Public health organizations in Denmark and most Organization for Economic Cooperation and Development (OECD) countries face an urgent demand for nursing staff. However, existing research does not yet offer a systematic understanding of how to attract talent to public healthcare organizations from the international labor markets.

In this study, the researchers will collaborate with a network of Danish public healthcare service providers led by the implementation partner Copenhagen Capacity to implement a large-scale megastudy (n=110,000), aimed at recruiting nurses and nursing assistants from across the European Union to Denmark. The study takes an employer-driven recruitment approach, co-designing and implementing a national recruitment campaign that uses social media job platforms to identify and contact potential recruits with a direct, personal message coming from a national recruitment agency.

The study aims to answer three research questions:

Research Question 1: Does receiving a behaviorally informed recruitment message (relative to a message with no employer value proposition) motivate a prospective employee to engage? Research Question 2: Does receiving any recruitment message (relative to receiving no message) motivate a prospective employee to engage? Research Question 3: What factors - that could be emphasized as employer value propositions in a short message - would motivate a prospective employee to engage?

Study participants will be randomly assigned to one of 10 treatment conditions in which they'll receive a recruitment message highlighting different employer value propositions or a simplified message including no employer value proposition, or to a control group that will receive no message.

Researchers will analyze differences across groups in terms of engagement after receiving a recruitment message, which can involve accepting, declining, responding to the message, clicking on a link within the message, and in this link registering to a national database to receive regular newsletters featuring job advertisements.

We define inclusion/exclusion criteria as follows:

• Their current/primary country/location listed on the public profile corresponds to any of the 26 European Union countries (all European Union member states except Denmark), as well as Norway and Switzerland.
• They have fewer than 10 years of work experience if their profile includes job entries with corresponding start and end dates. Profiles without job dates (preventing calculation of total experience) or with no listed work experience will also be included in the sample.
• Their public profile includes any of the following terms in their job titles, either as full terms or partial matches, in English or any local language: "Nurse", "Nursing Assistant", "Registered Nurse", "Intensive Care Nurse", "Emergency Room Nurse", "Surgical Nurse", "Travel Nurse", "Nurse Practitioner", or "Nursing Student" Additionally, their education section lists any of these terms, either as full terms or partial matches: "Nurse", "Nursing", "Registered Nursing/Registered Nurse" or "Nursing Science". We exclude job titles that indicate roles outside the scope of our interest. Examples include: "Midwife", "Dental nurse", "Veterinary nurse", "Occupational health/company nurse", "Cosmetology nurse", "Military/Fire nurse", "Retired nurse", "Nurse consultant/advisor", and "Nurse teacher/professor". In addition, we drop any positions or education records where the job/degree title appears 5 or fewer times in the sample.
• Individuals with five or more connections on their LinkedIn profile.
• Individuals with an active profile (i.e. a profile that has been updated) in 2024.

To avoid inadvertently disclosing participants' sensitive personal information, we cannot provide the full megastudy data publicly. However, we will provide a reduced dataset that includes the trial arm assignment, outcome variables, covariates, and an anonymized ID representing each individual. We restrict precise geographical information (e.g., replacing postal codes with broader regional descriptions) to ensure that no combination of variables can identify a group smaller than five individuals. This approach allows researchers to reproduce our main results without risking the potential de-anonymization of participants. De-identified data and materials will be made publicly available on Open Science Framework (OSF).

Recruiters interacting with candidates will see only the recruitment message text they are sharing but will remain unaware of specifics of the design and pre-analysis plan.