Ethical concerns for running a Survey
Graphic design is more than simply aesthetics; it is also about problem-solving and satisfying the needs of your target audience and to do that you need to understand your audiences and figure out what they like fully and don’t like. That’s where research comes in. According to Esposito’s (2018) article, she wrote on the importance of design research as a critical step in not just understanding customers’ behaviour but also a way to gather actionable insights to improve one's design. How does one gather said actionable insights? Through gathering data from users by conducting surveys (Williams, 2007) but one should bear in mind all ethical considerations to show respect for research participants and maintain one's integrity.
Why is ethical consideration crucial? Beyond its role in preventing harm and ensuring the protection of research participants, ethics also encompasses the need to be sensitive when asking questions that might be inappropriate (Takeda, 2021), assuring participants anonymity and being unbiased to achieve honest data.
In this blog post, I will be critically analyzing the major key points surrounding ethical concerns graphic designers must consider when conducting surveys.
1. Informed Consent
This is a crucial step when conducting any survey. Crow (2007) writes on the importance of informed consent whereby the participants must have full knowledge about the project they are participating in, consent freely to participate and can decline or withdraw at any point of the study without any consequences. Contrary to this, (Pols, 2017) article mentioned another way of getting user data in what is tagged as “Hidden Design” whereby informed consent isn’t needed but rather humans are studied in their natural setting to see how they react to a particular design. This is also a way of getting honest data that doesn't require a questionnaire. A good example is a street campaign ad where pedestrians are cajoled into engaging with a design. A designer could derive data from this by studying the psychological impact it has on the participants.
Recording the amount of people that engage with this image would inform a research on the average number of a certain demographic who would stand against physical abuse.
2. Consideration for Vulnerable Participants
Who can be classified as “Vulnerable”? These are participants usually below the age of 18, in a disadvantaged or marginalised position. It should be noted that the need to use vulnerable people for research especially persons below the age of 18 should only be essential if their data is required to answer the research question and consent is usually provided by their parent/guardian before them. Having vulnerable research participants might be very tricky due to the rules one must adhere but in most situations, they must be used to enable graphic design researchers effectively come up with solutions for research questions that are important to their health and social and mental well-being.
A good example is designing a stabilizing smart spoon that helps those with hand tremors eat with ease would require data collected from possibly disabled participants to fully understand the key attributes the spoon should have.
Source: Pinterest
Another example is creating an app design for young teenagers suffering from mental health. Data have to be collected from them to know suitable colours, imagery etc they would be comfortable with.
This is a brand logo of an app targeted to young teenagers suffering from mental health. I utilized a calming colour pallete and simple typography after researching on what would likely appeal to such users.
Source: Adrian Agbi Designs
3. Confidentiality and Anonymity
Keeping and respecting participants' confidentiality is another crucial aspect to consider when conducting a graphic design survey. Whether the participant is a family member or friend, it is important to protect the person’s intellectual property to build trust because participants may share sensitive information they might not want to disclose to other individuals (Hecker).
It is also advisable to also assure participants that their information will be kept confidential and anonymous. This assurance encourages them to provide truthful and complete responses, to obtain credible data results.
4. Privacy Protection
It should be known that it is the sole responsibility of the graphic design researcher to safeguard the participants’ data. Violation of privacy constitutes a threat to security which could incur the situation to be taken up by the law and can lead to the research paying hard costs such as financial penalties or soft costs such as damaging one's reputation (Lee,2017). Regardless of the information being safely protected, a situation could arise whereby some distinctive characteristics would make the participant identifiable and this is termed “deductive disclosure”. To tackle this, Kaiser (2009) suggests a “dominant approach” which entails three key points of protecting respondents' confidentiality whereby after data is collected, it is cleaned to remove any characteristics of it being used to identify a particular person before it is finally published. Below is an example of a wrong way of collecting data that can be identifiable and a right way data can’t collected cannot be identifiable:
Wrong X: A female graphic designer in Postgraduate design from Brazil.
