Today’s news is filled with stories of data misuse — for example, Cambridge Analytica harvesting personal data via Facebook and Amazon, Facebook’s Android app logging metadata on every call and text, and Grindr revealing its users’ HIV status to third parties. These companies are being roundly criticized for misusing their users’ data. In the midst of this uproar, companies are quietly revising and strengthening their data privacy policies. Yet this data privacy wave was prompted not by the media outcry, but by a law passed by the European Union.
The EU’s General Data Protection Regulation (GDPR) is the most comprehensive, far-reaching law for how companies must handle with online personal data. Though it only technically applies to EU citizens’ data, it will end up affecting almost every online service since there’s no “European internet”. One of its strongest rules — every time a company collects personal data on an EU citizen, it needs explicit and informed consent from that person. This consent must be opt in (not opt out), and include a clear explanation of how the data will be used. Violate this, or any of the other regulations in the GDPR, and companies will be subject to massive fines (4% of global turnover or $20 million, whichever is higher).
The GDPR, which regulates the online world, mimics the rules for informed consent in the offline world. In surveys, interviews and other forms of data collection, informed consent is the cornerstone of ethical research. No matter the study, a researcher shouldn’t carry out research or collect people’s data without a good understanding of what informed consent is and how to get it.
What is informed consent?
Informed consent is a process where researchers inform participants about a study, and the participants willingly choose whether they want to participate.
Informed consent has four main components:
- Information: Participants must be given all relevant information. This includes what it will take to be a participant, the risks and benefits of participating, how the data will be used and protected, etc.
- Understanding: It’s not enough to just be given information; participants have to adequately understand the information. This means that a key part of informed consent is making sure that the information is communicated well and people understand it.
- Volunteering: Participants have to genuinely volunteer to participate, not be coerced, manipulated or persuaded in any way.
- Decision-making capacity: Informed consent requires that participants can weigh the risks and benefits and come to their own decision about whether they want to participate. It’s important to remember that some groups, such as children and people with mental disabilities, may not have this full decision-making capacity.
Informed consent may seem time-consuming or tedious, but it’s incredibly important. It shows respect for people and their choices, and it helps ensure that your research process is fair, well-considered, and non-exploitative.
A good research plan should always include informed consent. Learn more about 6 more things you should consider while planning for data collection.
When is informed consent necessary?
If you’re doing any research that involves people and their individual data, you need to get informed consent. This is equally true for electronic or paper surveys, interviews or focus group discussions, and shallow or in-depth surveys.
Children aren’t legally allowed to give consent, so research with children must get informed consent from the children’s parents. However, older children (usually, 7 years old or above) should also be involved in the decision-making process. Be sure to inform older children about what’s going on and give them the ability to agree to the survey.
Informed consent can be waived under certain circumstances. For example, Cornell University’s IRB (institutional review board) says that informed consent can be waived if there’s minimal risk, and a consent document will de-anonymize an otherwise anonymous and confidential process. The American Association for Public Opinion Research’s IRB has a slightly different set of rules. It says that informed consent can be waived if:
- the people involved will face no more than minimal risk,
- their rights and welfare won’t be adversely affected,
- and the research can’t be conducted with informed consent.
However, informed consent is generally seen as mandatory. Most research review boards require a special request and explicit proof of extenuating circumstances to waive informed consent.
Even researchers that don’t fall under a review board (for example, M&E professionals in a nonprofit) should assume that informed consent is mandatory for any research that involves people and their personal data.
How can you get informed consent?
Informed consent should be the very first thing that happens in a survey or research exercise. It includes two parts: first, an explanation of the survey; and second, explicit written consent.
The explanation can be given verbally or in writing. There are pros and cons to each:
- With written explanations, consistency is no problem. It’s easy to ensure that every participant gets the same explanation, given in the same way. However, people are more likely to skim (and thus not fully understand) written explanations, rather than read them in-depth.
- Verbal explanations can be more engaging than written explanations. However, it’s important that each researcher reads off the same script, and there should be some way to ensure that researchers are following the script.
The consent can be recorded in different ways. The most traditional method is a simple Yes/No checkbox. However, digital data collection apps also allow for more detailed ways to record consent, such as an audio recording or signature.
With digital data collection apps, skip logic or conditionality can also be useful for informed consent. If someone gives their consent, the rest of the survey can then be shown. If not, the survey can end with a simple thank you.
Digital data collection apps make informed consent a breeze. Here are 5 other ways they can improve your M&E process.
Note: Some organizations prefer to put informed consent into a separate survey. However, we prefer to include it as the first part of the larger survey. This ensures that a person’s informed consent will never be separated from their survey results, and it’s easy to add multiple sections to a survey on digital data collection apps like Collect.
What should be included in informed consent?
Informed consent should include all information that is relevant to someone’s decision to participate in the survey or research study. This can include any or all of the following:
- The surveyor’s identity: What is your name? What organization are you from, and who are they affiliated with? What does your organization do?
