Dataset for: User-Centered Design for Open Government Data Platforms in Developing Smart Cities: A Discrete Choice Experiment in Adama City, Ethiopia

This dataset contains the raw, anonymized responses from a Discrete Choice Experiment (DCE) conducted in Adama City, Ethiopia, between [August, 2025] and [December, 2025]. The experiment was designed to quantify the preferences of key stakeholders for features of an Open Government Data (OGD) platform intended to support smart city development.

Study Overview:
The research investigates how potential users—specifically academic researchers, technology startup members, and final-year computing students—value different design attributes of an OGD portal. Understanding these preferences prior to platform development aims to improve user adoption rates and ensure the platform aligns with local needs and resource constraints.

Data Collection Methodology:

• Sample Size: 150 respondents.

• Participant Groups: The sample is stratified into three primary user groups: Researchers (n=50), Technology Startup members (n=50), and Final-year Computing Students (n=50).

• Experimental Design: Each participant completed 5 choice tasks. In each task, they were presented with 3 hypothetical OGD platform alternatives (Alternatives A, B, and C) and a "Status Quo" option (Alternative D). Participants were asked to rank these four alternatives from most preferred (Rank 1) to least preferred (Rank 4).

• Attributes and Levels: The alternatives were defined by five key attributes with varying levels:

  1. Search Functionality: (e.g., Keyword search, Keyword search with filters, Smart search).

  2. Download Format: (e.g., Basic files, Web APIs, Linked Data).

  3. API Availability: (e.g., No API access, Basic REST API, Full developer suite).

  4. Data Quality: (e.g., User-reported, Verified by source, Digitally certified data).

  5. Visualization: (e.g., Basic table preview, Static charts and maps, Interactive dashboards).

File Contents:
The dataset is provided in a single CSV file: ogd_survey_long_format_R.csv.

This file is structured in a long format, making it immediately suitable for analysis using statistical software such as R (e.g., with the mlogit, apollo, or gmnl packages) or Stata. Each row represents one alternative within a choice task, as ranked by a participant.

Variables included:

• ParticipantID: An anonymized unique identifier for each respondent.

• PrimaryRole: The stakeholder group of the participant (Researcher, Startup, Student).

• AgeGroup: Age bracket of the participant.

• Gender: Gender of the participant.

• Experience: Years of professional experience in a related field.

• UsedOGDBefore: Binary indicator (1 = Yes, 0 = No) of prior experience with OGD portals.

• Task: The choice task number (1 to 5).

• Alternative: The specific alternative being evaluated (A, B, C, or D).

• Is_Status_Quo: Binary indicator (1 = Yes) for the Status Quo alternative.

• Ranking: The rank (1 to 4) assigned to this alternative by the participant (1 = most preferred).

• Search_Functionality: The level of the search attribute for this alternative.

• Download_Format: The level of the download format attribute.

• API_Availability: The level of the API attribute.

• Data_Quality: The level of the data quality attribute.

• Visualization: The level of the visualization attribute.

Data Usage:
This data can be used to replicate the analysis presented in the associated manuscript, "User-Centered Design for Open Government Data Platforms in Developing Smart Cities: Evidence from a Discrete Choice Experiment in Adama City, Ethiopia." It is also suitable for secondary analysis, such as exploring different model specifications (e.g., latent class models) or subgroup analyses not presented in the original paper.

Ethics:
All personally identifiable information has been removed. Participants provided informed consent for their anonymized data to be shared for research purposes.