The Complete Guide to Qualitative Market Research

The Complete Guide to Qualitative Market Research

Qualitative research is one of the most prominent research methods in the ever-increasing research sphere. Running counter to quantitative research, qualitative research encompasses a distinct set of differentiating qualities (no pun intended). These attributes prove that these two methods ought not to be used interchangeably.

So what exactly is qualitative research? At a glance, this type of research method seeks to gather in-depth data about a phenomenon without focusing on numerical data or on quantities.

But there is much more to this kind of study method. Learn holistically about qualitative market research with this complete guide.

What Defines & Makes Up Qualitative Research?

Qualitative research is centered around experiences, ideas and opinions. As such, it does not focus on statistical or quantitative outcomes. Instead, it seeks out an in-depth understanding of an issue, occurrence or phenomenon.

Thus, this research method zeroes in on the “what” and more importantly, the “why” of a research subject. (Unlike quantitative research, which focuses on the “how much”).

Here are some of the applications of qualitative research:

  • Understanding an issue in greater depth

  • Finding the reason behind an occurrence (whether it’s desirable or undesirable)

  • Uncovering trends in target market opinions

  • Forming educated solutions to address customer/studied subject concerns

  • Discovering the causes of certain actions

Qualitative research generally relies on a smaller sample size in order to get a deep read of happenings, causes and motivations. This kind of research method functions through the usage of open-ended and exploratory questions.

Understanding the “why” behind an issue is then used to make decisions on how to resolve the issue or how to improve on an existing productive situation.

Qualitative data must occur in natural environments. This denotes a kind of environment in which participants discuss their opinions at length and at ease, which researchers use to gain deeper knowledge and form inferences around a topic.

Prior to the internet, this kind of research was conducted in-person, but with the advent of the internet and innovations in market research, qualitative data has been collected online. The digital space can also serve as a natural environment.

The Five Main Types of Qualitative Research

Just as with quantitative research, there is not a single approach to conducting qualitative research. On the contrary, there are five main varieties of performing qualitative research. Aside from their methodology, these sub-categories also seek different types of answers and conclusions.

1. Narrative Research

This research is used to form a cohesive story, or narrative, by way of consolidating several events from a small group of people. It involves running in-depth interviews and reading up on documents featuring similar actions as a means of theme-searching.

The point of this is to discover how one narrative is shaped by larger contextual influences. Interviews should be conducted for weeks to months and sometimes even for years. The narrative that the researcher uncovers does not have to be presented in sequential order.

Instead, it should be projected as one with defined themes that attempt to reconcile inconsistent stories. This method can highlight the research study’s ongoing challenges and hardships, which can be used to make any improvements.

2. Ethnographic Research

The most common qualitative research method, ethnography relies on entrenching oneself in various participant environments to extract challenges, goals, themes and cultures.

As the name suggests, it involves taking an ethnographic approach to research, meaning that researchers would experience an environment themselves to draw research. Using this firsthand observation, the researcher would not need to then rely on interviews or surveys.

This approach may seem to be far-fetched where market research is concerned, but it is doable. For example, you’d like to see the effectiveness or frustration that customers face when using your product. Since you can’t follow them home, you can request videos that show them using it. Many big brands have call-outs on their websites (ex: on product pages) for their customers to send in videos of their interactions with the products.

3. Phenomenological Research

This qualitative method entails researchers having to probe a phenomenon or event by bringing lived experiences to light and then interpreting them. In order to achieve this, researchers use several methods in combination.

These include conducting surveys, interviews and utilizing secondary research such as available documents and videos on the studied phenomenon. Additionally, as in ethnographic research, phenomenological research involves visiting places to collect research.

These will help you understand how your participants view your subject of examination. In turn, you will gain insight into the participants’ motivations.

In this research type, you would conduct between 5 and 25 surveys or interviews, then peruse them for themes. Once again, you would scrutinize experiences and sentiment over numerical data.

4. Grounded Theory Research

In contrast to phenomenological research, which seeks to fully form the core of an issue, grounded theory attempts to find explanations (the why) behind an issue. To achieve this, researchers use interviews, surveys and secondary research to form a theory around the issue/occurrence.

The sample of this study tends to be on the larger side, at 20-60 participants. Data extracted from this type of research is interpreted to determine the reasoning behind, for example, heavy usage of or frustration with a product. These types of studies help a business innovate an existing product by getting into the weeds of how it’s used.

5. Action Research

This type of research involves researchers and participants working collaboratively to bring theory to practice. Also called participatory research, collaborative inquiry, emancipatory research and action learning, this method entails the act of “learning by doing.”

This means a group of researchers come together to find and address a problem, resolve it and then study the success of their endeavors. If they underperformed or their outcomes don’t satisfy their expectations, they would then reattempt the process.

In action research, a researcher spends a considerable amount of time on collecting, analyzing, and presenting data in an ongoing, periodic process. This involves researchers coming up with their own surveys and interviews around a subject matter, then presenting their findings to one another to draw conclusions and solutions.

They would put into practice the means to improve a situation and continue measuring their success throughout the process.

Examples of Questions for Qualitative Research

When working within the capacity of any of the above research types, it’s crucial to ask the right questions. Here you’ll find the questions you can use when conducting each of the five types of qualitative research.

Bear in mind that some of these questions will appear to be similar in nature; some are even interchangeable. That is normal, as researchers may search for the same answers, but apply a different approach in their research method.

In any case, all of the below features questions that fit within the larger qualitative research framework.

Learn more about asking insightful market research questions. Here are a few examples of the questions within the five categories:

1. How do people who witnessed domestic violence understand its effects in their own relationships?

    Variable: Views of domestic violence on one’s own relationships

    Demographic: People in relationships, who’ve witnessed domestic violence

    Qualitative Research Type: Narrative

2. What are the lived experiences of working-class Americans between the ages of 20 and 40?

    Variable: Experiences and views of a working-class background

    Demographic: Working-class Americans ages 20-40

    Qualitative Research Type: Narrative

3. How do Asian Americans experience reaching out to address mental health concerns?

    Variable: The experiences in seeking out care for mental health

    Demographic: Asian Americans seeking help for mental health

    Qualitative Research Type: Ethnographic

4. What do you enjoy about this product or service?

    Variable: The positive experiences of using a particular product/service

    Demographic: The target market of a product or service

    Qualitative Research Type: Ethnographic

5. How have people who have experienced poverty changed their shopping habits when they entered the middle (or higher) class?

    Variable: The changes or stagnation in shopping habits

    Demographic: those who experienced poverty, but climbed the social ladder

    Qualitative Research Type: Phenomenological

6. What was it like when you had a negative online shopping experience?

    Variable: unpleasant shopping experiences

    Demographic: a group that is most likely to shop at a particular online store

    Qualitative Research Type: Phenomenological

7. What influences managers in private sectors to seek further professional advancement?

    Variable: Motivation for seniority

    Demographic: Managers in the private sector

    Qualitative Research Type: Grounded Theory

8. How do women in third world countries set up financial independence?

    Variable: Efforts at reaching financial independence

    Demographic: Women in third-world countries

    Qualitative Research Type: Grounded Theory

9. What impact does collaborative working have on the UX optimization efforts of a telecommunications company?

    Variable: effects of collaboration on the UX of a telecommunications company

    Demographic: workers in the telecommunications space

    Qualitative Research Type: Action Research

10. What strategies can marketing managers use to improve the reach of millennial customers?

    Variable: Strategies to improve millennial reach and their outcomes

    Demographic: Marketing managers

    Qualitative Research Type: Action Research

When to Use the Research and How to Analyze It

The qualitative research method has specific use cases. You ought to consider which is best for your particular business, which includes your strategy, your marketing and other facets.

The core of qualitative research is to understand a phenomenon (a problem, an inadequacy, and a slew of other occurrences) including its causes, its motivations, its goals and its solutions. Researchers do this by observing smaller portions of a population.

Researchers should use this form of research whenever you need to get the gist of a particular occurrence or event. It is particularly useful for studying how your target market experiences certain situations and how it feels about them.

There are several more specific ways that elucidate why this research style is valuable if not completely necessary. Here are some of the most crucial ways this method of research is vital:

  • Helps brands see the emotional connections customers have with them

  • Allows brands to find gaps in customer experience (CX) and user experience (UX)

  • Enables brands to create experiences that are more tailored to their target market

  • Helps businesses understand how they can improve on their product, service or CX

  • Finds experiences that customers had that highlight sensitive topics/language for them

  • Shows businesses how customers compare them to their competitors

  • Identifies possible solutions and innovations based on customer attitudes and experiences

To analyze qualitative research, you should first identify your subject of study and decide on the type of research you need to conduct based on the five types of research that fall under the qualitative category.

Then, brainstorm several questions that you can use to form the base of your studies. During the process make sure to jot down (either digitally or otherwise) your observations. For example, record interviews and store surveys in an organized database.

Make sure you ask open-ended questions in surveys, interviews, focus groups, et al. Aggregate secondary research such as government database documents, articles in your niche,  images,  videos and more.

Search for patterns or similarities within your findings. When you group them together and organize them by demographics, you can start drawing conclusions and proposing solutions.

The Benefits and Drawbacks of Qualitative Research

Qualitative research can be extraordinarily beneficial. But as with other aspects of research and beyond, it too comes with a set of drawbacks. As a business owner, marketer or market researcher, you should know both the pros and cons. Here are some notable ones:


  1. More intimate understanding of context and causation: besides understanding “what” in a granular way, you also learn the “why” and “how” of a particular situation.

  2. Understanding key experiences: Open-ended questions lead to unique answers, exposing things numerical-based surveys can’t answer.

  3. A foundation of deep insights: The design of the study is made to understand how customers relate to particular occurrences, events, ideas and products.

  4. Context-driven: Finding insights on motivation and past behaviors allows researchers to understand what their target market needs and what it tries to avoid.

  5. No need to find and create the correct measuring units: Open-ended questions don’t require a scale, a number range or any other measuring tools — one less thing to worry about.

