The objective is to measure the effects of different messaging strategies on intentions to nominate someone to the Order of Canada, as well as to survey different beliefs and attitudes about this Canadian honour, with a focus on populations of interest identified by prior research.

A randomized controlled trial will be conducted to compare the impact of four experimental messaging conditions on pre/post measures of intention to nominate to the Order of Canada, perceived barriers in the nomination process, as well as a post-messaging measure of interest in seeking out further information (e.g., clicking links to learn about the Order of Canada, dwell time).

1. Messaging will increase indication of willingness to submit a nomination to the Order of Canada.
2. Messaging will decrease indication of perceived barriers to the nomination process.

• nom_intent_pre / nom_intent_post: How likely would you be to do each of the following? (5-point Likert-type scale from Very Unlikely to Very Likely)
o nom_intent_pre_1 / nom_intent_post_1 / nom_intent_diff_1: Nominate someone to the Order of Canada
o nom_intent_pre_2 / nom_intent_post_2 / nom_intent_diff_2: Tell a family member, friend, or colleague about the Order of Canada
o nom_intent_pre_3 / nom_intent_post_3 / nom_intent_diff_3: Encourage a family member, friend, or colleague to nominate someone to the Order of Canada
o nom_intent_pre_4 / nom_intent_post_4 / nom_intent_diff_4: Notify or encourage a group or organization (e.g., civil groups, community or volunteer organizations, professional societies) that I know or am a member of to nominate someone to the Order of Canada
• nom_intent_pre_avg / nom_intent_post_avg: Index variables of the sets above, one average score for each.
o An additional difference score nom_intent_diff.
• nom_intent_pre_name / nom_intent_post_name: If you were to nominate someone to the Order of Canada, do you have someone in mind? (Yes / No)
• nom_barrier_pre / nom_barrier_post: What barriers do you anticipate if you were to try and nominate someone to the Order of Canada? Please select all that apply. (Selected; Not selected)
o The process of submitting a nomination taking too much time or effort
o Language barriers in the nomination process
o Not knowing enough about the nomination process
o Not trusting that the nomination process would be fair
o Not being confident that the nominee would be elite or prestigious enough to get recognized
o Not knowing if the nominee met the criteria to be nominated
o Not knowing if or assuming the nominee had already received the award
o Not being confident whether the nominee deserved it
o Not being confident that the nominee would be competitive
o Not being confident that the nominee would get recognized because of their background (e.g., ethnicity, culture, life circumstances)
o Not feeling like the award is relevant enough for me to complete a nomination
o Not feeling like the award represents my background (e.g., ethnicity, culture, life circumstances)
o Way of nominating being inaccessible or unintuitive to complete
o Encountering technical issues when trying to nominate
o Not being confident that my nomination would be processed in a timely manner
o Not knowing anyone to nominate
o Other (please specify)
• nom_barrier_pre_prop / nom_barrier_post_prop: Index variables of the sets above, one proportion score for each.
o An additional difference score nom_barrier_diff.
• Dwell time on pages of (1) messaging conditions or (2) links for more information.
• Number of clicked links on page with links for more information (range 0 to 3).

1. H1: Messaging will increase indication of willingness to submit a nomination to the Order of Canada.
a. t.test(nom_intent_diff)
b. chisq.test(nom_intent_pre_name, nom_intent_post_name)
2. H2: Messaging will decrease indication of perceived barriers to the nomination process.
a. t.test(nom_barrier_diff)

The survey panel sample (N=3762) breaks down as follows:
• N=3000 General population (primary sample): A representative sample of Canadians with nested quotas for age (18-24, 25-44, 45-64, 65+), gender (men, women), and region (13 levels = provinces and territories of Canada).
• N=250 Youth 18-24 oversample (~550 anticipated including primary sample)
• N=250 Rural or remote population oversample (~637 anticipated including primary sample)
• N=187 Member of visible minority group oversample (~982 anticipated including primary sample)
• N=75 Indigenous Peoples oversample (~225 anticipated including primary sample)
The overall sample, when divided among the 4 conditions, has sufficient power (0.8) to detect small paired-samples pre/post effects (d~.09). An omnibus ANOVA-type model with four conditions would detect small effects (f~.1) with over 0.99 power.
The oversamples were chosen based on results from prior research and discussion with the partner and survey supplier. The sizes of the oversamples were based partly on predicted rates of the oversample groups in the primary sample (leading to the “anticipated” numbers above) and partly on relative oversample costs in the context of the overall budget, rather than separate power analyses per se.

Participants will be excluded if they fall under the falling criteria:
• Incomplete protocols (< 100%) • Under 18 years of age • Not a resident of Canada • Flagged in Qualtrics internal quality control for removal • Taking fewer than ½ the median time of soft launch to complete the protocol • Taking more than triple the median time to complete the survey