How can you balance detail and brevity in a questionnaire?

Clarifying the core purpose and the key insights you seek. Clarity of Purpose: Clearly define the questionnaire's objective. Avoid unnecessary questions that don't contribute to your goal. Essential Questions: Craft questions that are concise, relevant, and to the point. Open-Ended Options: Incorporate open-ended questions to provide detailed responses. Consider the Respondent: Respect your audience's time and attention. Be selective in the details you request, while still extracting valuable information. Pilot Testing: Conduct pilot tests to ensure the questions are clear and that participants can respond easily. Use of Scales: For quantitative data, implement rating scales to maintain brevity while capturing a range of opinions.

The whole purpose of this article "balance detail and brevity in a questionnaire" has its foundation at this stage - Being crystal clear on the core objective of the study Here are the set of questions that have been most helpful for me to get that clarity. Ask the client: 1. What decisions are you going to make based on findings from this study? 2. Who/which departments will be making those decisions? 3. What exactly will influence your decision making? What will trigger it? What are the likely barriers? 4. How critical is that decision for you/your business? Do ask "Why?" few times for each of the above to get deeper Answers to these 4 questions should provide you with the deep understanding of the objectives that you have to deliver on

Every question does not need to be a full sentence. The beginning of the questionnaire should be the paragraph explanation. Each question after that should work just around the keywords and action words.

Achieving a harmonious equilibrium between detail and brevity in a questionnaire entails a meticulous approach. Begin by clearly defining the survey's objectives and the depth of information required. Prioritize essential questions, avoiding redundancy and superfluous queries. Use precise language and concise wording without sacrificing clarity. Employ logical question flow, starting with general inquiries and gradually delving into specifics. Pilot testing can reveal whether the questionnaire strikes the desired balance, enabling refinement for optimal data collection without overwhelming respondents.

Penning down key objectives and Qs under each key obj we want to answer can be a good starting point. Having a hypothesis or a set of hypothesis can also bring clarity regarding what is expected of the survey.

"Do you experience somnolence or excessive daytime sleepiness?" A doctor could say, "Do you feel unusually tired during the day?" Empathy, at the heart of design thinking, is the key to effective data collection. It means truly understanding your audience - their experiences, emotions, and unique perspectives. Using language they're comfortable with creates an inviting space for honest feedback. This approach not only improves data quality but also forges a stronger connection, ensuring insights are relevant and compelling. Add Also : Simplicity in questionnaire design mirrors the clarity of our inner selves.

Traditional MR questionnaires often use verbose and complex language. It's driven by researchers' need to get as much detailed information as possible while not sounding biased in their questioning. However, this approach can backfire, resulting in survey fatigue & lower comprehension. Also, very simply normal people don't talk like that! Here are some tips: 1. Use plain language: Avoid using jargon, technical terms and complex sentences 2. Be specific: No vague or open-ended questions 3. Use active voice: Its more concise and easier to grasp. This is how people normally talk! 4. Never use double negatives: Its confusing and exhausting 5. Pilot: As always, pilot to find avenues of further enhancement before you launch the fieldwork

Always imagine you're asking these questions to someone who has no background/context about your industry and who will see each question for the first time in their life (and will likely see it only once) Be extra clear and leave no space for ambiguity in your questions and answer options but always use the simplest words/phrases you can to do so.

Avoid numerous "grid-like" questions due to potential respondent fatigue. To improve data quality, use a "Multi-Select" question type over a "Yes/No" question. I generally don't recommend more than a few (2 or 3) open ended questions.

Here is a 'cheat-sheet' of the best & worst type of questions in context of 'balancing detail and brevity': BEST - Close ended qns: Likert scale, multiple-choice, yes/no, numerical response options arranged in ascending order - Ranking qns - Semantic differential qns: They use opposing word pairs/bi-polar to measure attitudes & perceptions e.g "good vs. bad" - Gamified association questions: e.g. drag & drop of words/phrases with brands that fit with them NOT GOOD - Open-ended qns: 1 such qn takes time worth ~3 close-ended qns - Grids + rating scale: 15 statements X 8 brands = 120 actions expected from respondents! - Leading or Double-barreled questions: Both are No-No - Qns with more than 40 words in it - Answers with >8 options

The best questionnaires tend to have a good mix of question types. Try to avoid having several of the same question type one after the other as respondent fatigue will likely become an issue. Also, when possible try to rephrase a yes/no question as it will likely improve the quality of data but will also provide you with more insights. Instead of asking "Have you bought shoes in the past 3 months? Yes/No" ask "Which of the following, if any, have you bought in the past 3 months? Shoes, Shirt, Socks, Coat etc." You still use one question but get more insights and you ask it in a less leading way.