  6. Smaller sample size: Smaller sample sizes allow researchers to study responses more thoroughly to form more accurate hypotheses and conclusions.

  7. Inspirational: The responses received can also help researchers form new studies.

  8. Flexible and detail-oriented: Since questions aren’t based on scales and other units, you can ask more creative and in-depth questions. Questions focus on details and subtleties for robust insights.


  1. Relies on researcher experience: It relies on the researchers’ experience; not all are familiar with industry topics.

  2. Not statistically representative: Only collects perspective-based research; does not provide statistical representation. Only comparisons, not measurements can be executed.

  3. Difficult to make copies of data.  Individual perspectives make it hard to replicate findings, making it it more difficult to form conclusions.

  4. More likely to have researcher bias: Both conscious or subconscious of the researcher can affect the data. The conclusions they draw can thus be influenced by their bias. (This can be avoided by using controls in data collection.

The Final Word

Market research is a wide-spanning undertaking. It has a wide swath of aspects, practices and applications. As such, researchers should know its main categories and qualitative research is one such category of significance.

As opposed to quantitative research, which has four methods, qualitative research has five — not all of which will be of use to your particular market research needs. In any case, this type of research involves imbuing as much context and particularities around a phenomenon as possible.

As such, researchers should create questions more specific to the aforementioned examples of this article. That is because those are more encompassing, generalized questions that researchers can attempt to answer after conducting all of their research and parsing of the findings.

But prior to that, researchers should ask several related questions around a particular topic and tailor those questions as best as possible to the target audience.

Frequently asked questions

What is qualitative research?

Qualitative research is a type of research that is conducted to gain deep or unexpected insights rather than focusing on numeral or quantitative data.

Why is qualitative research conducted?

Qualitative research is conducted to find the “why” of the research subject, rather than the “what’ of that subject. For example, qualitative research might be conducted to understand an issue more deeply, to understand why something is happening, or to learn how to address a target market’s concerns.

What is narrative research?

Narrative research is a type of research that is used to create an in-depth story about a phenomenon or event. It is conducted by interviewing a small group of people who were directly involved in the event.

How is ethnographic research conducted?

When conducting ethnographic research, the researchers use firsthand observations of an environment to more deeply understand the goals, challenges, or opinions of the target audience.

What is action research?

Action research is a type of qualitative research in which researchers and participants collaborate to better understand a phenomenon. Together the group works to find and solve the problem by gathering information on an ongoing and evolving basis.

The Complete Guide to Quantitative Market Research

The Complete Guide to Quantitative Market Research

Quantitative research is a chief category in the research sphere, along with qualitative research. An encompassing aspect of market research, it can include both primary and secondary methods. 

Although used interchangeably with qualitative research, quantitative research is a distinct process that should not be confused with its counterpart. In fact, it is the opposite of qualitative research.

Let’s navigate through the waters of quantitative research in this complete guide.

What Defines & Makes Up Quantitative Research?

As its name suggests, quantitative research is the process of aggregating quantitative, or numerical data for research purposes. This data is used for a number of applications. These include:

  • Quantifying opinions, behaviors, attitudes and problems
  • Making generalizations
  • Forming predictions
  • Discovering patterns
  • Determining averages
  • Testing relationships

Quantitative research generally relies on a larger sample size in order to quantify any issue or variable. In order to achieve this, this research method involves using mathematical and statistical means. 

This type of research answers the “what” and the “how much” of a subject within a research endeavor. As it forms generalizations, this type of method involves surveying a larger population, using measurable data and processing all the data first and then analyzing it from a statistical standpoint.

The Four Main Types of Quantitative Research

There are four main ways to perform quantitative research. Aside from their methodology, these sub-categories also seek different types of answers and conclusions.

1. Descriptive Research

This is used to determine the state of variables. It describes the situation and environment surrounding a variable or topic. As such, it is used for arranging comparisons, outlining sample characteristics, overlooking emerging trends and confirming existing phenomena.

The data is collected by way of observation. Descriptive Research is used to form a hypothesis, but only after having aggregated all the necessary data.

2. Correlational Research

This research method is used to examine the relationships between different subjects and variables. Analyzing relationships is necessary to either test a hypothesis or a prediction. Because this research focuses on relationships between fixed variables, other outlying variables are not part of the investigation.

Correlational research is in direct opposition to experimental research, as none of the studied variables are manipulated. Correlations can be either positive or negative, with different degrees of the relationship’s strength.

3. Experimental Research

This method is used for finding whether there is a cause and effect relationship among variables. This kind of research relies on the scientific method. Unlike correlational research, experimental research involves manipulating variables.

Researchers would manipulate a variable to uncover its effect on another one. This method is frequently referred to as true experimentation, as no experimental undertaking leaves all variables unchanged; at least one must be influenced in some way. 

This includes manipulating, randomizing or reverting back a variable. The variables are then measured, calculated and compared.

4. Survey Research

The final research method is crucial to understanding behavior. In market research, it is often used to acclimate a brand with its target market’s desires, needs, points of contention and behaviors.

Surveys allow researchers to ask pointed questions to either discover their target audience or get a granular sense of their opinions. As such, they can be conducted within one group or many, for the sake of comparison.

Instead of turning to survey panels, which are likely to have skewed or biased results, researchers should use a random sample of people. A non-panel-based survey will garner more respondents that aren’t motivated by professional compensation.

Surveys can be administered by mail,  in-person, on the phone, or digitally. The latter has even more options: online surveys, third-party surveys, emails and in-app.

Examples of Questions for Quantitative Research

Survey research has a far larger scope of questions than do the other three types, as researchers can ask practically anything to conduct their studies. However, there are some best practices in survey questionnaires, such as focusing on your industry, your product and the desires of customers.

Learn more about asking insightful market research questions. Here are a few examples of quantitative research questions in the three other categories.

  1. Is working from home the best option to improve productivity for employees with long commutes?Variable: Working from home and in-office
    Demographic: Employees with long commutes
    Quantitative Research Type: Experimental
  2. How has the coronavirus changed employment for white-collar workers?
    Variable: Employment types and statuses
    Demographic: White-collar workers
    Quantitative Research Type: Experimental
  3. How often do working people travel for a holiday?
    Variable: Amount of times respondents travel during a holiday
    Demographic: working people
    Quantitative Research Type: Descriptive
  4. How much would you pay for a subscription to an entertainment magazine?
    Variable: payments for a magazine subscription
    Demographic: women aged 14-44, those interested in celebrities
    Quantitative Research Type: Descriptive
  5. What is the difference in smartphone usage between Millennials and senior citizens?
    Variable: Time spent on using a smartphone
    Demographic: Millennials and seniors
    Quantitative Research Type: Correlational
  6. Does the leadership style of car shop owners predict the job satisfaction of car salespeople?
    Variable: Leadership style and job satisfaction
    Demographic: Car shop employers and salespeople
    Quantitative Research Type: Correlational 

When to Use Quantitative Research and How to Analyze It

The quantitative research method has specific use cases. You ought to consider which is best for your particular business, which includes your strategy, your marketing and other facets.

The core of quantitative research is to quantify a phenomenon (a problem, an inadequacy, and a slew of other occurrences) and understand its prevalence. Researchers do this by observing large portions of a population.

You should use this form of research whenever you need to be presented with the state of things at a higher level, or from a bird’s eye view. This Is because this type of research can identify links between various factors, look for correlations and discover cause and effect relationships.

Researchers can then use the results of their findings to form predictions. This is useful in market research when launching a new product, brainstorming product ideas or innovations or growing a customer base.

To analyze this research, it should first be made quantifiable and objective. Researchers should pin down the scales and units of measurements in their various studies. Then, they should organize them into easily interpretable formats.

For example, once you gather the numerical data, you can enter it into a spreadsheet. Thereafter, you can organize it by desegregating it into graphs, charts and tables. Finally, you should draw data-based conclusions from your study. You can also do further sleuthing via advanced analytics.

The Benefits and Drawbacks of Quantitative Research

Quantitative research has a bevy of benefits; it also has some hindrances. You should peruse both the positive and negative qualities of this research type before setting out on any major research project. The following may help you choose one form of research over the other, or use aspects of both.


  1. Larger sample pools: the larger the group of respondents, the more accurate are the results.
  2. Highly structured: Surveys, questionnaires, and other tools for recording numerical data
  3. Focused: The design of the study is determined before it begins
  4. Theory-based: Research tests a theory to provide support/proof
  5. Designed to Be Analyzed: Numbers/statistics exist as tables, charts, figures and other non-textual forms for easy analysis.
  6. Objective: Steering clear of bias as the research is separated from the data & only objective responses are sought.
  7. Direct comparisons of results: The study can be set in different cultural environments, times or different groups of participants with a statistical comparison of results.


  1. Focuses solely on numbers: This can be limiting as researchers may overlook other data and larger themes.
  2. Superficial Representations: It cannot adequately describe complex concepts (ex: feelings, opinions) it only shows the numbers behind them. 
  3. Several factors can invalidate results: A hypothesis and a model for collecting/ analyzing required; any mistake can lead to bias and inaccurate illustrations.
  4. Erred Structure: If any data is missing or if measurements are not clear, biases easily take precedence.

The Final Word on Quantitative Research

Market research is far too encompassing to fully complete, especially in a limited amount of time. To tackle market research, begin with a research method. Quantitative research is often a good starting point, as it shows you the existence of a problem by way of quantifying it.

Aside from confirming the existence, it can help confirm a hypothesis, find correlations and prove cause and effect relationships. A hard set of data can also help you make educated predictions.

While the three types of quantitative research methods are useful, they do have several disadvantages. The fourth one, ie, survey research helps fill in the gaps and inadequacies of numerical limitations. Interestingly enough, they too can be a source of hard data and numbers. 

Either way, market research is sure to benefit from incorporating surveys as part of the processes.

Frequently asked questions

What is quantitative market research?

Quantitative market research utilizes the techniques of quantitative research in order to better understand the target market. In quantitative research, the information gathered from surveys and questionnaires is converted into numerical values so it can be easily analyzed.

What types of questions do quantitative research answer?

