Polls can be used to collect opinions from users on some concerns for organizations.

Depending on the type of poll, users can vote on one [or more] of the options available.

Users can only vote once in a poll.

Preparing for a Poll

Polls may be published for a variety of reasons. The poll may have been requested by someone in the Municipality, or it may assist in planning for future interactions with community. Polls can also be used to gather feedback from users.

When you are asked to publish a poll, or are thinking of publishing one, ask the following questions, and help the poll requester answer them if needed:

1. What information are we looking to gather from this poll?

Is this poll asking users for feedback on an event or project that has already happened? Does the poll requester want community input in choosing floral designs for a roundabout? Does the municipality want to learn more about the community’s needs?

Before phrasing the question and the answers, ensure you have a clear idea what sort of information is required.

2. What decision will be made based upon the results of this poll?

Polls in Freejna are designed to inform decision makers, and are a channel for public participation. As such, polls shouldn’t be used for pure marketing purposes, and should be designed with a specific decision-making process in mind. Make sure there is a specific decision being made based on the results of a poll. This brings meaningfulness to the poll, and builds up a sense of public participation and buy-in in the community.

3. What action will be taken based on the results of the poll?

Will the feedback inform how future events or projects are executed? Will a specific design be implemented based on the community’s choices? Will learning more about the community’s specific needs with regards to playground infrastructure be used to inform and direct specific planning and development projects?

The poll requester should have a clear execution plan for each possible result of the poll. This execution plan should be communicated in a poll response. All polls posted on Freejna should, upon completion, receive a response.

At a minimum, the plan should have a timetable of execution for the resulting options. According to the decision being made, additional information should be available such as quotes for the implementation of each option, funding allotments, the executing contractor and other relevant information.

Do not publish a poll which does not have a clear plan of action, or which may have a result that is not executable by the poll requester.

 

Before proceeding with the poll, all three of these questions should have clear, coherent answers. A poll that is published without clear answers to the above three questions is a bad poll.

 

Designing a poll

Freejna provides many options for polling your communities. Polls can include text, images and videos, the answer options can include text and images, and you can provide up to five answer options. Polls can be published in multiple languages, to address communities that speak more than one language.

There are some best practices you should follow, and some pitfalls you should avoid. these are listed below.

 

Questions

When designing a poll question, take into account the following best practices:

 

• Provide the context.

What is the context of your question? All poll questions should start with informing the reader what the context of the poll is. This serves a twofold purpose: making sure the user is clear on what the question is about, but also informing users and letting them know about things they may not hear about otherwise.

• Give agency to the community.

When phrasing the question, make sure to give agency to the community by using sentence structures that give weight to their choices. If the poll is consultative, ask about what action should be done, not what action is preferred. Be specific about who is responsible for enacting the result.

• Be clear about why this question is being asked.

The poll question should inform users about the reason for it being asked. The question should have a clear rationale for being answered via a poll on Freejna. This is sometimes clear from the context and other information provided, but other times has to be specifically mentioned. If a user is asked ‘why is this question being asked’, they should have a clear answer.

 

BAD EXAMPLE: Which bench do you prefer?

GOOD EXAMPLE: The municipality is re-designing heritage park. Work will start on the 20th of June, and should conclude on the 15th of August. As part of the redesign, benches will be placed in locations a, b and c of the park. Which of these benches do you prefer?

By providing the context of the question, users know exactly what they are being asked about, and you have also informed them of the project. The user can also infer from the information available that the municipality wants to know which bench design is preferred by the community, so that it can install it.

BETTER EXAMPLE: The municipality is re-designing heritage park. Work will start on the 20th of June, and should conclude on the 15th of August. As part of the redesign, benches will be placed in locations a, b and c of the park. Which of these benches should we install?

This improves upon the example by giving the user agency. Rather than asking for their opinion, you are asking for their direction. This creates a sense of ownership.

This also avoids saying ‘Which of the benches should be installed’, and uses ‘Which of the benches should we install’ instead, which shows responsibility towards the community for responding to their choice.

BEST EXAMPLE: The municipality is re-designing heritage park. Work will start on the 20th of June, and should conclude on the 15th of August. As part of the redesign, benches will be placed in locations a, b and c of the park. We have asked contractors for quotes, and after receiving 50 proposals, we have whittled them down to the five below. Of these five benches, which one should we install?

This further improves upon the example by informing the user of the process, and shows them that they are taking part in something real. This creates a feeling of participation and buy-in.

 

There are common pitfalls in polling which you should avoid when designing the question for a poll. Make sure you avoid them when designing a poll question:

• Leading questions 

Poll questions should be phrased so that they do not induce bias in the answers provided.

Leading questions are questions that implicitly or explicitly guide respondents to answer in a specific way. For instance, by saying ‘we are working towards being the highest rated provider’ before asking about how the responded would rate their service. this should be avoided, as it means respondents will not be providing unbiased information. Leading questions could also be embedded in the question itself, for instance, by asking ‘how good is our service’.

• Double-Barreling

A poll question should only ask about one thing. In the bench example before, also asking about paving is bad practice. Instead, ask about paving in a separate poll.

 

BAD EXAMPLE: How much did you like last night’s event?