If there is only one female graphic designer from Brazil in Postgraduate design her privacy is breached and she is identifiable which could lead to the participant not giving an honest answer to the survey questions. A cleaner approach would be to remove her country of origin.
Right ✔: A female graphic designer in Postgraduate design.
In this scenario, she might not be the only female graphic designer in PG design and would feel assured she cannot be identified as a participant therein able to give honest answers.
5. Unbiased Questioning
When preparing survey questions, it should be meticulously structured to enable the respondent to give genuine opinions. Unbiased survey questions are meant to prevent the researchers from making assumptions and enable the respondents a chance to give impartial feedback. It should be simple, neutral and straightforward. Two major characteristics that could tint your survey questions are using leading words and assumptions (Zight). The discussions below will provide some examples of unbiased questions and will analyse the impact of the questions:
Wrong X:
Leading words: Like most designers, do you feel Adobe Photoshop is better than Affinity designer for designing?
Assumptions: Are your favourite software Adobe Photoshop and Affinity designer?
Questions like this usually lead the respondent to agree with the general public view or gear towards a particular biased answer. Unbiased questions should aim for neutrality and should be void of leading words or assumptions. Fixing the above questions it is better to ask:
Right ✔:
Which software do you usually use?
What is your favourite software?
The questions above are not partial to any particular software or aren’t trying to say a particular software is bad or good but are only aimed at gathering the specific data of knowing the respondent's preferred software. In summary, it is important to have unbiased survey questions when designing survey questions to have accurate, reliable, and ethical data.
Special attention must be paid to vulnerable populations to prevent any potential harm. Participation should always be voluntary, and power dynamics should be recognized and mitigated, particularly when dealing with sensitive topics.
Conclusion
Conducting ethical surveys shouldn’t just be seen as following some rules and guidelines but rather as tools needed to build trust, show respect and ensure the safety of your participants. It is advised that graphic designers memorize these key points to avoid violation of the guidelines.
Which guideline stood out most to you?
What are the best ways to create questionnaires and interview questions?
In most situations, respondents taking a survey might be left confused, and frustrated and end up leaving the survey halfway. This could be because the survey questions were not drafted out appropriately. Whether in a traditional face-to-face interview, a telephone interview, or a questionnaire. This guide provides you with the knowledge and skills needed to create and design questionnaires and interview studies and to enhance your research. I hope this guide will help you to create effective questionnaires and interviews and will help you gain valuable insight into your research.
Define your Objectives
First of all, do your research. Graphic designers should keep their main objective in mind and know the kind of information they wish to gather from the questionnaire. In some situations, there might have been similar works done by other researchers that one could borrow some ideas or draw inspiration from. As a graphic designer, the main objective of your research is usually to get data which would help aid the design thinking process. In other to get this data you should consider the choice between using the probability sampling method of selecting respondents at random (anyone in any place can participate) or the non-probability method of selecting participants through non-random criteria (users/consumers age, location, traits, characteristics etc). Most research is usually targeted to certain groups and these groups might pose certain guidelines and restrictions to also bear in mind.
Be simple and straightforward
Now that you have concluded what your research objective would be and also identified who your target participants would be, the next step should be tailoring your questions to suit your participants. Your questions should be simple and easy to read. Avoid using jargon, uncommon words or abbreviations and putting two questions together to enable the participants to understand the questions easily. For interview questions avoid words that can easily be misheard for example a question like “What is your opinion about sects?” might end up getting a different reply.
Having to craft the right words is one of the hardest tasks in copulating your questionnaire because you might stand a risk of introducing bias or leading questions if you’re not careful or make the questionnaire too boring to participate in.
Always keep in mind a questionnaire should be concise and should take a maximum of 15 minutes to answer.
Choosing your question type
Keeping in mind the two major types of questions which are open-ended questions and closed-ended questions. Open-ended questions are questions participants are required to answer in their own words for example:
Question: “Why do you like the logo?”
Ans: I like the logo because….
Consider how long it would take a participant to phrase an answer. Open-ended isn’t usually recommended because it take time and effort which might lead to the participant not completing the questionnaire.