- Surveyor’s contact information: Be sure to include some sort of contact information, where people can go for additional questions. This can be as simple as an email address. However, the more information you provide, the more secure respondents will feel. Providing a phone number and address as well can go a long way toward making an unknown organization feel more reputable and trustworthy.
- The survey’s purpose: Why is this survey being conducted? What question is it trying to answer?
- How participants are chosen: Why is this person being asked to participate? How many people in total are participating? How have they been chosen?
- Voluntary participation: It’s important to explain that participation is entirely voluntary, that participants have the option to not respond to any questions, and that they can withdraw at any time without giving a reason. If there’s any sort of relationship between the respondent and surveyor, it’s also helpful to explain that participation or non-participation won’t affect that relationship. (For example, beneficiaries won’t be kicked out of a program if they don’t answer an M&E survey.)
Monitoring and evaluation can be a tricky process. Before you start your next M&E survey, don’t forget to check out the 5 best project M&E guides.
- Survey process: Explain what someone should expect if they participate. For example, what will the survey cover? What will participants be asked to do? How long will it take (or how many questions does it include)? What will happen after completing the survey (if anything)?
- Risks and rewards: If there are any disadvantages, costs or discomforts that come with participating, it’s important to explain them clearly to the participant. It’s also important to explain any benefits (such as financial incentives) during informed consent.
- Results: If the results of this research will become public, be sure to include a notice about this, along with information on when the results will go public and where they can be accessed.
- Data privacy: Explain how the data will be handled after the survey is complete, and what will be kept anonymous, confidential and/or private.
- Data use: Explain how the data will be used after the survey. For example, will it be anonymized? Will it be aggregated or kept at an individual level? Will participants be able to see the results? If you plan to use quotes or photos in public results, be sure to explain that too (and explain if and how they’ll be anonymized).
- Permission for recording: If you plan to record anything (either with audio or video), it’s important to inform the participants that they will be recorded, how they will be recorded, and what will happen to the recordings. You should include a specific informed consent question on this (e.g. “I agree to this interview being recorded on video.”)
- Thank you: Don’t forget to thank participants in advance for spending their time to help you with your research.
Best practices for getting informed consent
When setting up informed consent, it’s important to remember that the information must be understandable, not just explained.
- Make it short. It’s helpful to keep your script or text to something people can go through in just a minute or two. People have short attention spans, and their minds are likely to wander in the middle of a long explanation.
Keeping your text short is essential for building a great mobile survey. Learn 7 more tips for making mobile-friendly surveys.
- Keep it simple. Any information should be given in simple, non-technical terms so anyone can easily understand it. Keep text at an 8th-grade reading level or lower. (If you’re surveying kids, use an even lower reading level.)
- Break up text. Don’t be afraid to deviate from standard paragraphs. Bullets, lists, tables, pictures and charts are often easier to understand than lots of written text.
- Use “you”. Write the introduction in second person (e.g. “You’ll be asked to rate 10 questions on a 1-5 scale”) to make it clearer for the listener. The only time you should deviate from second person is the actual “informed consent” recording, which should use first person (e.g. “I voluntarily agree to participate in this survey”).
Many online services use opt-out consent, meaning that you automatically agree to sharing your data unless you say otherwise. However, this is not a good practice for informed consent. It should always be opt-in, meaning that you have to explicitly give your permission (in writing) to be included. Anyone who doesn’t respond or declines to participate should be excluded from any data collection or research.
Lastly, don’t forget to consider whether you’ll be asking participants private information about other people. If so, it is also important to get informed consent from the other people. For example, if you are surveying men in a village, you can ask them about their habits and preferences under their informed consent. However, if you ask them about their wives’ preferences, you should also get their wives’ informed consent (or just interview the wives directly).
Example: Informed consent script
Imagine you are working for an off-grid solar lighting distributor called Lightish. Lightish operates in communities in eastern India with the mission of empowering rural livelihoods through solar power. Lightish is seeking investment, and investors are eager to fund the organization’s impact. However, there is a problem. The enterprise has not collected data on its impact. Investors have balked at the idea of blindly investing in an organization that can’t validate or quantify its impact.
The company has asked you to design a survey that will collect data that on the life-changing effects of its products. As an experienced researcher, you know you must start the survey with informed consent. What might it look like?
Here’s an example of that informed consent script:
Hello, my name is [name here] and I work for a solar-lighting company called Lightish. Lightish was founded with the mission of improving life for people in off-grid communities by providing them with high quality solar lighting options. Currently, we are trying to gather data from our past customers to determine how well we’re fulfilling this mission. We’re doing this because a partner organization wants to understand our impact before they give us money to expand our program. As a customer, we were hoping that you might be able to provide us with some information about your experience with our products.
This survey will be conducted completely anonymously, as we are going to group all of the information we’ve gathered together. All surveys will only be identifiable down to the village level, which means that we will be able to identify trends within your village. We will post the summarized results publicly on our website and share them with our partner organizations. This information might be used to determine whether or not our work in your village should be continued and expanded, which is something you should consider before you agree to take our survey. Would you be willing to participate in a short 5-10 minute survey to help us understand this important information?