GOOD EXAMPLE: How do you feel about last night’s event?

 

Answers

Each Freejna poll question can have two to five answer options. Each answer can contain text and an image. Consider the following best practices when setting answers:

 

• Equal prominence

Answers should be equally prominent, and should not bias respondents via factors irrelevant to the decision being made based on the results of the poll. There are a few factors that should be taken into account:

Answer length

The length of all answer options should be similar.

Image use

All answers should either have images, or not have them. Do not have some answers with images, and some without images.

If images are used, make sure they are all of the same high quality. Use images that are all the same file type, with identical dimensions and specifications.

Images should differ only with regards to the choice that is being made. EG: if respondents are being asked to chose between three different logo options, the three images used should have the same background color and be renders. Do not have one option with a black background, another with a white background, and the third option be a printout.

All images used in the same poll should be of the same type. EG: Do not mix photos and renders.

 

• Balanced options

When asking about respondents’ feelings towards a subject or when soliciting feedback, make sure you use answer options that are worded in a manner that is unbiased towards one value or another. EG: When asking for a respondent’s opinion on service quality, provide options for bad, neither bad nor good, and good. Providing options which are limited to the positive end of the spectrum such as ‘good, very good and excellent’ is bad practice and should not be done.

In such opinion polls, also ensure that the median value is phrased neutrally. Do not use ‘acceptable’, use ‘neither good nor bad’.

 

• Neutralising options.

Know when to include an option for the respondent to say “I don’t know/No opinion” (IDK/NO). IDK/NO answers can provide significant information and are encouraged in the context of Freejna polls. To use the example of benches from the previous section, a significant amount of IDK/NO responses in such a poll would indicate low interest in all of the options, and would be grounds for reissuing the poll with different answer options. An attitudinal poll can include IDK/NO answers to avoid fabricated opinions, or to allow respondents to exclude themselves if they do not match the poll’s criteria.

 

Publishing a Poll

You can save draft polls without having to publish them. Use this to prepare and edit a poll before publishing.

When publishing a poll, you can set a time and date for publishing. select a date and time that is sensible. Use experience and Freejna’s admin tools to understand when users are active.

When setting the poll length, avoid setting a length shorter than 3 days, and longer than two weeks. setting a duration that is too short can exclude users who use Freejna infrequently from participating in the poll, and polls that are too long can hold up any decisions that will be made based upon the poll.

 

 

Poll Completion

While poll results are sometimes viewable before the completion of a poll, it is best to wait for poll completion before communicating poll results with stakeholders.

Upon poll completion, you should communicate the results with the poll requester.

Every poll that is published should be followed by a poll response. A response should be posted within a week of the poll closing.

Poll responses are used to communicate the actions that will be taken as a result of a poll. Before publishing a poll, make sure the poll requester has provided at minimum an outline of the responses that can be posted based on the answer options available. Help the requester write these if necessary. This will help you make sure that the poll requester will be taking the poll seriously, and will maintain the integrity of Freejna polls for your community.

 

Writing a Poll Response

Once a poll closes, you can see the poll results and should share them with the poll requester. You can also publish a poll response.

Poll responses are an important part of the polling process in Freejna. All polls should be responded to upon completion. You should aim to respond to a poll within a week of it closing. Do not leave polls without a response!

Poll responses are a tool that let you communicate to the community how their opinion will be taken into account. Poll responses are an important part of community engagement, as they explicitly inform the community how their opinion is part of the decision making process.

When writing a poll response, be sure to include the following elements:

• An introduction, mentioning the poll and linking to it.
• A summary of the poll results.
• The actions that will be taken as a result of the poll. This should include a timetable if applicable.

It is advisable to also include:

• A description of the process involved in creating the poll.
• An explanation of how the poll results were factored into the final decision, especially if the actions being taken do not match the most voted option.
• Images and other associated media that can highlight how the results of the poll will be implemented.

 

Considerations for polling via Freejna

There are some things you should consider when using Freejna to conduct a poll.

Freejna does not provide information about the respondent demographics of ethnography.

Freejna does not have verified information about a community’s demographics. Information provided by users about their age, location and such has not been verified.

Freejna does not support multi-question surveys.

Freejna polls will always have some degree of self-selection bias.

Freejna polls do not provide the ability to randomly sample from a selected population.

 

Managing polls for a Freejna community

There are some things you can do to make the most out of Freejna polling.

Due to how the timeline in the Freejna app works, polls generally receive the most votes the day they are published. Time your publishing so that the poll has the most visibility when your community is most likely to be using the Freejna app.

Avoid posting more than one poll at a time. Try to space polls out so that one or two are posted in your community at most on the same day. This reduces poll fatigue, and allows your community to fully engage with the poll before answering it.

Try to time the publishing of new polls, so that they are published a few hours before a live poll closes. This can increase the votes for the closing poll.

Try to always have a live poll for your community to vote in. Depending on how active your community is and how many members it has, you may want to have more polls active at the same time.

Understand what the poll requester needs from the poll. What is a statistically significant sample for your community? How many people usually vote in your polls? What factors impact engagement? Help increase the usefulness of Freejna polls by building an understanding of these factors.