Close-ended questions on the other hand are more flexible and less time-consuming and usually the best choice for collecting data and quantitative variables. It gives the participant an option of choice and can be structured in different ways (Single response, multiple choice, Rating scale, ranking scale, Likert scale). Example of a close-ended question:
Do you prefer to work remotely?
Yes
No
Maybe
Both types of questions undoubtedly have their pros and cons and some research questionnaires end up having a mixture of both.
Avoiding Bias/Assumption
Always keep in mind one of the end goals of any quantitative research is to get genuine responses from your participants. To achieve this, endeavour to make your questions unbiased and as neutral as possible. Avoid leading questions because they tend to guide respondents towards answering in a biased way rather than how they particularly feel. Also keep in mind the geographical/regional differences of your participants. A question like “On a scale of one to 10 how would you rate 3D billboards?” wouldn’t make sense to people in developing countries that probably have never seen a 3D billboard before. Finally, asking sensitive questions is also very tricky being that participants may lean towards answering to save their virtue. For example:
“Do you like gambling?”
Most participants who gamble would either avoid answering this question or answer dishonestly.
A better approach should be:
“How many times have you gambled in the last six months?”
Keep all these in mind for integrity in your research.
Arranging your questions
This is a key factor to consider being that a disorganised layout of a questionnaire might make the participant not complete the questionnaire. Do not rush into asking complicated questions but rather start with simple questions like age, gender, race e.t.c before moving into more complex questions in other to build trust and make the participants feel comfortable. Also, a not well-arranged question order might lead to biased responses because the earlier questions might influence the reply to the current question. For example:
1. Which famous brands can you recognise?
Apple
Nike
Zoe Franco
Adidas
Pandora
2. Name 3 famous brands
The arrangement of these questions would most likely affect the outcome of the answers and lead to biased answers.
Testing it out
Questionnaires and interview questions should go through a screening or pre-test stage usually by colleagues, friends and family to check that the questionnaire reads well and is understood, it is easy to follow, how long it takes to fill the questionnaire (remember to aim for 15 minutes max) or answer the interview questions and finally to fix any errors but grammatical or phrasing. It is also advised to discuss with pilot participants to gain knowledge of their experience. This evaluation phase is very crucial because it is before the final launch of your questions.
Example of Questions for Questionnaires
Participant Statement (Example)
You are being invited to complete an online survey as part of an MSc course being undertaken by Adrian Agbi, a Graphics Design and Branding student at the University of Hertfordshire, UK.
Please read the following information carefully before deciding whether to take part.
You are eligible to take part in this study if you are 18 years and above.
Research Topic:
The Influence of Brand Identity Design on Customer Recall and Loyalty in United Kingdom.
Research Aim/Objectives
This research aims to investigate the impact of brand identity design on customer recall and loyalty specifically in the United Kingdom, where it explores how visual elements, such as logos, typography, and colour palette, shape customers' perceptions and attitudes towards a brand. The objective of this research is to identify specific design elements that have the most significant impact on customer recall and loyalty.
Sample Population
The research is targeted towards a diverse group of members of the public specifically 18 years and above, preferably living in highly populated city areas who engage with brands through various mediums (online or physical stores). The research aims to invite 400 people to take part in the survey and expects at least a 25% response rate to get a minimum of 100 respondents.
Method
This quantitative research will be conducted as an online survey by using a non-probability sampling method because the respondents will be selected based on non-random criteria (age, location and brand engagement). The research also targets this demography of respondents because of the higher estimated percentage of literates in the area and first-hand experiences with both online and physical stores.
Risks
There are no expected risks for participants. Any data provided by the respondents in this research will be treated as confidential and the participants are assured of anonymity (no one will be able to identify the respondents or their answers). Notwithstanding, using the non-probability sampling method has its drawbacks of running at a high risk of sampling bias but this is mitigated by having already defined my target participants, aiming for a high population of survey invitees and utilizing both open-ended and close-ended questions to save participants time and encourage them to leave honest answers and ensure validity of the research.
