Jump to content

Survey Support Desk/Learning

From Meta, a Wikimedia project coordination wiki

The Survey Support Desk
for Wikimedia Foundation staff, Wikimedia affiliates, and volunteers

[edit]
Contents

Survey Learning

[edit]

This page centralizes knowledge in about doing surveys in the movement and at the Wikimedia Foundation, using Learning Patterns to collect the information. Be sure to explore the links and add your own information as you learn about doing surveys. Learn more about Learning Patterns


Strategy, management, and communications

[edit]

Is a survey the right choice? How long does a survey take?

Strategy

[edit]
  • Defining the scope and scale of the survey. The scope and scale of a survey will depend on your overall strategy: Why do you need to do this survey? By when do you need the data? Answering these questions involves talking to other stakeholders.
  • Other types of surveys. Interviews, focus groups, consultations. These are actually types of surveys that can help you listen to people.

Communications

[edit]
  • Communications for surveys. There are many different forms of communications around surveys. Knowing each is important to help you plan ahead, and you might want to involve certain people, like external news organizations.
  • Involving communities and partners in your surveys. There are many great reasons for involving people in doing your survey, like improving your questions and getting additional helping hands.
  • Thank participants. It is hugely important to thank participants. They have volunteered their time to help you out. There are many fun and free ways to thank users.

Managing a survey

[edit]

Reaching people with surveys

[edit]

What is the best way to reach people with my survey

Choosing your audience

[edit]
  • Who should you survey? Sometimes, it is not entirely clear who you need to survey. Determining who you need to survey is based on the goals of your survey and how you will reach those people.

Reaching users online

[edit]

In-person surveys

[edit]
  • Using pen and paper surveys. Pen and paper surveys have their own design principles and process. Like having a few techniques to make sure people fill out the surveys
  • Group surveys or focus groups. Focus groups are great for brainstorming, learning about a new community, or other information at the group level. Recruiting and gathering data from focus groups is unique.

Translating Surveys

[edit]


Writing and designing questions

[edit]

What does a good question look like? What does a bad question look like?

Design process

[edit]
  • Asking the right questions. Asking the right questions is difficult. It involves making sure that the survey goals are clear, and that the questions and response options are designed well. If you have no experience with survey question design, the learning module "Designing Effective Questions" is a great place to get started.
  • Testing survey questions. There are generally three to four rounds of testing survey and survey questions. The first is testing with colleagues. The second is testing with the users you want to reach, which includes both question review and testing of the final survey software.
  • Specialized questions. Some questions are very particular. Whether its questions for WMF Grant programs, or questions related to private information (e.g. gender, age, etc.), or other questions that might be particular to the Wikimedia Movement.

Question tools

[edit]
  • Ordered response options No need to invent response scales (e.g. Agree/Neutral/Disagree). Try these ones first!
  • Survey templates '(coming soon).' Since many Wikimedians have been doing surveys, we are collecting surveys and making standard templates that can be used and re-used.
  • Use the Question bank for ideas. If you are looking for survey questions, use the survey bank to find question you might need.
  • Tools for question design. There are many, many resources out there to help you learn how to write a good question. Here are some favorites.

Privacy and data

[edit]

How do I ask for personal data, like gender? What are the policies for storing data?

[edit]
[edit]
  • Writing legal disclaimers. If a survey links from a Wikimedia page, you must include a disclaimer. The type of legal disclaimer will depend on where you live, whether inside the U.S. or outside the U.S.
  • Contests, giveaways, and sweepstakes. Many times, surveys include free gifts for random participants. There are many U.S. laws based on giveaways and sweepstakes. Make sure you are informed.

Open data

[edit]
  • Collecting data with personal information. Working with personal information (e.g. gender, race, age, etc.) has many laws. Learn the laws and practices for working with this data.
  • Open access policy. The Wikimedia Foundation has an open data policy. This includes posting information about your research on meta.wikimedia.org as well as sharing your data publicly.

Privacy of participants

[edit]

Privacy is an important value not only for the Foundation, but for Wikimedians as well.

  • When creating a survey, you should always be transparent about the types of information you will be collecting from participants.
  • When setting up a survey with a third-party service provider, consider reviewing their privacy policy carefully before you use their products or services.


Analyzing and sharing survey results

[edit]

How do I process the data? What is the best way to report survey data?

Reporting quantitative data

[edit]
  • Quick guide to presenting data. Most surveys should always present a basic list of data, like response rate, completion rates, and descriptives to help readers understand the quality of the data being presented.
  • Creating basic descriptives. Surprisingly, you can learn a lot with basic descriptive data. Learn about what this means and how to present it.

Reporting qualitative data

[edit]
  • Overview of qualitative analysis. Learn how to do some basic qualitative analysis.
  • Tools for qualitative analysis. There many tools to help you with qualitative data. Dedoose is great, and you can even use spreadsheets for simple qualitative analysis.