Executive summary

The digital transformation of public administrations requires much more than simply publishing content online: it calls for the information to be understandable to all members of the public.

In this project, STAR Servicios Lingüísticos participated in the linguistic evaluation of a wide range of institutional websites to verify their compliance with plain language principles. To do this, we designed an analysis methodology based on specific indicators that allowed us to evaluate the clarity of the content, the structure of the information, the terminology used and the ease of understanding for users.

Furthermore, we helped to improve an artificial intelligence system designed to automate future evaluations, providing the linguistic expertise needed to train the model using robust, consistent and reliable criteria.

The result was a methodology that combines the efficiency of artificial intelligence with the expert judgment of specialised linguists, contributing to improving the accessibility of public information on a large scale.

Project details

Sector: Public administration

Type of organisation: Public institutions

Services provided:

• Evaluation of content according to plain language criteria
• Language consultancy
• Design and validation of evaluation indicators
• Expert review for AI model training
• Preparation of technical reports

Applied technologies: Artificial intelligence and linguistic evaluation methodology

The challenge

Public administrations generate a large volume of digital information aimed at citizens with very diverse profiles, knowledge and needs.

The main challenge was to systematically evaluate the linguistic quality of numerous institutional websites, identifying aspects that made the content difficult to understand and proposing objective criteria for improvement.

At the same time, the project sought to develop a methodology that could serve as a basis for future automated evaluations using artificial intelligence, without losing the rigour required by specialised linguistic analysis.

Objectives

• To evaluate the linguistic quality of institutional websites
• To verify compliance with plain language principles
• To design objective and consistent evaluation indicators
• To generate reports useful for improving digital content
• To contribute to the training of artificial intelligence tools specialised in document evaluation

STAR’s solution

Initial analysis

Our team defined an evaluation model based on specific criteria related to the clarity of the content, the organisation of the information, the use of terminology and the ease of understanding for the public.

Working methodology

The evaluation was carried out by plain language specialists, applying a standardised methodology that ensured the consistency of the results across the various sites analysed.

At the same time, the linguists reviewed and refined the evaluation criteria themselves, identifying potential ambiguities, proposing methodological improvements and documenting observations that would enable them to obtain high-quality data.

Artificial intelligence integration

The knowledge generated during the project was used to train an artificial intelligence system aimed at automating future document reviews.

Human intervention proved essential to ensure that the criteria used by AI were accurate, consistent and applicable in different contexts.

Project development

The project was developed in five phases:

1. Definition of the evaluation criteria
2. Linguistic analysis of institutional websites
3. Review and improvement of the evaluation indicators
4. Generation of detailed reports with proposals for improvement
5. Validation of data for training the artificial intelligence system

Results obtained

The project allowed us to:

• Obtain consistent and objective evaluations of a wide range of institutional websites
• Identify specific opportunities to improve the clarity and accessibility of public content
• Generate useful reports for those responsible for communication and digital transformation
• Create a solid methodological basis for training artificial intelligence systems specialised in document evaluation
• Demonstrate that combining technology and linguistic expertise significantly improves the quality of analysis processes

The added value of STAR

This project demonstrated that artificial intelligence is a valuable tool for accelerating document evaluation processes, provided it is supported by expert knowledge.

The expertise of STAR’s specialists made it possible to define consistent linguistic criteria and resolve complex interpretative issues.

FAQs

What is a plain language content evaluation?

It is a systematic analysis that verifies whether the information is understandable to the intended audience, applying objective criteria of clarity, structure, terminology and readability.

Can artificial intelligence evaluate the linguistic quality of a website?

Yes, but its effectiveness depends on having been trained with criteria defined and validated by plain language specialists. Human supervision remains essential to guarantee reliable results.

Why is plain language important in public administrations?

Because it facilitates access to information, improves the citizen experience, reduces errors in interpretation and promotes more accessible and inclusive communication.

Related services

• Plain language
• Language consultancy
• Evaluation of digital content
• Artificial intelligence applied to language services
• Translation for public administrations

Would you like to improve the clarity and accessibility of your content?

At STAR Servicios Lingüísticos, we help public and private organisations improve the quality of their communication through solutions that combine linguistic expertise, a specialised methodology and artificial intelligence. Request advice.