Quantitative research seeks to define “what” and “how much.” It is used for identifying patterns, making predictions, establishing averages, and quantifying opinions, attitudes or behaviors.

What are the four main types of quantitative research?

The four main types of quantitative research are survey research, correlational research, descriptive research, and experimental research.

What type of surveys are used for quantitative research?

Quantitative surveys are best suited for quantitative research. In this type of survey, there are no open-ended questions, and all responses can be assigned a numerical value. In most cases, a quantitative survey is distributed to a large and random sample of individuals.

Why are large sample sizes important when conducting quantitative research?

A small sample size can lead to inaccurate results. The larger the sample size (i.e. the group of individuals who receive the survey), the more likely it is that the results will be statistically significant and accurate.

Your Ultimate Guide To Writing Amazing Survey & Poll Questions


Section 1: Set Your Data Goals

Section 2: Question Writing 101

Section 3: Respondent Experience

Section 4: Avoid Questionnaire Bias

Your Ultimate Guide To Writing Amazing Survey & Poll Questions

Whether it is your first time creating a survey on your own, or you are a seasoned research professional, one thing is for sure: if you can’t write good survey questions, you won’t get helpful consumer insights.

But just what is a “good” survey question anyway? In this guide, we break down the four key elements that you need to master in order to write better survey questions.

A “good” survey question is:

  1. One that achieves your survey and data goals.
  2. One that is clear and direct.
  3. One that carefully considers user experience.
  4. One that is free from bias.

Throughout this guide, we will show you the tactics you can use to make sure you are delivering quality survey questions. To help you along, we will also show you what not to do, so you can easily spot the difference between a good question and one that needs some work.

For example, let’s say we want to avoid bias. Which of these is the better choice?

Reveal Answer

This is an example of a direct, straightforward question that has eliminated bias. Nice work!

Reveal Answer

This is an example of Prestige Bias. By letting the respondent know that an authoritative source says smoking causes cancer, you are biasing the respondent to agree with the source.

Try out these question quizzes at each section to test your knowledge as you go. Let’s get started.

Set (and Achieve) Your Survey Goals

Everyone wants to write good survey questions–they want to be clear, direct, engaging and get respondents excited about answering. 

But these are not the most important reasons to take care when writing survey questions. 

The reason is if you build a survey with the wrong questions or question types, you could get stuck with a bunch of data you can’t use. 

That’s why it is important to set data goals in advance and tailor your survey to achieve those goals from the very beginning. 

How To Set Survey Data Goals

On the surface, this may seem simple: just state what you want to learn from your respondents and go from there. 

But different surveys deliver different types of data. So you need to know how you plan to use the data you get from your questions first. 

For example, let’s say your goal with your survey data is to create a map so you can continuously survey consumers every month and map their responses over time. 

For this to work, you will need to combine demographic data with raw numbers of respondents and their answers.  

You also want to ensure that your questions are simple and not prone to shifts in understanding based on geographic location or time passing. 

Knowing these data goals, which of these two questions would be better for your survey?

Reveal Answer

By adding a specific type of summer location, you assume that your respondents go to the beach or have been recently enough to have a favorite activity there. This will bias your survey towards people that live closer to a beach. Also, by allowing an open-ended question at the end, you will make one answer choice unusable for your map.

Reveal Answer

By keeping responses general, you avoid bias towards one geographic area. And by having a fully quantitative survey but also including a “Something Else” answer choice, you don’t force your respondents to give an answer that doesn’t apply to them, while still ensuring you only get data you can use for your map.

In order to discover the question types that may work best for your survey goals, check out this sample results page. Do an export of the data so you get a sense of what you will have to work with once your survey is completed.

Next, you’ll want to check with your stakeholders to ensure that you are aligned on the questions you want to get answered, how you want to use and ingest survey data and what you are hoping to be able to deliver at the end. 

How To Ensure You Meet Your Goals 

With Pollfish, you have 11 different question types to choose from. All of these question types deliver data slightly differently. 

It’s also critical to deliver these questions in a structural order that makes sense to the respondent and reduces bias from their answers (more on that later). 

Depending on your survey goals, you’ll want to choose these questions wisely to ensure you have a data set you can use. 

The most important difference between survey question types is the difference between Quantitative and Qualitative research. 

From Qualitative vs Quantitative survey questions 

Quantitative research is about collecting information that can be expressed numerically. Researchers often use it to correlate data about specific demographics, such as Gen Z being more likely to focus on their finances than Millennials. Quantitative research is usually conducted through surveys or web analytics, often including large volumes of people to ensure trends are statistically representative. 

Even when the survey audience is very large, quantitative research can be targeted towards a specific audience, usually determined by demographic information such as age, gender, geographic location.

Qualitative research focuses on personalized behavior, such as habits or motivations behind their decisions. This can be gathered through contextual inquiries or interviews to learn more about feelings, attitudes, and habits that are harder to quantify but offer important additional context to support statistical data.

When quantitative and qualitative research are paired, a complete set of data can be gathered about the target audience’s demographics, experience, attitudes, behaviors, wants and needs.

Based on this description, can you tell which is an example of a qualitative question type? 

Reveal Answer

Qualitative surveys don’t have set answer choices. Instead, they allow respondents to speak in their own words about experiences with your product or service.

Reveal Answer

Qualitative survey questions allow respondents to speak in their own words about your product or service without selecting from pre-selected answer choices. You can use these questions to gain a deeper understanding about your consumers and their feelings towards your product outside of what you’d normally ask them. Feel free to ask more broad questions and let them open up.

But differences between quantitative and qualitative are not the only differences between question types. You can still gather a broader range of emotions from respondents without going to a fully qualitative survey. 

Question types like matrix questions, where respondents can rank the importance of different product or service features can provide more quantifiable data while still giving a broader range of interest than a simple one-by-one ranking. 

Lastly, some researchers choose to combine qualitative and quantitative research. By allowing an open-ended “other” field to single selection questions, for example, researchers collect quantifiable data, while still allowing the option for further context. 

Based on your research goals, this added field can either provide needed context for deeper understanding of your audience, or create unnecessary noise that can’t be quantified and throw off the data set. 

It all depends on you. So think it all the way through and have a plan in place before you begin creating your questionnaire. 

Question Writing 101: Clear, Direct, Well-Tested

When writing anything, clear, direct communication should be the ultimate goal. You want to ensure that, regardless of any flowery language or interactive content features, the core of what you are trying to say is communicated loud and clear. 

With survey questions, this becomes even more important.

Once you have selected your survey goals and you have a good idea of what kind of data you want to collect, the goal of your survey questionnaire is to get respondents to completely fill out your survey as accurately as possible. 

Market research best practices show that it is best to let respondents decide if they want to finish your survey or not. With Pollfish, respondents can escape the survey at any time. Therefore, all responses are voluntary and of higher quality than if respondents were forced to finish the survey. 

What that means for you as the researcher is you have to ensure that your questions don’t drive respondents away. 

Don’t worry, Pollfish is here to help you with that. First of all, each questionnaire is reviewed by our research experts prior to launch. But you can help by following these three steps to survey question writing: be clear, be direct, and test on real people. 

Be Clear

In order to get the types of answers you expect, you must ensure that your target audience will fully understand what you want from them. 

What does that mean? Below are a few common mistakes researchers make when writing their first questionnaire. 

Too Much Jargon

When writing for a niche audience, industry jargon can be a welcome addition. Things like common abbreviations and loaded terms help build rapport and allow you to establish a personal connection with your readers. 

Not so in market research. 

Imagine you are reading an article and you come to some jargon you don’t recognize. All it takes is a quick Google search and you have learned some new terminology.

But if you come to a survey question that contains an acronym or abbreviation you don’t recognize, not only will you feel like you can’t answer that question, you will feel like maybe you aren’t qualified to answer the rest of the survey. You may attempt to guess at the meaning, or escape the survey all together. 

Even if your audience is screened to only be in your industry, don’t assume everyone uses the same terminology you do. Take care when using any kind of jargon, acronym or abbreviation in your survey questions. 

Extra Explanation

Most survey question types are fairly self explanatory. If there are 5 answer choices and the instruction is “Select One”, there isn’t much room for interpretation there. 

But when questions get more complicated, extra explanation is essential. 

For complicated concepts or ideas that respondents may only have some ideas about, a “description” question type can be helpful in clarifying which part of the idea a respondent should focus on.

Matrix questions are one question type where you have to be careful of your labeling and explanation. Let’s take a look at two Matrix questions and see which is the most clear. 

Reveal Answer

How are your respondents supposed to know if 1 is good or bad? When you use matrix questions, ratings scales and any other rating or ranking question, be sure to label questions clearly.

Reveal Answer

As you can see, everything here is labeled so it is very clear for each amenity what you are choosing when you select each answer. When you use matrix questions, labels are very important. From the row titles to the answer choices, make sure it is very clear what respondents are selecting.

Questions that contain media like images, audio files and videos also often require further explanation. Let’s take a look at the video questions below and see which is the most clear.

Reveal Answer

With questions containing required media, it is always a good idea to give explicit instructions as to what to do with that media. Once you have done that, you want to be very specific with what you ask respondents to deliver. If your ask is too vague, you are likely to get a lot of short, unthoughtful responses.

Reveal Answer

When you use Pollfish for video questions, respondents are required to watch the entire video before answering. Still, you want to be clear about what you want. If you want them to watch the entire video, say so. Also, be specific with what you want them to pay attention to. Otherwise, you are likely to get a lot of vague, unthoughtful responses.

If it is necessary to consume all media to understand the question being asked, make sure your instructions clearly state this. You can even include your media as a separate question. 

Create non-leading questions

It can be tempting to write questions in a way that seems obvious to you, but survey questions should not follow the same patterns we have when speaking to each other. 

Although plain language is advised, inserting your own opinions—even subtly— into the survey question can lead respondents to feeling more inclined to agree with the answer you want. 

Remaining neutral towards answer choices improves data quality.

The second question remains neutral towards either show. Since the goal of your research is to uncover the most accurate data, you want to avoid pushing respondents towards one answer over another.

