What is Social Desirability Bias? | Definition & Examples
Social desirability bias occurs when respondents give answers to questions that they believe will make them look good to others, concealing their true opinions or experiences. It often affects studies that focus on sensitive or personal topics, such as politics, drug use, or sexual behavior.
Social desirability bias is a type of response bias. Here, study participants have a tendency to answer questions in such a way as to present themselves in socially acceptable terms, or in an attempt to gain the approval of others.
It is especially likely in self-report questionnaires, but it can also affect the validity of any type of behavioral research, particularly if the participants know they’re being observed. However, there are ways to detect and reduce research bias in your research design if you know what to look for.
Table of contents
- Why does social desirability bias occur?
- Types of social desirability bias
- Why social desirability bias matters
- Social desirability bias examples
- When does social desirability bias occur?
- How to reduce social desirability bias in your research design
- How to detect social desirability bias
- Other types of research bias
- Frequently asked questions about social desirability bias
Why does social desirability bias occur?
While social desirability bias may be caused by the nature or setting of the experiment, it’s important to remember that the desire to act in a culturally appropriate and acceptable manner is deeply rooted in human nature. For this reason, the mere presence of a researcher or other participants may trigger some level of socially desirable responding. It is inherent to conducting human-centered research.
However, each individual respondent may also have their own reasons to want to be perceived a certain way (e.g., seeking approval or desiring praise), as well as expectations regarding how their behavior will be evaluated by others.
Types of social desirability bias
In general, there are two types of social desirability bias:
This distinction is important because it accounts for both situational factors (related to a situation) and personal factors (related to personality traits) that can result in socially desirable behavior. While situational determinants can be influenced by the researcher, personal determinants are less easily controlled for. These can usually only be detected after the fact.
Self-deceptive enhancement occurs when the respondent believes something to be true when it is not. In this case, the respondent is neither aware of portraying themselves positively nor consciously trying to, but does so anyway.
On the other hand, when people engage in impression management, they are aware of their overconfident self-appraisal and intentionally seeking to keep up with social or group norms in order to avoid negative evaluation or judgment.
Why social desirability bias matters
Social desirability bias is one of the most common sources of research bias. It leads to over-reporting of socially desirable behaviors or attitudes, and under-reporting of socially undesirable behaviors or attitudes. As a result, reported answers will differ from true answers.
Socially desirable responses can bias results in three main ways:
- Social desirability can cause a spurious correlation to occur between variables, making them appear to have a causal relationship when they do not.
- Social desirability can hide relationships between variables by acting as a suppressor variable.
- Social desirability can act as a moderator variable, even though it may be uncorrelated to either the independent or dependent variables.
Social desirability bias examples
You need to consider social desirability bias when deciding what research design would work best for you.
When does social desirability bias occur?
Unfortunately, it’s often not possible to fully prevent or remove social desirability bias from your research. However, it is important to identify and control for the influence of this bias, starting with your research design. The first step here is to recognize and anticipate conditions where bias is particularly likely to occur.
These can include:
- Research designs that incorporate self-report measures
- Studies involving personal or sensitive topics
- Situations in which subject anonymity is compromised or not guaranteed
- Instances when subjects anticipate that their responses will result in judgment from others
- Situations in which participants belong to similar social groups as the interviewer
How to reduce social desirability bias in your research design
There are a few strategies that you can use to help you reduce social desirability bias in your research design.
When asking about sensitive topics such as drug use or breaking the law, it’s important to reassure the study participants that their identities will be protected. If your results are appropriately anonymized, you may receive more truthful answers.
Be careful of leading questions that can influence a respondent’s answer. The phrasing of a questionnaire item can trigger a socially desirable response, even when the respondent doesn’t have the tendency to respond in such a fashion.
Giving respondents the opportunity to fill out a questionnaire at a time and place where they are undisturbed by others may lead to more truthful answers. As mentioned above, the presence of the interviewer or other participants may give them conscious or unconscious cues to answer in a socially desirable manner.
Online surveys or questionnaires in particular are a great way to minimize socially desirable responses. However, note that the absence of an interviewer here limits the usefulness of self-administered surveys to cases where the questions are not complex. You are also limited to questions that are mainly closed-ended.
Indirect questioning asks respondents to answer a set of structured questions from the perspective of another person or group. A typical indirect question asks respondents to make predictions about how someone like them would think or act in a particular situation.
The underlying idea here is that indirect questions can reduce social desirability bias. As respondents feel that they are giving information about situations based on objective facts rather than their own opinions, they project their attitudes into the response situation
A forced-choice question requires the respondent to provide a specific answer, without giving a “nonresponse” option such as “no opinion,” “don’t know,” “not sure,” or “not applicable.” Depending on the specific forced-choice design, respondents are either asked to:
- Choose an item within each block that is more descriptive of them
- Choose the most descriptive and the least descriptive item
- Provide a full ranking of items within each forced-choice block
How to detect social desirability bias
You can detect and measure social desirability bias using two methods:
Social desirability scales
A number of social desirability (SD) scales have been developed in an effort to detect and measure socially desirable responses in data collection. These include the Marlowe–Crowne Social Desirability Scale, the Martin-Larsen Approval Motivation Scale, the Self-Deception Questionnaire (SDQ), and others.
These scales involve a number of true or false statements concerning personal attitudes and traits. They describe socially desirable but statistically unlikely behaviors, such as “Before voting, I thoroughly investigate the qualifications of all candidates.” Based on their answers, each respondent is assigned a social desirability score. Individuals who agree with these statements will have a higher score.
After checking for correlation with social desirability scores, you can then use three tactics:
- Reject the data from subjects you deem to have scored too highly (outliers).
- Correct data from high scorers (for example, by using data of a control group).
- Note or register the impact of social desirability bias in your research paper.
Rating of item desirability
Item desirability is measured by asking subjects to rate an item (e.g., a question or a statement) on a desirability scale. For example, if “being happy” is an item, subjects are asked to indicate how desirable being happy is to them. This rating is added to each of the questions in the survey.
Keep in mind that adding a rating to each question can be an inconvenient method if there are many questions, as the length of the questionnaire is effectively doubled.
Other types of research bias
Frequently asked questions about social desirability bias
Cite this Scribbr article
If you want to cite this source, you can copy and paste the citation or click the “Cite this Scribbr article” button to automatically add the citation to our free Citation Generator.