcomes will be measured on a 7-point Likert scale. The interest in collecting more information is assessed by asking participants whether they would like to learn more about the occupation (yes/no). If they answer “yes,” a link is displayed directing them to the corresponding occupation page on a job board as it would also be found by a regular search for this occupation. To analyze the effect participants will be asked how likely they would be to apply for the position (intention to apply) and how likely they would be to recommend the apprenticeship to a friend (word-of-mouth intention). Both outcomes will be measured on a 5-point Likert scale. The interest in collecting more information is assessed by asking participants whether they would like to learn more about the occupation (yes/no). If they answer “yes,” a link is displayed directing them to the corresponding occupation page on a job board as it would also be found by a regular search for this occupation. The choice-based outcome is measured by participants’ stated preferences across four choice sets in which they select between two alternative apprenticeship offers.
Planned Number of Observations We estimate that the final sample will comprise approximately 2,200 to 4,400 individuals. (This expectation is based on the fact that, in 2025, the job board counted 55,000 active users within our target group, all of whom will be invited to participate in the survey. Drawing on response patterns from previous surveys, we anticipate a response rate of about 5 to 10 percent. Among those who respond, earlier survey experience suggests that roughly 80 percent will complete the questionnaire in full. This anticipated range informs our expectations regarding the final number of observations.) We estimate that the final sample will comprise approximately 2,750 to 4,950 participants. This projection is based on results from a pilot survey conducted on the same platform, in which the response rate was around 5 percent without reminders and after accounting for attrition. With the use of additional measures such as reminder messages, we expect the response rate to increase by 2 to 4 percentage points, resulting in an overall response rate of approximately 7 to 9 percent. In 2025, the job board counted around 55,000 active users within our target population, all of whom will be invited to participate in the survey. Applying the expected response rate to this population yields the anticipated sample size range stated above.
Sample size (or number of clusters) by treatment arms approx. 350-730 individuals per treatment arm. approx. 305-550 individuals per treatment arm.
Intervention (Hidden) We present participants with a hypothetical apprenticeship description where the training firm offers an apprenticeship in one of three common Swiss training occupations. They receive information on an occupation that highlights opportunities to use and learn about AI, while the control group receives information on traditional workplaces and learning. Afterwards we ask respondents to evaluate their interest in applying for such an apprenticeship, in recommending it to friends (word-of-mouth intentions) and in collecting more information on the respective occupation. We present participants with a hypothetical apprenticeship description where the training firm offers an apprenticeship in one of three common Swiss training occupations. They receive information on an occupation that highlights opportunities to use and learn about AI, while the control group receives information on traditional workplaces and learning. Afterwards we ask respondents to evaluate their interest in applying for such an apprenticeship, in recommending it to friends (word-of-mouth intentions) and in collecting more information on the respective occupation. In addition, participants complete a short choice-based vignette task. They are asked to choose between two alternative apprenticeship offers for the same occupation, presented side by side. Each offer is described by a small set of attributes. The AI-related attributes are randomly varied between the two options, while the remaining job characteristics are held constant. For each choice task, participants indicate which apprenticeship they would prefer. This allows us to study how AI-related features in otherwise comparable apprenticeship offers affect adolescents’ stated job preferences.
Back to top