Make Answer Choices Clear & Distinct

Once you have made sure your questions and instructions are clear and direct, make sure your answer choices are distinct from one another. 

One common reason respondents bounce from a survey is they cannot decide between two similar-seeming answer choices. 

This can also reduce data quality because it may result in respondents offering answers they don’t really mean. 

Although “yes” and “no” can seem like two distinct answer choices, these also don’t offer gradients within the answer options to appeal to respondents who don’t feel strongly or are unsure of an answer. 

We recommend avoiding “yes/no” questions in favor of a Likert Scale or by writing answers out entirely to ensure less ambiguity and reduce bias. 

Which of these questions seems correct?

Reveal Answer

This question not only assumes the respondent drinks beer, but drinks beer enough to know the difference between these varieties. As a researcher, you don’t even know at this point if your respondent drinks at all. Answer choices need to be clearly distinct to all potential respondents, unless you have narrowly filtered your audience to ensure their knowledge.

Reveal Answer

While this question is not as specific as it’s counterpart, it allows all possible respondents a chance to answer, while even making concessions for those who don’t drink at all. Once respondents answer here, you can get more specific (while ensuring the ability of respondents to participate) by using skip logic. Have the respondent answer and, for the ones who select beer, send them on to a question about which variety they prefer. That way, you can be more assured they will know the difference.

Be Direct

Good survey questions contain two parts: the question and the instructions. 

For many Pollfish survey question types, some level of instruction is built in. But you want to be sure that when you deliver instructions, you are speaking directly, in clear, declarative statements. 

Using too much passive language and indirect phrasing can create unnecessary confusion. Let’s take a look at the two examples below. Select which you think is the most direct. 

Reveal Answer

While these two questions may seem to say the same thing, using softer, more imprecise language opens up the potential for confusion. Using very direct, plain-spoken language removes any such ambiguity. Or to put it more simply, it makes questions (nearly) idiot-proof.

Reveal Answer

While these two questions may seem identical, the indirect language here leaves room for confusion. For example, who am I choosing for? Without a direct subject, I am left to wonder why I might be inclined to choose one flavor over another. Maybe when I buy ice cream, I only buy it for my husband because I am on a diet. If you want your respondent’s personal preference, ask for it, as clearly and directly as possible.

Always keep your Survey Goals in mind

As we learned in part 1, survey goals are at the core of all research projects. To make sure you’re writing the right survey questions to meet the right goals, consider the flow and structure of the survey and how the respondent will be presented with each question. Your goals may also dictate the type of question respondents should be given. 

Example: Survey goal is brand awareness

It’s best to begin a brand awareness survey with an open-ended question for respondents to recall brands that are top-of-mind, prior to being presented with a series of questions that might trigger their recall. 

If the survey goal is press for a specific cereal brand, ranking questions can help uncover consumer preferences.

Ranking questions ask respondents to order answer choices, forcing them to choose one over another. These are best for establishing stronger opinions or validating a strong preference.

The data from a ranking question could be “85% of consumers believe that Trix is the best children’s cereal”, providing a good headline for a research story.

Matrix questions allow respondents to apply a ranking scale towards each cereal, but don’t force them to choose one over another or compare them. These are a better question choice to measure other sentiments.

The possible data to come from a matrix question could be that nearly ¾ of 90’s kids recall Frosted Flakes’ “Tony the Tiger”, however, the Cheerios bee has largely been forgotten.

Test On Real People

While it is important to align with all stakeholders internally on survey goals, you also want to test out your final questionnaire on some real people. 

This can mean either sharing it with co-workers who are not familiar with the research project, or emailing your questions to a friend. 

Getting outside help can be essential in identifying unclear directions that cause confusion for your respondents. 

Respondent Experience: Think Like An App User

Because nearly everyone has an internet-enabled device with them at all times, survey companies have identified the opportunity to use this technology to reach respondents.

Unfortunately, not all surveys are created equal when it comes to offering a good respondent experience on mobile devices. 

Poor respondent experience can have disastrous consequences, impacting your survey response and completion rates, and even impacting the data quality of your study. 

So how can you follow best practices for mobile surveys? 

Build on a Mobile-First platform

By using Pollfish, you are already off to a great start. That is because every Pollfish survey question type is optimized for mobile out of the box. 

Our surveys are not only optimized to fit your mobile screen, they are delivered exclusively inside mobile apps, meaning each question type is designed by app developers to mirror the experience of an app. 

Pollfish also provides a mobile previewer for each question in your survey, so you can test how your question will look on a phone in real time, right inside the questionnaire builder. 

<Maybe add a screenshot of the questionnaire builder here?>

Shorten your questionnaire

Because so many users prefer to take surveys on mobile devices, it is important to think like a mobile user. 

Think about your app usage–do you stay in one place on your phone for a long time, or do you jump quickly from place to place? Do you often have your phone open for hours or do you use your phone in quick bursts? 

Mobile phone users want to move fast. They could be on the go. They could only have a few minutes to kill while waiting in line at the bank. 

That’s why it is important, when you can, to keep surveys short. 

Shorter surveys get faster results and ensure full engagement with your questions from start to finish. 

Our survey experts review every survey before it is published as well, so they may recommend you shorten parts of your study if they feel respondent experience may be impacted. 

Keep a mobile mindset

While all our survey questions will look great on mobile, there are a few things you can do to make your questionnaire more engaging to the mobile user. 

Start by favoring question types that are easier to answer on mobile phones. Multiple-choice questions that users can tap, rather than open-ended questions where typing is required makes answering easier for users. Question types like rating stars and using video can keep users engaged and motivated throughout your survey. 

Keep in mind that while many survey tools that offer a “mobile survey” experience also offer matrix questions, you’ll want to verify that these have been designed for mobile distribution. 

Matrix questions in a typical survey present as a table where respondents are asked to rank a series of answers. In mobile, these don’t translate into a good experience for respondents. 

Ensure that your mobile matrix question utilizes vertical scrolling and doesn’t require respondents to zoom in or drag the table horizontally to see all the answer possibilities.

While more qualitative questions may be helpful to provide more context and emotion to survey respondents, mobile respondents would rather tap or swipe than type. 

If you were a mobile user, which of these would you prefer to answer?

Prevent Survey Bias

One of the most important parts of building the perfect survey is reducing survey bias. 

From Your Ultimate Guide To Survey Bias

Put simply, survey bias impacts data quality. By eradicating survey bias (as best you can), you build trust in the data you’ve collected. 

That’s no small thing, as companies across the world are using consumer insights data to inform everything from content creation to product timelines to strategic planning.

So how do you remove bias from your questionnaire? 

Well, for starters, it is important to remember that preventing all survey bias is impossible. 

That’s right. We are all biased. We have unconscious thoughts that cause us to subtly indicate our true feelings to those around us. And that applys to writing survey questions as well. 

The key is to do everything you can to remove the most pernicious forms of bias from our surveys so we can ensure top-quality data. 

With this section, I’ll take you through a few of the more common ways to avoid bias in your questionnaire. 

Remove Double-barreled questions

Double-barrelled questions happen when researchers blend two questions into one, and then allow for only one answer. This creates a fallacy for the respondent and often results in incomplete or biased answers. 

Can you tell which question is biased here? 

Reveal Answer

While this question may not appear to cover as much ground as it’s counterpart, the other question here doesn’t allow the respondent to choose different answers for the Pool and Bar. It is always better to split double-barreled questions into two separate questions to ensure specificity.

Reveal Answer

This question doesn’t allow respondents the ability to select different ratings for the pool and bar. Even if you consider the pool and bar to be the same, a respondent may not. Double-barreled questions provide incomplete answers. You’ll risk losing critical consumer insights. Make sure you split them in two.

Avoid Assumptions

Depending on how you write a question, you could be assuming things of the respondent, or even forcing them to give an answer they don’t believe. 

One way researchers fall into this trap is by asking loaded questions. Loaded questions make multiple assumptions about a person and then force them to answer a question based on those assumptions. 

Can you spot the Loaded question? 

Reveal Answer

This question presupposes the respondent likes attending baseball games, because they have to fit the premise of the question in order to answer. Before you can ask this question, you must ensure your respondents like attending baseball games. You can do this by asking such a question and then adding skip logic so that only those who have previously attended baseball games will be shown this question.

Reveal Answer

Before a respondent can answer a question about snack preferences at a ballpark, you have to first establish that your respondents attend or have attended baseball games. You can ask the above question and apply skip logic so only those who fit the premise of your intended question are shown it.

Don’t Use Biased Language

Similar to Loaded questions, biased language (sometimes called Leading questions) pushes respondents towards one response or the other. 

This removes the impartial nature of a survey and influences respondents, leading to poor data quality. 

Can you spot the Leading question?

Reveal Answer

The words “new” and “old” bias the respondent towards saying the newer design is easier to use. You need to ask questions from a neutral position and do your best to avoid biased language that lead respondents towards one answer or another.

Reveal Answer

By asking, simply and directly, which design is easier to use, without letting the respondent know which is newer or older, you ensure that you remove bias. By adding responses like Not Sure or No Difference, you open up the question to additional answer possibilities for a broader range of respondents.

Do Your Best To Include All Available Choices

As a general rule, you want to avoid forcing respondents to give an answer they don’t really mean. 

So when you are writing a closed-ended survey question with a finite answer pool, make sure you think hard about additional answer possibilities that may not come immediately to mind. 

Once you have done that, there are two additional ways you can reduce bias on these kinds of questions. 

The first is to add an open-ended “Other” field, where respondents can include any additional choice you may have forgotten. If there is enough consistency in responses, you may even be able to manually quantify the data. If not, you have still provided some much-needed context to the current answer choices. 

If you want to stick with your closed-ended question type, you can make it clear in your instructions that respondents are to select the answer closest to what they think. This is imperfect and provides a tough-to-quantify variable to your data. But it also will keep respondent experience and completion rates up. 

