The Impact of Cognitive Bias on Student Surveys

March 16, 2022 Watermark Insights

the impact of cognitive bias on student surveys - woman with pen working on laptop

Survey response bias is a challenge for school administrations when collecting accurate data. However, understanding the types of bias in student surveys and course evaluations can help you learn how to avoid them, creating more precise and dependable research for your higher education institution.

What Is Cognitive Bias?

Cognitive bias is the subconscious decisions and judgments we make every day. Because cognitive bias is subconscious, we are unaware of how it shapes our decisions, making actions influenced by cognitive bias unintentional.

This type of bias originates from the brain's attempt to process and understand all the information it receives every second. Psychologists estimate that every second, our brains receive around 11 million bits of information, but we can only process about 40 pieces of information. Because of this, the human brain creates mental shortcuts to filter information and allow the brain to remain in the present. These mental shortcuts often cause cognitive bias, affecting several brain functions, like memory and attention. 

In contrast to cognitive bias, conscious bias is when people make decisions or judgments and know how their opinions and beliefs result in their actions, behaviors, and attitudes. Conscious bias is an essential aspect of formulating a sense of self and identity within a community. Since both go into decision-making and judgment, they both impact survey results. However, since cognitive bias is unconscious, it comes with unique challenges requiring researchers' awareness.

keyboard with paper on top that says "bias"

What Is Cognitive Bias in Surveys?

When conducting course evaluations at your higher education institution, it is essential to understand what is cognitive bias in surveys and what it looks like. While people encounter cognitive bias frequently in their everyday lives, it often looks different in survey and research settings and can take on several forms, including:

  • Sampling biases: Biases at this level involve how researchers target and reach their audience for taking their survey. Researchers can encounter bias if they only market their survey to a small group of people, tailoring responses to reflect how only that group would respond. Further, if researchers limit how they market their survey, they can also encounter sampling bias.
  • Response biases: Researchers and college administrations can encounter survey bias at the response level in many ways. Researchers need to be careful how they frame and order questions to ensure they don't lead to specific responses. However, respondents will also bring their cognitive and conscious biases to surveys, affecting how they interpret questions and interact with survey elements.
  • Researcher biases: Researchers also have unique preferences and desired results for their surveys. To prevent this type of bias, researchers need to be careful about how they create their surveys and evaluate results. Researchers can write questions with the desired inquiry in mind or ignore research to get the information they want from their data.

Why Is Cognitive Bias Important in Understanding Survey Results?

Researchers need to understand where they can encounter bias in surveys to help protect the surveys' validity and accuracy. Especially when looking for bias in course evaluations, a school's administration can help avoid common researcher bias and be aware of how improvement suggestions might not be an accurate reflection of student needs.

Learning about cognitive and survey bias can help your administration better understand how responses may have linked preferences.

7 common types of response bias in surveys

7 Common Types of Response Bias in Surveys

Cognitive bias takes many different forms in survey responses, which are essential for researchers and higher education administrations to understand as they collect survey responses. When evaluating your survey results, you will want to keep these types of response biases in mind to understand better how they may affect your results and analysis and discover how to avoid bias in surveys.

1. Recency Bias 

With recency bias, survey respondents will emphasize recent events rather than older ones, even if older ones are more relevant or impactful than current ones. This occurs because the brain remembers recent events better than older ones, allowing people to feel more confident about their responses when talking about a recent event.

Because we receive so much information every second, our memory has flaws. Recency bias is one way that memory's flaws show in surveys. If you include questions on your survey that ask your respondents to recall events, you will want to consider these biases when analyzing your results. Recency bias in course evaluation surveys may look like asking students to reflect on the semester and students responding in a way that only represents the past couple of weeks, creating less accurate data.

2. Confirmation Bias

Confirmation bias is one of the more common forms of cognitive bias in surveys and primarily affects researchers and data analysts. Confirmation bias occurs when people receive new information and use it to confirm their existing opinions or beliefs. However, it often manifests in people ignoring information that contrasts their views and ideas, allowing them to continue receiving confirmation from information that supports their ideologies.

In research, confirmation bias typically looks like researchers intentionally ignoring specific information to get the results that support their initial hypothesis or receive the desired response. When conducting surveys, it is essential for faculty and administrators to be aware of confirmation bias when evaluating survey results to ensure they gather accurate data reflecting the given answers. In course evaluation surveys, administrators may ignore undesired student feedback in favor of responses that support the changes they already want to make.

However, many people experience confirmation bias every day and outside of research settings, and researchers should be aware of this too. People receive information across various platforms every day — including social media, which often uses algorithms to tailor feeds to match beliefs and interests. Because of this, people often believe the information presented to them as fact because it supports something they already believe.

3. Anchoring

While many types of response bias come from external and personal responses, how you frame and order questions can also impact biased responses. Anchoring bias occurs when people face multiple options when making a decision and base their judgment and understanding on the first factor they encounter, also called the anchor. Like confirmation bias, people face anchoring bias daily.

For example, people often face anchoring bias when shopping. If they are searching for a specific object, like a new coat for the winter, they will use the price of the first coat as their baseline for all other coats they encounter while shopping. While many people see this method as helpful for finding a good deal, the first object they encounter might not properly reflect the market value or may cause them to overlook other factors, like quality.

In surveys, anchoring can impact people when presenting them with too many choices. Instead of giving each question their full attention, they may overlook or ignore information, leading to results that don't properly reflect their opinions or beliefs. Instead, they base their responses on an anchor response or factor, determining how they answer all other questions. In course evaluation surveys, asking students about all their classes on one questionnaire can result in them picking one course as their anchor for their responses.