"Ideal/optimal # of questions in a MR survey" - This is a topic I have researched deeply in last 3 years. I found that it's not about the number of questions in the survey but rather the amount of time it takes to complete the survey. You could well have a single question - a brand rating grid. Assume 15 attributes X 8 brands = 120 actions which will take upto 6-10 mins just for this single question (and an exhausted respondent too)! Here is the summary of my recommendations: IDEAL: 5 min. Today, we can't even wait for 5 sec to skip the pre-reel ad on our favourite online video platform, why would we endure a long survey? OPTIMAL: 6-8 min PRACTICAL: 10 min UPPER LIMIT: 14-15 min max $ incentives have to go up as we go down the above

One way to cut unnecessary questions is to complete this sentence for every question: “the answers to this question will help me understand _________”. The desired insight for each question should be MECE and directly contribute to your overall research objectives. Repeating the same desired insight across multiple questions is a sign that your questions are redundant and can be trimmed.

It is good to understand the difference between 'need to know' and 'nice to know' questions. Helps avoid temptations of overreaching and keep the questionnaire strictly adhered to the objectives of the study. Always ask 'is this question necessary to answer any of my objective?'

I'll add this age old philosophical gem “Do not do to others what you don’t want to be done to you.” as a reasonable starting point. Also this is based on the assumption that as a current generation we want things quick, snappy and with least amount of effort. If you are the kind that loves answering a survey for as long as is needed of you, clearly my opening comment doesn't apply. You are a precious minority and the majority of the world, unfortunately is not as generous with their time :) This simply means ask the 'absolutely' essential questions and keep to under 5 mins (~ max 15 simple close ended Qs) If you really need to ask more, if it is that important, be ready to field another survey. KISS principle is a winner here

Some really excellent points in the comments here. I wanted to add that if your survey has more than ~25 questions there is a good chance you are trying to ask too many things for one survey. It's often tempting to try to address all the things you need to know in a single project/survey but consider splitting those into 2, or even 3, clear themes and therefore split that one big survey into a couple smaller ones with a greater clear focus, if possible.

Disagree with the notion that "a great way to test and refine your questionnaire is to distribute it internally within your company". Its undoubtably a convenient option but it's not the right approach. Why? People working in a MR/marketing/support firms are what we would call 'professional respondents'. People who write surveys or work with them on a regular basis are not at all normal respondents and hence their input/feedback is not representative/valid. Speak to the real target respondents instead, even if takes a bit of effort. The inputs gathered from them will have much higher value and relevance vs. the easy option

Distributing the survey internally within your company can help with spotting inconsistencies/spelling mistakes and similar errors but I agree with Nishant that they are not a representative sample that can accurately test the clarity of your survey. Most platforms allow you to initially test the questionnaire with just a few respondents (pilot) and then increase the sample as you wish. Having said that, respondents are not trained to spot ambiguity in the way a question is phrased, tell you if the survey flow has leading elements, and whether the Likert scale you used is the right one so there is definitely value in having your survey checked by someone who has knowledge in survey design and best practices, if you have access to one.

A great way to test and refine your questionnaire is to distribute it internally within your company. Get other teams involved. Ask them to take the questionnaire on both smartphones and desktops. Ask them to make notes on what they observe: - Is the questionnaire too long? - Are the questions too vague? - Does the questionnaire work well on mobile? These feedback from different people in your company can help you refine your questionnaire better. A small caveat with internal review is to put a time frame on it to prevent internal feedback from delaying the launch and distribution of your questionnaire.

1. Good design can go a long long way. This includes look and feel aka aesthetics 2. Think device agnostic consistency. Whether it's on a mobile or a laptop, that shouldn't sway a respondent's propensity to respond at all or sway the way they do respond.

A general good rule of thumb is to be an empathetic questionnaire writer, designer, and programmer. This aspect isn’t just important for ensuring happy survey takers; it’s important for ensuring high quality results. Think about the barriers you have faced when taking surveys. Don’t reproduce those barriers for others.