Market Research Terms Glossary


Section 1: Market Research Terminology

Section 2: Question & Survey Types

Section 3: Survey Bias

Section 4: Pollfish Features

Market Research Terms Glossary

You can’t become a market research expert without learning the lingo. That’s why we created the Pollfish Market Research Terms Glossary.

This alphabetical listing of key market research terms, concepts and ideas has been segmented into four sections: Market Research Terminology, Survey Design & Question Types, Survey Bias and Pollfish Features.

For basic research concepts, check out the Market Research Terminology section. Survey Design & Question Types gives simple definitions of the different types of surveys and questions you can build with Pollfish. The Survey Bias section takes you through the ways you can bias your results. And the Pollfish Features section gives you definitions of Pollfish-specific tools and features so you’ll always know what’s going on.

Check it out!

Section 1: Market Research Terminology

The more you read about market research, the more surveys you build and the deeper you sink into the world of consumer insights, the more you translate the language surrounding it.

Here, we define some basic terms that you will need to have an understanding of to become a market research expert.

A   |   C   |   D   |   E   |   F   |   I   |   L   |   M   |   N   |   O   |   P   |   Q   |   R   |   S   |   T


Agile market research

An approach that values numerous small experiments over a few large bets, rapid iterations over big-bang campaigns and responding to change over following a plan.



A fully completed survey that has been screened for bias, insufficient responses and bots.

Completion Rate

The rate at which surveys are completed as compared to the number of surveys started by respondents. To calculate completion rate, divide the number of completes by the number of starts.

Consumer Insights

Valuable information on the preferences, opinions, habits and emotions of your most valuable customers. Consumer insights usually encompass insights related to a product or service.

Control Group

Survey participants can be split into two groups–an experimental group, exposed to a product or service, and a control group that is neutral. A common example is ad effectiveness testing, where researchers can track a respondent’s exposure to an ad through cookies. They then split those who have viewed the ad and those who have not into separate groups, asking the same questions to see how responses differ.

Cost Per Complete

The price you pay per completed survey. This calculation is based on certain factors of your survey, including the number of screening questions, quotas, demographic/geographic filtering and more. Learn More >

Cross Tabulation (Crosstab)

A feature of your survey platform that presents data in a table with rows and columns, designed to help researchers observe two or more variables at the same time. Crosstabs are useful when you want to divide your respondents into subgroups to see how a dependent variable changes the results.



Any information collected by your survey, along with any outside information collected, observed, generated or created in service of your research goals.

Data Cleaning

Removing unqualified, biased or incomplete responses from a survey. This process improves data quality and protects against survey bias. Learn More >

Device ID

An individually assigned ID given to a Respondent’s device to differentiate them from other respondents. Many survey platforms who deliver surveys via mobile devices will collect or create Device IDs. At Pollfish, these IDs are passed to us from the mobile app publishers where we deliver your survey. This ID, along with demographic information respondents give on their first survey, are collected and used for targeting and filtering purposes in the future. Learn More >

DIY market research

Market research conducted using a self-service platform, as opposed to partnering with a market research agency or research consultant. Most DIY market research is conducted in-house to avoid the speed and cost limitations of working with outside entities. Learn More >


When a respondent begins a survey and doesn’t complete it. These are also called Starts. Drop-offs are not counted as completes and, therefore, you will not be charged for them.



Engagement has many meanings. It can refer to a high degree of focus and interest in stimuli. In digital marketing and technology, engagement often refers to metrics surrounding use of platforms, content features or apps (think clicks, time on page, etc). In market research, engagement refers to how users interact with your survey. Does their time spent on each question indicate they are confused or don’t understand how to choose an answer? If so, your completion rate could be impacted.


Feasibility Study

A feasibility study is designed to determine the likely success of a project, product or service. There are many factors that go into a feasibility study, including existing competitors, production limitations, timing, estimated pricing and more. Brands or researchers may conduct feasibility studies to determine the market interest in a new product or service, or even to help determine the feasibility of a future research project.


Fielding refers to the distribution of the survey questionnaire. When using Pollfish, you can watch your results roll in in real time using our Results Dashboard.


Implicit data

Implicit data refers to information that is not provided from respondents directly, but is gathered from available data. For example, Pollfish collects location data and information on a respondent’s mobile carrier from the App publisher, not from the respondent directly.

Incidence Rate

Incidence rate is the measure for the rate of occurrence or the percentage of persons eligible to participate in a survey, based on the targeting criteria selected.


Longitudinal Research

Researchers performing a longitudinal study will run the same survey many times over short or long periods, in an effort to observe how the opinions, behaviors or habits of the same population change over time. The population can also be randomized to see how time impacts the questions being asked, regardless of population.


Margin of error

Margin of error, also called the confidence interval, is a statistical measurement of difference between survey results and the population value, expressed as a percentage. Within the survey ecosystem, the margin of error measures the difference between your survey results and how accurately they reflect the views of the overall population. Learn More >

Market research

Market research refers to the gathering of consumers’ needs, preferences, habits, behaviors and more in an attempt to better understand a company’s potential customers, brand positioning and potential interest in a product or service.

Mobile ID

Also known as the Advertising ID, this unique ID number is how mobile advertisers are able to cookie users and keep track of engagement with mobile ads. At Pollfish, we use this id in a similar way, delivering surveys to users in much the same way an app delivers ads. We use the advertising ID to ensure we don’t send the same survey to the same user more than once. Learn More >


Non-Probability Sampling

Non-probability sampling excludes some of the population in your sample, and that exact number can not be calculated – meaning there are limits on how much you can determine about the population from the sample. These methods include convenience sampling, quota sampling, judgement sampling and snowball sampling. Learn More >


Online panel

Online panels collect responses either via a fully opt-in structure, including a signup page, or start with some form of digital outreach to potential respondents who have agreed to take surveys in advance. Panelists are then recruited to participate in specific surveys, for example via email invitation to the page of the panel provider. Pollfish avoids the pitfalls of traditional online panels by asking users to take surveys while they are using apps or games in real-time, increasing data quality and reducing biased responses. Learn More >



A panel is a collection of potential respondents who have agreed to take a survey in advance of the survey’s fielding process. These respondents are typically promised some type of incentive in exchange for joining the panel, which would effectively pay them for their time. Learn More >


Piping allows researchers to personalize surveys by ‘piping’ an answer from a previous question into a later question. For example, you can ask a respondent their name or occupation on the first question, and then add that name or occupation to future questions to make the questions more personalized.


The population is the total group of respondents who you attempted to survey. If they complete your survey, they become part of your sample.

Primary Data

Primary data refers to the data collected by researchers directly from respondents using surveys, interviews or direct observation.

Primary research

Primary research refers to the methodology of using only data collected directly from respondents, rather than relying on data collected during previous research or from some external source (government agencies, employment records, etc).

Probability Sampling

Probability sampling refers to a randomized method of respondent selection. In order to utilize probability sampling, researchers have to have a method that ensures every member of the population has an equal chance of being chosen to participate (like picking names out of a hat).


Unlike demographics, which explain who your respondents are, psychographics seek to explain why they do what they do. While any quantitative study, group of screening questions or even secondary location data can net you demographic data, psychographics are more often culled from qualitative studies. For example, is your respondent concerned with health and appearance? Do they enjoy socializing or are they more introverted? You can get answers to these questions from scale-based quantitative questions, but open-ended questioning or interviewing often provides more depth to these groupings. Depending on how you plan to use them, you should consider this before creating your survey. Learn More >

Public Opinion Research

Public opinion refers to the opinions of a majority of people in a certain population. Polling the public opinion requires taking as broad of a study as possible and asking direct, quantifiable questions about specific issues.


Qualification rate

The qualification rate is the estimated percentage of people you expect to qualify for your survey based on your targeting criteria, screening questions and other filters.

Qualitative research

Qualitative survey questions aim to gather data that is not easily quantified such as attitudes, habits, and challenges. They are often used in an interview-style setting to observe behavioral cues that may help direct the questions. Learn More >

Quantitative research

Quantitative research is about collecting information that can be expressed numerically. Quantitative research is usually conducted through surveys or web analytics, often including large volumes of people to ensure trends are statistically representative. Learn More >


Your questionnaire is the list of questions you plan to ask your respondents. There are many different types of survey questions you can ask, depending on your survey goals. Learn More >


Quotas are limits you can set for the number of responses your survey collects from a particular group. They can be set across the entire survey or on a given question or segment. Unlike weighting, which ties your quotas to existing data sets like a national or local census, quotas can be chosen by the researcher to match the goals of your survey. If, for example, you’d like to survey a population that is 75% female, you can do that.



A respondent is a person who meets your targeting criteria and completes your survey in full.

Response Rate

The response rate is the percentage of the total targeted population who responded to your survey. Learn More >



Your sample refers to the respondents who matched your targeting criteria and completed your survey.

Sample size

Your sample size is the number of completes your survey receives. Learn More >

Secondary data

Secondary data refers to data that has been collected outside of the bounds of a researcher’s survey, but which the researcher can use to add context. For example, DMV records, national census records and other government information can be used for things like weighting and quotas, or just to show the disparity between perception and reality within a studied population.

Secondary research

Secondary research refers to the summary or synthesis of existing research towards a new research goal. In this practice, previous primary research projects are used as sources.


Segmentation studies seek to separate larger audiences into smaller segments based on similar tastes, interests, perceptions and other secondary factors like education, employment or lifestyle.

Statistical significance

A measurement to quantify whether a result is just due to chance or due to some significant factor. Getting a survey that is not statistically significant comes from sampling error. That’s when your sample doesn’t accurately reflect the population of your survey and, therefore, may cause skewed results. There are two things to contend with when trying to remove sampling error: sample size and variation. The larger the sample size, the less chance there is in the result so that reduces sampling error and increases significance. Controlling for variability of your sample also can impact sampling error–the more variability in your sample, the more prone to error the study will be.


In its most basic form, a survey refers to the questionnaire–delivered either in person or online–that a researcher administers in service of a research study. When using Pollfish and other online survey platforms, a survey refers to the targeting you select, the questionnaire you create and the results files created once you run your survey. Each survey’s full info is grouped together in your account dashboard.