4. Framing Effect

When writing higher education surveys, you can help lessen and understand response bias by examining how you frame and present information in questions. Framing questions can lead people to answer a specific way on surveys, causing higher chances of bias in results.

The framing effect significantly impacts criminal investigations and judicial processes. If police or lawyers frame questions in a way that leads respondents to the answer they want, judges can throw out responses and information because they exemplify framing effect bias.

When developing your surveys, you can help prevent researcher-level bias and the framing effect by watching how you frame questions and ensuring they can lead respondents to answer how they wish. Framing in course evaluation surveys may look like asking students if they believe they were highly satisfied with their course instead of asking them to rank their satisfaction.

5. Response Biases

Response bias can come from how researchers and administrators frame and order questions, but to prevent bias on their end, they should understand that respondents can also have bias. When answering surveys, individuals can experience many types of response bias, including:

  • Extreme response bias: This type of bias applies to questions asking respondents to rate or rank items. Under this bias, respondents will only choose factors on the extreme sides of the scale. This can cause inaccurate representations of opinions and beliefs in surveys with these questions.
  • Neutral response bias: Neutral response bias is the opposite of extreme response bias. Instead of only choosing the extremes, respondents will only choose neutral responses. You can experience both types on the same survey since each person will experience different factors leading to their cognitive biases.

In higher education surveys, the school administration may ask students to measure their satisfaction with their courses and professors. However, response bias can lead students to respond using only extreme or neutral options, decreasing the accuracy and making it harder for administrators to judge course and professor performance. Researchers can do their best to prevent response bias through careful question framing, ordering, and wording. However, since respondents have their cognitive biases, researchers should be aware that survey answers will still have some bias.

6. Order Bias

Also called the response order effect, order bias is similar to the framing effect. Instead of being about question wording and premise, it involves how researchers order questions together. Under the order bias, respondents will answer questions differently if researchers order the questions differently.

For example, if a survey asks students how happy they are with their academic program and then follows up with how happy they are with their professors' performance, they may answer the second question differently than they would have if researchers placed the second question elsewhere. The placement may lead students to link their program with their professors, and even if they feel differently about their professors, they may answer with this link in mind.

7. Nonresponse Bias

Nonresponse bias is sampling bias and how researchers understand their sample group. While you may target your survey to all desired demographics and take the proper measures to ensure you are reaching them all, there is no guarantee that respondents will take your survey.

Many researchers think there is something fundamentally different about the people who don't respond and put a lot of effort into understanding those reasons. However, people have many reasons for not wanting to take surveys, ranging from not having the time or interest to worrying about response anonymity. Instead, researchers should focus on understanding their respondents and their answers.

man wearing dress shirt and tie talking on mobile phone

How to Avoid Bias in Student Evaluation Surveys

Avoiding cognitive bias can make it challenging for school faculty members to create effective course evaluation surveys. However, you can use many tips for determining how to avoid bias in a survey, including:

  • Provide open-ended response options: You can help prevent response bias and lead questions by offering students more freedom with their responses. Consider offering answer selections like “not applicable,” “other,” “all or none of the above,” and “I don't know.” These options will allow students to select how they feel instead of choosing an answer because it's available.
  • Mix topics throughout the survey: Prevent anchoring or order bias by scattering questions throughout your survey, decreasing the chances of your students linking certain questions. With this method, questions will have independent importance and provide more accurate data.
  • Avoid emotional language or statements: Emotional language or answer selections, like asking students to choose their agreement level, can lead to various types of cognitive survey bias, including recency bias or extreme and neutral response bias. Instead of asking students to rate their agreement about their course's success rate, you could ask them to rank school success values for that course.
  • Ask only one question per survey item: Preventing confusion is essential for avoiding the framing effect. When students understand what you are asking them, they can more accurately answer your questions. Separating questions that ask multiple things can help keep confusion and the framing effect at a minimum.
  • Keep researcher interactions neutral: Regardless of how you conduct your course evaluations, you will interact with your students as they respond. Prevent researcher bias in course evaluations by ensuring all language and instructions are neutral, so respondents can answer how they feel instead of how you want them to feel.

Your school administration can even prevent nonresponse bias in your marketing efforts. Since students will reject your survey for many reasons, you can try to reach initial nonresponders by frequently contacting them. Students who didn't have time or forgot about the survey the first time can have the opportunity to take it. Further, offering incentives like raffle entries can motivate students to take the survey, helping reduce nonresponse in your course evaluation bias.

request a demo

Streamline Survey Data Collection With Watermark

At higher education institutions, course evaluations and student feedback are vital for faculty to understand where to implement improvements for upcoming semesters on the course level. Ensuring your assessments and surveys have minimal bias can help strengthen survey data and insights. While understanding the types of bias occurring is essential, you can also implement more neutral habits into your survey to prevent bias.

Our Course Evaluations & Surveys platform helps you easily and efficiently collect crucial information about your courses. Our software rapidly collects data by integrating it into your resources, so you don't need to input information manually. Further, we design our software specifically for higher education faculty and administrators and streamline your many processes for more efficient planning and implementation. 

Request a demo from Watermark today.

About the Author

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut euismod velit sed dolor ornare maximus. Nam eu magna eros.

More Content by Watermark Insights
Previous Article
Live Solution Tour: Applying Indirect Measures to Outcomes Assessment
Live Solution Tour: Applying Indirect Measures to Outcomes Assessment

See how to link student feedback to institutional learning and non-academic outcomes to gain deeper insight...

Next Flipbook
The Essential Guide To HLC Accreditation
The Essential Guide To HLC Accreditation

This guide offers simple steps to align your assessment process with HLC accreditation requirements and sta...