Below I have attached a copy of an example of a questionnaire. I hope it can be helpful (Click to download).
Critical Reflection/Conclusion
As a graphic designer, it might be a challenging quest to get information on how to prepare a survey for quantitative research because there is hardly any information on the internet that points directly to how graphic designers should survey. Most information available is usually generic and encompasses all fields of study (but can still be used as a foundation/guideline to follow). But not to worry, as a professional in the field of graphic design, I have been able to interpret generic findings to suit graphic designers.
The first step to consider is how you draft out the questionnaire. Graphic design is a field that aims at creating visual content to communicate ideas and messages. The end goal of graphic designing is to communicate with the public and this should be one of the deciding factors in creating your questionnaire. Your questionnaire should be tailored towards exploring consumer/user preferences or getting feedback/experiences when using a product or service. The data collected should be geared towards aiding the design thinking process in other to create designs that are appealing to consumers/users.
Having that in mind, the next aspect to consider when creating a questionnaire is who it is targeted to. You definitely cannot collect data from everyone (you’re not doing a national survey, lol) you need to be specific by using some personality characteristics to narrow down the population sample like demographic information (age, location, gender etc.), preferences (for example a preference between two brands), or behavioural traits (creativity, psychological etc).
The next step is to look at the layout of the questionnaire. The language should be clear, concise, and straightforward and each question should be relevant to your study. The layout should have a flow usually starting with demographic questions (age, location, gender etc.) and easing the respondents into the major research questions.
Avoid using a single method of asking your questions, be creative and fun! Explore different question types like the Likert scale, multiple choice, open-ended questions, closed-ended questions etc to increase respondents' completion and genuine responses. Also avoid identifiable questions like name, address, and date of birth because these could affect the aim of assuring anonymity and prevent the respondent from giving honest answers. My suggestion would be to refine your questions a couple of times till it fits just right. Do a pilot test by asking friends, family or colleagues and use their feedback in amending grammatical errors or tone of voice
So you know your sample population and your questionnaire layout is ready…time to conduct your survey right? Wrong! Make sure you have all ethical considerations in check and you have received ethical approval especially if you are conducting your research at a university. Do this by considering if there are vulnerable participants and the risks involved, ensure all your respondents give their consent and are assured that they can opt out of the survey at any given time with no reason before answering your questionnaire to save you from any legal issues that may arise (better safe than sorry) and finally assuring respondents that their answers are private and anonymous to avoid bias or dishonest answers.
Okay, now you are ready! I would like to hear from you after your survey. How was your experience? Write it in the comments.
Reference
Beck, C. (2019). How to Design With Ethics. Medium. Available at: https://medium.com/ux-power-tools/how-to-design-with-ethics-dd0b6ec9c503 [Accessed 10 Apr. 2024].
Bhandari, P. (2022). Questionnaire Design | Methods, Question Types & Examples. Scribbr. Available at: https://www.scribbr.co.uk/research-methods/questionnaire-design/ [Accessed 28 Mar. 2024].
Crow, G. et al. (2006) ‘Research Ethics and Data Quality: The Implications of Informed Consent’, International Journal of Social Research Methodology, 9(2), pp. 83–95. doi: 10.1080/13645570600595231.
Esposito, E. (2018). 4 types of research methods all designers should know | Inside Design Blog. [online] Invisionapp.com. Available at: https://www.invisionapp.com/inside-design/research-methods-designers/ [Accessed 9 Apr. 2024].
Farnsworth, B. (2021). How to Design a Questionnaire - iMotions. Best Practice. Available at: https://imotions.com/blog/learning/best-practice/design-a-questionnaire/ [Accessed 10 Apr. 2024].
Ghooi, R.B. (2011). The Nuremberg Code-A critique. Perspectives in Clinical Research, [online] 2(2), p.72. doi:https://doi.org/10.4103/2229-3485.80371.
Gossett, S. (2021). What Exactly Is Ethical Design? builtin.com. Available at: https://builtin.com/design-ux/ethical-design [Accessed 12 Apr. 2024].