Targeting refers to the criteria you select to screen potential survey respondents. Once targeting is selected, a population is created and your survey is delivered.

Tracking Study (Tracker)

Tracking studies use the same questionnaire, delivered over time, to track brand awareness, monitor customer satisfaction, study consumer interest in new products or services, analyze the effectiveness of advertising creative and more. Tracking studies may be delivered to the same populations (to gauge how perceptions of the same group changes as time goes on) or different populations to view time as just one factor impacting shifting perceptions.

Section 2: Survey Design & Question Types

When creating your survey design, it is important to start with the type of survey you want to run. Understanding the small but significant differences between survey types is a great first step.

Here, we want to review the different types of studies you can run with Pollfish, as well as the ideal question types for each one.

A   |   B   |   C   |   D   |   I   |   L   |   M   |   N   |   O   |   P   |   Q   |   R   |   S   |   U   |   V


Ad Absorption

A study designed to show the effectiveness of certain advertising creative. These studies tend to use narrow audience targeting screening questions and even control groups to determine if new ad creative is having statistically significant impact on user experience or understanding of a product, brand or service. Learn More >

Attitude & Usage Testing

A study that aims to understand the available market for a product or service. This typically involves things like market sizing, general understanding of the product category, questions understanding brand decisions, gathering targeting information and more.

Audience Profiling

Profiling is a process of collecting demographic information to define an audience or population. Researchers may need to conduct surveys of a population to determine if there are enough available survey takers for the demographic or other targeting they want to run. At Pollfish, we use a rolling profiling model to keep our audience up to date. We also collect most of the demographic data you will need up front, including demographic, gender and mobile usage data.Learn More >

Audio question

A question that requires the respondent to listen to an audio clip in full before answering.


Brand Awareness

Brand awareness is the extent to which consumers are familiar with a brand, the extent to which the brand’s intended perception matches reality and the extent to which a company’s brand is helping or hurting sales.


Closed-ended question

Closed ended questions are those that offer a limited selection of answers to choose from, such a single or multiple-selection question, matrix, or scaling question type. Learn More >

Competitive Analysis

Competitive analysis testing helps you better understand how your potential customers are discovering your competitor’s product, feature preferences, user behavior and more. Learn More >

Concept Testing

The process of testing a big idea, concept testing allows you to run ideas past a sample of your target market before producing them. These ideas can be new logos, new product ideas, new ad campaign ideas and more.

Content Creation Surveys

Conducting surveys can reveal gaps in news coverage that can help digital media brands develop a unique editorial voice. By surveying outside of existing readers to prospective readers and even competitive readers, digital media companies can gauge interest in, and existing knowledge of–certain topic areas. Learn More >

Creative Testing

A test allowing a creative team to gauge the effectiveness of their work at conveying the feelings and emotions they were aiming for. This includes logo testing, ad creative testing and more. Creative testing may ask questions about brand effectiveness, gauge tone and consumer enjoyment and gauge the likelihood of someone to purchase before and after viewing creative assets.


Description question

A question in your questionnaire that contains a basic description. This question type is usually used to provide directions to the respondent. Learn More >


Image question

A survey question requiring respondents to look at and respond to an image or a group of images.


Likert Scale

A Likert Scale is typically a 5 or 7 point scale that asks a respondent to express how much they agree or disagree with a statement. Learn More >

Logo Testing

A type of creative testing focused on changes, updates or just current opinions of a brand or product logo. This type of survey may contain questions gauging how appealing, authoritative or on-brand a logo is, or even reveal possible changes to a brand logo and ask potential customers to weigh in. Learn More >


Market Analysis

Studies that focus on the size, scope and potential of a market. Before launching a new product, companies may want to figure out the size of the potential market by broadly surveying interest for that product across key areas and demographic groups. This helps companies determine how large a product rollout may be needed.

Matrix multiple selection question

A matrix question asks respondents to make selections for multiple options on a scale. A matrix multiple selection question lets you select multiple responses for a single option. For example, a matrix may contain elements of a hotel, asking respondents to rate elements like the pool or the lobby bar using a variety of potential responses like “Clean” or “Fun.” A multiple selection matrix will allow you to check all descriptions that apply. Learn More >

Matrix single selection question

A matrix single selection question will ask you to select one answer per option on a scale. For example, a matrix may contain elements of a hotel, asking respondents to rate elements like the pool or the lobby bar using a variety of potential responses. Because responses are single-selection, researchers will often ask something more definitive, like “How likely were you to use these different elements of our hotel?” And ask for a scale of 1-5. Learn More >

Multiple selection question

Multiple selection questions contain a list of options and ask respondents to select all that apply. For example, the question “what kind of music do you like?” will contain a list of music types so respondents can select all types they prefer. Learn More >


Naming Tests

Studies concerning the names of things–new product names, new website URLs, new brand names or even the name of a new film or TV show–naming tests help reveal potential perception issues that can occur. For example, does one title effectively convey what the product or service is? Does the name defy expectations and surprise the user?

Net Promoter Score (NPS)

Net Promoter Score is a widely used customer/consumer experience test that asks a simple question: on a scale of 1-10, how likely are you to recommend this business, brand, product or service to a friend or colleague? Respondents who rate 0-6 are detractors, 7-8 are considered passive and 9-10 are considered promoters. Learn More >

NPS Survey

An NPS Survey asks the Net Promoter Score question to a specified target population. This allows companies to gauge how well they are doing on the NPS scale with their specific target customer or potential customer. Learn More >

Numeric open-ended question

An open-ended question that requires a numeric answer. For example, researchers may ask how much money you’d potentially pay for a product or service. Learn More >


Open-ended question

Open-ended questions do not require the respondent to select from a specific list of responses, but instead asks them to type their response into a text box. Open-ended questions, therefore, are less definitive, seeking to gather more qualitative responses, getting at the feelings and broader opinions from the respondent in their own words. Learn More >


Package Testing

Studies that provide consumer feedback on product packaging. Would you be more or less likely to buy a product with new packaging? Does the packaging effectively convey what is inside? How do your potential customers feel when they look at your new packaging?

Pre/Post Studies

A pre-post study examines whether participants in an intervention improve or become worse off during the course of the intervention, and then attributes any such improvement or deterioration to the intervention.

Product / Market Fit

Product/Market fit is a the degree to which a particular product meets the demands of the market it is in. Have you created the minimum viable product for your market, that solves a problem or need that exists? Product/Market fit studies aim to survey early adopters or potential customers to see if the product meets previously identified criteria for satisfaction within a market.

Product Testing

A final test of a product, product testing gives users access to a prototype and seeks to identify opinions–positive or negative–of a product before it goes to market.


Qualitative survey question

Qualitative research focuses on personalized behavior, such as habits or motivations behind their decisions. This can be gathered through contextual inquiries or interviews to learn more about feelings, attitudes, and habits that are harder to quantify but offer important additional context to support statistical data. Learn More >

Quantitative survey question

Quantitative survey questions are those that can be expressed numerically, meaning they require respondents to select from a pre-selected list of potential responses. Learn More >


Ranking question

Similar to a single-selection matrix, ranking questions ask users to rank different elements of a product or service on a numerical scale. Learn More >

Rating stars

A question where respondents answer using a scale of a number of stars. Learn More >


Screening question

Screening questions help researchers ensure they are getting the exact sample they want using filtering that falls outside of available targeting criteria. For example, if you only want to include people in your survey who have seen the movie Pulp Fiction, your screening question would ask potential respondents if they have seen the film, and screen them based on their answers. Learn More >

Single selection question

A question providing a list of potential responses, from which respondents may choose one answer. Learn More >

Slider question

A question where respondents answer by dragging a button across a slider to provide a rating. Learn More >


Usability Testing

Usability testing aims to test user experience changes on real users to determine if the experience has been positively or negatively impacted by any changes. Unlike more traditional product testing, usability testing usually focuses more on design changes and more commonly relates to app or software development.

UX Testing

Any testing related to the overall user experience of a product (usually app or software). UX testing tends to be more holistic than Usability testing, running qualitative and quantitative studies of each phase of the user journey.


Video question

A question where respondents much watch a video in its entirety before answering.

Section 3: Survey Bias

One of the most important factors in DIY survey creation is avoiding bias. Survey bias reduces data quality, threatening the integrity of your consumer insights.

We have gone in-depth on ways to root out survey bias here. But to fully understand the different types of bias you may have to contend with, here are some helpful definitions.

A   |   D   |   E   |   L   |   O   |   P   |   S


Acquiescence Bias

Also referred to as “yea-saying,” Acquiescence bias occurs when respondents are overly agreeable, to the point of contradicting themselves. You can avoid this by limiting the number of questions asking respondents to simply agree or disagree, as these statements often result in over-agreement. Learn More >


Alternators are survey takers that fill in responses in familiar patterns, simply alternating responses but not engaging with the questions. You can avoid this by mixing in some open-ended, audio or video questions. If responses appear to be biased, Pollfish will remove them from your sample. Learn More >


Double-Barreled Questions

Double-barrelled questions happen when researchers blend two questions into one, and then allow for only one answer. The respondent must agree in total with a statement they may only partially believe, creating incomplete or biased responses. For example, “Did you find the product interesting and helpful? Yes or no?” Learn More >


Extreme Responding

If given the option, some respondents will select only the most extreme option. Make sure to mix up your question types. If you have several scales or rating questions, make sure to mix in some others that require a bit more attention. Learn More >


Leading Question

Leading questions use biased language to influence respondents in a direction. Keep questions as simple and direct as possible, taking care to remove any editorializing adjectives that may betray a preference towards one response or another. Learn More >

Loaded Question

Loaded questions make assumptions. For example, asking “what kind of candy do you get when you go to the movies?” assumes the respondent goes to the movies, eats candy and gets candy at the movies. This creates a biased environment where the respondent may not be able to answer accurately. Learn More >


Order Bias

The order you list question responses in a quantitative survey question can bias respondents towards choices higher on the list. If your survey platform has the ability, use answer shuffling to remove the potential for this bias. Learn More >


Prestige Question

Inserting prestige into a question by citing an expert opinion one way or another biases the respondent towards the expert’s opinion. For example, “Doctors say smoking causes cancer. Do you agree?” or “Do you support the President’s policy on Zimbabwe?” Learn More >


Self-Selection Bias

This occurs when respondents are allowed, through whatever means, to opt out of taking a survey based on its content. This biases the results by oversampling people with a previous interest or overly polarized opinion on a given topic. Learn More >

Social Desirability Bias

This occurs when respondents will attempt to discern what the researchers want them to say and respond accordingly. This often occurs when participants see the brand or topic ahead of time and try to give answers the company or researcher may want to hear. Avoid exposing this information, or asking any leading or loaded questions that may betray an interest. Learn More >


Respondents who complete surveys far too fast to have actually read the questions or truly contemplated answers. Pollfish removes these respondents through AI and machine learning algorithms. Learn More >

Stereotype Question

When researchers infuse bias into survey questions, they open the potential for respondents to agree with or defy that bias. If, for example, a survey question reminds respondents that gender stereotypes exist around math, that question may have planted a seed in the mind of respondents that changes responses. Avoid this type of language at all costs. Learn More >


When respondents answer using the first available choice for every question. Respondents answering in familiar patterns in order to complete surveys faster will be removed from the Pollfish sample. Learn More >

Section 4: Pollfish Features

Sometimes, we get so excited about the features of Pollfish, we forget to define them all for you!

