Research Proposal

Introduction to Plain Language in Digital Public Health Campaigns

Clear language in digital health campaigns is essential for ensuring that accurate and safe health information reaches the public. However, research indicates that many health communication materials exceed the comprehension levels of individuals with limited health literacy, thereby excluding a large portion of the population from receiving critical health information (Rudd et al., 2004). The use of plain language, which is clear and easy to understand, can bridge this gap between public health campaigns and their intended audience by presenting visual and written materials in digestible ways (Randell et al., 2025).

Nonetheless, the effect of plain language on trust, credibility, and behaviour intention is under researched in digital health materials. Current research by Sayfi et al. (2024) and Stallwood et al. (2023) has been carried out on plain language presented in a static, written format. And while these studies have demonstrated that participants preferred plain language, the evidence is mixed on the relationship between clarity and trust (Stableford and Mettger 2007).

The purpose of this research study is to address the gap in knowledge between digital health campaigns and audience perceptions of the messages in relation to credibility, trust, and behavioural intentions. By using real-world examples of public health messages such as videos, infographics, and social media posts presented in plain and standard language versions, this study aims to contribute empirical data to the ongoing debate between clarity and trust of digital health communications.

 

Research Questions and Hypotheses

Research Questions

1. How does the use of plain language in digital health campaigns, compared with standard language, affect the perceived credibility of the message?

2. How does the use of plain language, compared with standard language, affect the perceived trustworthiness of the message?

3. Does the use of plain language increase the likelihood of participants acting on recommendations in the message?

4. How do participants perceive plain language versus standard language in digital public health messaging?

Hypotheses

1. H1 (Credibility)
   Plain language health messages will be rated as more credible than standard language messages.

2. H2 (Trustworthiness)
   Plain language health messages will be rated as more trustworthy than standard language messages.

3. H3 (Behavioural intention)
   Plain language health messages will lead to higher self-reported behavioural intention than standard language messages.

4. H4 (Perception)
   Participants will express positive perceptions of plain language messages than of standard language messages in their open-ended responses

Data Gathering

A mixed methods study consisting of two parts will be conducted. The first part will consist of a quantitative survey using Likert-scale questions, which provide more insight than simple responses to questions (Batterton and Hale, 2017). The second will be an optional interview with open-ended questions. The example questions for the survey and interview are provided in appendix 1.

The materials selected for the study will be categorised using plain language standards set by the Centers for Disease Control and Prevention (2025), which recommends that communications are:

• Written in active voice.

• Use words that the audience is familiar with.

• Limit the average sentence length to twenty words and one idea.

• Limit paragraphs to five sentences and one topic.

• Use “you” and other pronouns.

 

The participants of the study will be recruited via social media and university mailing lists, with an expectation of 100 quantitative surveys and 10 interviews completed over the course of three weeks, with two reminders sent during that time. Participants will be asked to identify basic demographic information such as age, gender, location, and highest level of schooling. Participants will receive all relevant information outlining the purpose, procedures, and voluntary nature of the study, with no personal identifying or sensitive health information collected or stored at any point. Participants may choose not to complete the study at any time. 

The materials will consist of four digital public health campaign samples from real-world campaigns: two in plain language and two in standard language. The samples will be categorised as plain language and standard language versions, using the CDC’s recommendations for plain language messaging. The materials will be sourced from official government social media accounts from English-speaking countries, such as the United States, Canada, Ireland, and the UK. An example of an official social media communication is provided in appendix 2.

A multi-part survey will be created via Google Forms, broken down into sections for each sample message. Each section will contain Likert-scale questions, with an optional section at the end for users to opt-in to be contacted for an interview.

Procedure

1. Participants will view a series of digital public health campaign messages presented in random order. Half of the messages will be written in plain language and half in standard language. All messages will be selected from real-world digital health campaigns.

2. After viewing each message, participants will rate it on several Likert-scale items measuring clarity, credibility (believability), trustworthiness (reliability), and behavioural intention (likelihood of acting on the recommendation).

3. Upon completing all message evaluations, participants will answer demographic questions, including their age, location, and highest level of education.

4. When submitting the survey, participants will have an option to provide their email to be contacted for an interview.

5. Participants who provided their email will be contacted and asked to respond to open-ended questions about each message in an interview setting, allowing them to elaborate on their impressions.

Data Analysis

Participants will rate four digital public health campaign messages (two in standard and two in plain language) on four Likert-scale variables: clarity, trust, credibility, and behavioural intention. The resulting quantitative data will be analysed using repeated-measures analysis of variance (ANOVA) to determine whether the four messages differ substantially. If a significant systematic difference is identified for any variable, follow-up post-hoc comparisons will be conducted.

Descriptive statistics, including means, standard deviations, and confidence intervals will also be calculated using open-source statistical software (Jamovi). Finally, the results will be presented in tables and figures, including bar charts showing mean ratings with error bars to illustrate variability.

The qualitative data from the open-ended interviews will be coded using thematic analysis following Braun and Clarke’s (2006) guidelines. This method involves systematically identifying, organising, and interpreting patterns of meaning across the data. Using thematic analysis will provide a deeper interpretation of participants opinions on plain versus standard language and their relation to the variables of clarity, trust, credibility, and behavioural intention.