Hecker, J. and Kalpokas, N. (n.d.). Confidentiality and privacy in research. The Ultimate Guide to Qualitative Research - Part 1: The Basics. Available at: https://atlasti.com/guides/qualitative-research-guide-part-1/confidentiality-privacy-research#:~:text=Why%20is%20confidentiality%20important%20in,to%20unauthorized%20individuals%20or%20entities [Accessed 6 Mar. 2024].
Kaiser, K. (2019). Protecting Respondent Confidentiality in Qualitative Research. Qualitative Health Research, [online] 19(11), pp.1632–1641. doi:https://doi.org/10.1177/1049732309350879.
Lavery, D. and Mendez, J. (2021). Ethical considerations for surveys. ArcGIS Blog. Available at: https://www.esri.com/arcgis-blog/products/survey123/constituent-engagement/ethics/ [Accessed 17 Feb. 2024].
Lee, W., Zankl, W. and Chang, H. (2016). An Ethical Approach to Data Privacy Protection. ISACA, [online] 6, pp.1–2. doi:https://www.isaca.org/resources/isaca-journal/issues/2016/volume-6/an-ethical-approach-to-data-privacy-protection.
Leonard, N. and Ambrose, G. (2018). Conducting research. First Edition ed. [online] www.bloomsburyvisualarts.com, Lausanne: AVA Publishing SA, pp.96–106. Available at: https://www.bloomsburyvisualarts.com/encyclopedia-chapter?docid=b-9781350088771&tocid=b-9781350088771-chapter4&pdfid=9781350088771.ch-004.pdf [Accessed 21 Apr. 2024].
Mora, M. (2019). 10 Things to Consider in Survey Design. Relevant Insights. Available at: https://www.relevantinsights.com/articles/survey-design/ [Accessed 10 Apr. 2024].
Pols, A.J.K. (2016). May Stakeholders be Involved in Design Without Informed Consent? The Case of Hidden Design. Science and Engineering Ethics, 23(3), pp.723–742. doi:https://doi.org/10.1007/s11948-016-9811-0.
Research, A. (2020). How To Design A Good Questionnaire. Anpar Research Ltd. Available at: https://www.anparresearchltd.com/post/how-to-design-a-good-questionnaire [Accessed 12 Apr. 2024].
Shariat, J. and Savard Saucier, C. (2017). Tragic Design [Book]. In: www.oreilly.com. [online] O’Reilly Media, Inc. Available at: https://learning.oreilly.com/library/view/tragic-design/9781491923603/ [Accessed 16 Nov. 2022].
Simkus, J. (2023). Sampling Bias: Types, Examples & How to Avoid It. Simply Psychology. Available at: https://www.simplypsychology.org/sampling-bias-types-examples-how-to-avoid-it.html [Accessed 13 Apr. 2024].
Singh Gil, N. (2022). Visual Perception in User Experience Design | 2022. www.xenonstack.com. Available at: https://www.xenonstack.com/blog/visual-perception-in-design#:~:text=Introduction%20to%20Visual%20Perception%20in%20Design [Accessed 21 Apr. 2024].
Takeda, A. (2022) ‘Uncovering ethical concerns through reflexivity—ethics in practice in fieldwork’, International Journal of Social Research Methodology, 25(3), pp. 419–424. doi: 10.1080/13645579.2021.1889828.
Williams, C. (2007). View of Research Methods. [online] Clutejournals.com. Available at: https://clutejournals.com/index.php/JBER/article/view/2532/2578 [Accessed 4 Mar. 2024].
Zight (2020). How to Ask Unbiased Questions in User Research. Zight. Available at: https://zight.com/blog/ask-unbiased-questions/ [Accessed 7 Mar. 2024].
Zulfiqar, Z. (2004). Beyond informed consent. Bulletin of the World Health Organization, [online] 82(10), pp.772–773. doi:http://www.scielosp.org/pdf/bwho/2004.v82n10/771-777/en.
Comments