So we decided to explain them all in one place. Feel free to bookmark this section so you can return to it as you read through

A   |   P   |   Q   |   R   |   S   |   W


Advanced branching

Also known as skip logic, advanced branching allows you to show respondents different questions based on responses to earlier questions. For example, if a respondent says they have tried your product, you can show them one set of questions and if they have not, you can show them different questions. Branching provides an alternative to screening questions. Learn More >

App Monetization

App monetization refers to the method or methods app publishers use to make money on their apps. There are a variety of app monetization strategies and platforms. Pollfish provides a way for app publishers to immediately monetize their creations by installing the Pollfish SDK and exposing their user base to Pollfish surveys. Learn More >


Programmatic Advertising

Programmatic advertising refers to the automated buying and selling of online advertising. Pollfish functions on a programmatic advertising model, delivering surveys in much the same way advertisers deliver ads in apps. Learn More >



If a survey must meet a set percentage of participants from an age or gender group, you can set quotas.


Radius Targeting

When setting geographic targeting on your Pollfish survey, you can select a city, congressional district or town. You can also use our Radius Targeting tool to select a specific radius on a map to target your survey. Learn More >


Skip Logic

Also known as Branching, skip logic allows you to show respondents different questions based on responses to earlier questions. For example, if a respondent says they have tried your product, you can show them one set of questions and if they have not, you can show them different questions. Skip logic provides an alternative to screening questions. Learn More >

Survey Stratification

Stratified sampling involves dividing a population into smaller sub-populations called strata. There are many ways of accomplishing this, including balancing using predetermined quotas before you run your survey, or weighting your results after the fact. Learn More >



A type of stratification where quotas–often pegged to pre-existing data like a local or national census–are used to weight respondent pools to more accurately reflect the population being surveyed. With Pollfish, you can weight your survey results to the local or national census (where available) with the push of a button. Learn More >

Survey Bias: Your Ultimate Guide To Reducing Bias And Increasing Data Quality


Section 1: Sampling Bias

Section 2: Questionnaire Bias

Section 3: Survey Fraud

Section 4: Pollfish Can Help!

Survey Bias: Your Ultimate Guide To Reducing Bias And Increasing Data Quality

Survey bias is a broad term. It encompasses all parts of a survey that can contain bias. And, in case you didn’t know, there are a lot of trouble spots in the average survey. 

But not to worry. Once you know what to look for, and have the right tools to root it out, survey bias will quickly become a thing of the past for your market research team. 

But first things first: why should you care? 

Put simply, survey bias impacts data quality. By eradicating survey bias (as best you can), you build trust in the data you’ve collected. 

That’s no small thing, as companies across the world are using consumer insights data to inform everything from content creation to product timelines to strategic planning.

But where to begin?

Disclaimer: Preventing All Bias Is Impossible

Before we go any further, we should clear something up: no survey is 100% free of bias. Little things from the incentives and influences of the questionnaire writer to the words used to various forms of sampling bias can negatively impact results when a questionnaire is distributed to hundreds of users. 

Not all bias is preventable, but there are a few key areas and best practices you can use to prevent bias in your survey. 

Once you have done everything you can to ensure your survey is taken from a representative sample, contains a variety of unbiased questions and is delivered to real consumers, you will be ready to get great consumer insights.

Let’s start with the basics.

The Different Types Of Survey Bias

Survey bias falls into three categories: Sampling Bias, Questionnaire Bias and Survey Fraud. There are other kinds of bias out there to watch out for (some of which we’ll get into at the end) but most forms of bias you should watch out for fall into these three categories. 

We’ll start with sampling bias.

What is Sampling Bias (and How to Avoid It)?

Sampling bias, quite simply, refers to bias that occurs in the selection of your survey respondents, also known as your sample.

There are many different types of sampling bias to watch out for. 

Representation Bias

If you select your sample of a town based on a respondent’s presence at the mall, you could be excluding key members of the local population. For example, you may be over-representing healthy people of a certain age who get out of the house, people with the disposable income to shop, etc. 

How to avoid it

Digital sampling methods help with this. By collecting responses from users on their phones, through social media or by email, you avoid the types of sampling bias that an in-person surveyor would contend with. If you want to take it one step further, Pollfish Survey Stratification can help (more on that later). 

Post-stratification allows results to be weighted for census data in a single click, reducing sampling bias.

Exclusion Bias

Removing trial participants can create sampling bias. If, for example, your survey requires follow-ups but excludes original participants who have since moved out of the study area, you could skew the results.

How to avoid it

Keep initial and follow-up samples the same. 

Pre-Screening Bias

This occurs when you advertise for volunteers within particular groups. 

How to avoid it

One way to avoid this would be to reduce targeting on your survey collection method of choice, and instead use screening questions on your survey to screen out people who may not have the qualifications you are looking for.

Survivorship Bias

An over-emphasis on “surviving” respondents can create sampling bias. For example, doing an economic study of the business climate and only including current businesses, without including ones that failed.   

How to avoid it

When doing customer satisfaction surveys, for example, it is important to get feedback not just from your current customers–who are likely satisfied–and potential customers–who don’t know how they feel about you–but from your former customers who you likely failed to satisfy. This way, you can truly know how satisfying your product or service actually is.

 More On This…

weight-survey-resultsHow to weight your survey results in Pollfish


Other Ways to Avoid Sampling Bias

Stratified sampling

Stratified sampling is the process of dividing a population into subgroups and selecting a balanced sample audience from those groups to decrease the sampling error. These populations are usually large and diverse. Because random sampling from a given audience can be unbalanced, stratified samples apply weights to audience characteristics against a control group to ensure that the responses are representative of a given population.


  1. If you wanted to survey the population of Dallas, TX, you could weight your survey sample audience demographics against census data for Dallas to create a more accurate reflection of the demographic makeup of that city.
  2. If you were conducting a political survey and needed representation of various minority groups (such as race or religion) that may not appear in a random sample, stratified sampling could ensure that some members of those groups are included and their responses adjusted to present a proportional segment of the population as a whole.


Quotas are another way to weight an audience and reduce sampling bias, but instead of applying a weighted number to audience characteristics to estimate proportionality to the population, quotas fix the proportions of an audience so that responses can be collected from people who exactly fit the criteria. While they typically take longer, quotas offer more precise data. They’re better used for a smaller sample population where the weighted demographics are known.


    1. In a survey for a magazine where readership was known to be 80% female, quotas to survey an 80% female audience would be used.
    2. If a product is geared towards millennials, age quotas might be used to ensure that the survey is only completed by a randomized sample of those between the ages of 25-40.

What is Questionnaire Bias (and How to Avoid It)?

When building your survey questionnaire, the way you ask a question matters just as much as, if not more than, what you are asking. 

Questionnaire Bias (also known as Response Bias) can come in many forms. Where most researchers get in trouble is they bias respondents unintentionally, due to how they ask a question, what question type they use or how they administer the questionnaire. 


Self-Selection Bias

Also known as Non-Response Bias, Self-Selection Bias can occur when respondents are allowed, through whatever means, to opt out of a survey due to its content. For example, not everyone who eats at a restaurant is forced to leave a review. Therefore, most restaurant reviews are biased toward people who had a very good or very bad experience. You also see self-selection bias when there is a knowledge gap on a topic. If the topic or subject of the survey is known and people can opt out, the survey will get more participants with advanced knowledge of or interest in the topic or subject. 

How to avoid it

You can avoid this type of bias in two ways–you can either remove the ability for participants to opt-out once they’ve seen the brand or topic of the survey, or you can keep the brand or topic hidden throughout the survey.  

Acquiescence Bias 

Also referred to as “yea-saying,” Acquiescence bias occurs when there are too many questions asking participants to agree or disagree. Respondents have been shown to overly agree, even when this causes contradictory responses. 

How to avoid it

Using a variety of question types will remove this type of bias. Don’t rely too heavily on any one question style. And if you can, review responses and throw out any with deeply contradictory “yea-saying.”

Order Bias

The order in which you list selections in a single-selection or multiple-selection question biases respondents towards choices higher on the list. Similarly, responses change based on the order survey questions appear. 

How to avoid it

If your survey platform has the ability to shuffle questions and/or responses, take advantage of this feature. You will instantly remove this type of bias from your survey.

Shuffling answer order reduces bias

Extreme Responding

Some respondents just like extremes. So, if given the option, they will always select the most extreme option. 

How to avoid it

Similar to acquiescence bias, Extreme Responding is the result of too many of the same type of question. If you are going to use scales or sliders that welcome extreme responses, use other types of questions to gauge user feeling as well. This will give you a more nuanced understanding.

Social Desirability Bias

There are different forms of this bias. For example, some participants seem to believe there is a way in which they are “supposed to” act when participating in a survey. Typically, this comes from respondents who believe in the importance of research and want to be good subjects, so they will try to discern what the researchers hope will happen in the study and give them the answers they want. Or the respondent will overly-report behaviors they believe the researcher may want to hear for other reasons. The most famous example of this is in political polling in the US in 2016, where subjects hid their intention to vote for Donald Trump from pollsters.   

How to avoid it

You can start by keeping the brand or topic hidden from participants, and just asking them to respond so they don’t know who the answers are for. Secondly, eliminate leading or loaded questions that may convey an intent or interest.

Question Types to Avoid

Loaded Questions

Loaded questions make assumptions. For example, “What is your favorite snack when you go to the ballpark?” is a loaded question, as it assumes the person being asked likes sports, has been to a ballpark and has eaten snacks at a ballpark. 

How to avoid them

Assume nothing. If you want to survey people who enjoy snacks at the ballpark, try a screening question or two. This will ensure you are only asking people who fit the profile. 

Set targeting first and screen respondents from a qualified audience.

Leading Questions

Leading questions use biased language to push respondents toward a response. 

How to avoid them

Keep questions simple and remove any language that may imply intent. Complimenting or insulting a product or service should be off-limits. If you can, have a friend with some distance from the subject review your questionnaire.

Double-Barrelled Questions

Double-barrelled questions happen when researchers blend two questions into one, and then allow for only one answer. This creates a fallacy for the respondent and often results in incomplete or biased answers. For example, “Did you find the product interesting and helpful?”

How to avoid them

Make sure each question asks for a single answer. If you have double-barrelled questions, they should be split into two questions, i.e. “Did you find the product interesting?” and “Did you find the product helpful?”

 More On This…

How to write good survey questions


Prestige Questions

These questions insert knowledge from broadly accepted experts into questions, which biases respondents to agree with the expert. For example, “Doctors say smoking causes cancer. Do you agree?” or “Do you support the President’s policy on Zimbabwe?”

How to avoid them

Avoid asking questions beyond the cognitive grasp of the average person. If your question assumes a level of knowledge, use screening questions to remove people who don’t consider themselves knowledgeable. If you provide information on a topic you are asking about, be sure to remove bias from your explanation and only ask participants to respond to the information included. When possible, use multimedia (video, audio, images), as these can be less biased than adding prestige to the information.

Negative Questions

These questions force respondents to remember a traumatic event or think about something negative. Asking about health issues, car accidents, money woes or even divorce all qualify.

How to avoid them

If these types of questions exist in your survey, you can remove them (as they will likely bias respondents towards responding more negatively), or you can move them to the end.

Stereotype Questions

Many researchers may unintentionally stereotype their respondents. For example, reminding respondents that racial stereotypes exist around driving or that gender stereotypes exist around math ability may bias respondents towards acting more stereotypical, or trying to defy those stereotypes instead of delivering honest responses. 

How to avoid them

Remove this kind of language from your survey, and don’t ask respondents to describe themselves by race, ethnicity, age and the like, unless you have to. 

Absolute / Pushy Questions

Questions with Yes or No answers can bias respondents because they don’t allow for any range in between. This may cause respondents to not answer or answer Yes when they really mean Only Sometimes. For example, do you eat breakfast in the morning, yes or no? If someone doesn’t eat breakfast 100% of the time, they may answer “No” where another person may answer “Yes” because they do sometimes eat breakfast. The result is imprecise and biased responses. Taking this even further, some researchers will make a declarative statement and ask the respondent to agree or disagree (Spain is too hot in the summer, agree or disagree?). This presents bias because survey takers who don’t have a strong opinion will be more inclined to agree with the statement. 

How to avoid them

Remove Yes or No questions from your survey. Then, read your questionnaire from a user perspective and remove all statements that may lead the user toward one response or another. Remove all declarative statements, and simply ask questions.

Ambiguous / Confusing Questions

Unclear wording, referring to previous questions and using indirect language are just some of the ways researchers confuse respondents. For example, “Was your breakfast not incorrect when it arrived at your table?”

How to avoid them

Read your survey questions out loud. This will allow you to hear how they may sound to a reader. You should be able to spot examples of confusing language. 

What is Survey Fraud (and How to Avoid It)?

Traditionally, market researchers would convene panels– groups of people who have agreed to take surveys. In most cases, panelists were convened in person and offered some sort of semi-lucrative incentive. Think airline miles, gift cards and even cold, hard cash. Panelists were convened in person, preventing things like panelists taking the same survey multiple times.

However, researchers eventually sought more randomized sampling. Enter random digit dialing, where an auto-dialer machine would call random people to collect responses for surveys. This reduced bias as respondents were selected at random and screened after the fact.

Then came the internet. 

 More On This…

Random Device Engagement and Organic Sampling


Suddenly, research panels re-emerged. Now online, backed by global reach and offering respondents all over the globe the ability to take surveys for pay, Panels were back with a vengeance. 

This incentive structure, combined with the scale of these efforts, created an opportunity for fraud. Survey takers, looking to accumulate as many rewards as possible in as short a period of time as possible would take as many surveys as possible, barely reading the questions before providing answers. 

While most reputable panels seek to remove these fraudulent responses, panel participation has been on the decline, and panels often fall short of the number of respondents necessary to create a representative sample for their customers. 

This requires panels to bridge this gap wherever they can find it by purchasing sample from wherever they can find respondents. When they lose control of the panel vetting process in-house, they have less information about the respondents, recruitment methods, or whether or not they are an appropriate choice for the survey’s target audience.  

But who are these problematic survey-takers and how can you spot them? 

At Pollfish, we created AI and Machine Learning algorithms to find, locate and remove these fraudulent responses (more on that later.) 

If you see these types of responses popping up in your surveys, make sure you get rid of them as they don’t contribute to a representative sample. 


Some companies pay people to take surveys. The amount per survey is often low, so professionals take a ton of surveys to make money as part of their overall income. These respondents will often have accounts on multiple sites, multiple aliases, etc. Their fast, haphazard movement through your survey can be disastrous to data quality.    

How to avoid them

Don’t use traditional panels. Survey platforms like Pollfish deliver your survey inside mobile apps, to active consumers. Respondents receive in-app benefits, not monetary incentives. 


Rule-Breakers skip key instructions, offer incomplete responses or contradict themselves. 

How to avoid them

It can be hard to know if this is due to speeding through towards a reward or just a mistake, but either way, if there is too much rule-breaking, it is best to filter them out. 


Much like professionals, speeders can’t possibly have completed your survey accurately in the time they took. But if you aren’t tracking for this, how would you know? 

How to avoid them

Set a time threshold and timestamp respondents. If you determine that they moved through your survey too quickly to successfully complete it, filter them out. 


Straight-liners are some of the easiest fraudulent respondents to catch. These respondents give the same or similar answers for every question. Whatever the first choice is, that’s what they select. 

How to avoid them

If you spot this in your responses, or your platform can flag it, remove these respondents as there is a high likelihood their responses are fraudulent or unhelpful to your study.


Some survey takers know they may be filtered out and not get a reward if they answer like a Straight-Liner. So near-straight-liners will answer all but 1 or 2 questions in a straight line.

How to avoid them

Same as above. Remove. 


Straight-Liners, Near-Straight-Liners and Alternators will select their answers in familiar patterns in order to complete surveys quickly. The Pollfish Survey Methodology uses AI and Machine Learning to screen out these low-quality responses.



Remember the kid in class that didn’t study for the test so they’d just fill in the bubbles on the scan sheet into diagonal lines (think A,B,C,D,C,B,A) all the way down? That kid was an alternator.

How to avoid them

These can be a little harder to spot. It would help if your survey platform could do this work for you. 


For the more computer savvy fraudster, creating bots to take surveys is the easiest way to game the system. While most popular panels have installed protections against bots and have abandoned particularly vulnerable methodologies like River Sampling (using banner ads as a recruitment process without verifiable participant info), these panels may still buy respondents from companies who don’t do enough to protect against bots. 

How to avoid them

Find out where your survey respondents are coming from. Make sure none of the providers in your sample are using outdated methodologies like River Sampling. If your survey platform has bot detectors and can throw out common bots, all the better. 

Fake Accounts

Just like bots, some survey takers will create fake accounts, allowing them to take the same survey multiple times. While most survey panels check IP addresses to protect against this, these protections are easily circumvented. 

How to avoid them

Check if your survey platform allows accounts to take a survey more than once or repeat or restart a survey once they have started. Make sure to have a mechanism in place to filter by IP address, or look for other patterns that may indicate fake accounts. Surveys that allow link sharing are more likely to suffer from repeated survey attempts, whereas those that use a randomization methodology such as RDD or RDE approach the respondents and use a unique identifier to prevent repeats.

Biased Respondents

This last one is nearly impossible to prevent if you are working with a traditional online panel. Forums and other online meeting places allow biased respondents to find pre-recorded answers to your survey shared by other respondents. This allows them to quickly answer surveys without adding their own, natural feedback. 

How to avoid them

The internet is a big place and you, as a researcher, cannot easily protect against this. You can prevent survey takers from cutting and pasting inputs, you can prevent screenshots and other measures that prevent sharing responses. But if you want to prevent this problem, it may be best to avoid survey panels all together.

Pollfish Can Help

AI / Machine Learning

Pollfish AI and Machine Learning removes bad survey takers. Read more about that here.

Individual Review

Pollfish individual review (by a person) ensures your survey follows best practices. Each survey you submit is subject to a brief (30 minute) review to ensure you have not egregiously biased your study.  

Survey Stratification

Recently, Pollfish launched Survey Stratification, which allows you to tie weighting of your survey sample to the US Census. With the push of a button, you can weight your respondents to match the Census demographics. This data is available on a national level in the US, as well as in all 50 states. It is also available across the EU and UK, as well as Iceland, Liechtenstein, Norway and Switzerland. This feature can reduce sampling bias. Read more about this here.