Question

There are many types of bias that can affect research outcomes. Research and discuss the following types of bias that can affect the credibility of resources. Please give specific examples of how these types of bias can alter results of studies.

• Sampling bias
• Selection bias
• Interviewer bias
• Response bias
• Observation bias
• Leading questions and wording bias
• Sponsor bias

Answer 1

1. Sampling bias:

Sample selection bias / biased sampling refer to errors that occur in research when the researchers do not select their participants properly and / or if it systematically favors some outcomes over the others.

It is very important for the researchers to follow sampling methodology and chose participants randomly still adhering to criteria of the study. If researcher fails to select their participants using specific sampling methodology, they will have a risk of severely impacting the validity of their research results and findings as their participants does not reflect the population of interest accurately.

An example of sampling bias:

Sampling bias is very common and can even be committed by most experienced professionals. Below is one of the most powerful examples of sampling bias being committed on a grand and impactful scale occurred during the Truman-Dewey US presidential race of 1948.

During the race, nationwide political telephone survey was conducted. Results of the survey implied that Dewey would win over Truman in a heavy-handed landslide; however, the study failed to account for the fact that telephones were still a fairly revolutionary and expensive form of technology. Due to the cost of telephones in 1948, only a small number of wealthy families owned them. Therefore, the political telephone survey was only presented to participants that were part of relatively wealthy families, and at the time, wealthy families tended to support Dewey while lower-middle class to lower class families were more likely to support Truman. Failing to consider the population of Americans that owned telephones in 1948, the researchers conducting the telephone survey committed sampling bias. As a result, they received severely skewed response data. Instead of distributing a more effective survey to a sample that more accurately represented the population of the United States at the time, the researchers ended up with inaccurate and unrepresentative insights. The conductors of the survey were confident that Dewey would win the presidential race with ease, while in the end, it was Truman that ended up becoming the leader of the free world.

2. Selection bias:

Selection bias is a distortion in a measure of association due to a sample selection that does not accurately reflect the target population. Selection bias can occur when investigators use improper procedures for selecting a sample population, but it can also occur as a result of factors that influence continued participation of subjects in a study. In either case, the final study population is not representative of the target population – the overall population for which the measure of effect is being calculated and from which study members are selected.

Selection bias occurs when the association between exposure and health outcome is different for those who complete a study compared with those who are in the target population.

There are many sources of selection bias:

• Selective survival and losses to follow-up
• Volunteer and non-response bias
• Hospital patient bias (Berkson’s Bias)
• Healthy worker effect

An example of selection bias:

In a case-control study of smoking and chronic lung disease, the association of exposure with disease will tend to be weaker if controls are selected from a hospital population (because smoking causes many diseases resulting in hospitalization) than if controls are selected from the community. In this example, hospital controls do not represent the prevalence of exposure (smoking) in the community from which cases of chronic lung disease arise. The exposure-disease association has been distorted by selection of hospital controls.

3. Interviewer bias

Interviewer bias is a type of non-sampling error caused by mistakes made by the interviewer. These may include influencing the respondent in some way, asking questions in the wrong order, or using slightly different phrasing (or tone of voice) than other interviewers. It can include intentional errors such as cheating and fraudulent data entry.

When human beings interact with each other in a social setting, they are bound to have a set of preconceived notions about the person they’re interacting with even before they talk. You might like the person even before you’ve had a chance to interact with them because they are good looking or because they went to the same college as you. Maybe you already view them in a negative light as you know they come from a certain part of the country. While these biases are a part of daily social interaction, if you are not conscious of your biases, you might end up in selecting someone on the basis of what pleases you about them or if they have very similar traits or interests as you rather than your research requirement. This is referred to as interview bias.

A few common interview biases that recruiters should keep in mind are listed below:

• Cultural Noise
• Stereotyping Bias
• Generalization Bias
• Halo/Horn Bias
• Recency Bias
• Contrast Effect
• Gender and Racial Bias

An example of interviewer bias:

Interviewer bias is more likely when disease status is known to interviewer. An example of this would be a patient with Buerger's disease enrolled in a case control study which attempts to retrospectively identify risk factors. If the interviewer is aware that the patient has Buerger's disease, he/she may probe for risk factors, such as smoking, more extensively (“Are you sure you've never smoked? Never? Not even once?”) than in control patients.

4. Response bias

Response bias (also called survey bias) is the tendency of a person to answer questions on a survey untruthfully or misleadingly. For example, they may feel pressure to give answers that are socially acceptable. The respondent may not be aware that they aren’t answering the questions in the way the researcher intended: the format of the question or the nature of the previous questions may have an unwanted impact on how a person responds to a survey.

Self-Reporting Issues:

People tend to want to portray themselves in the best light, and this can affect survey responses. According to psychology professor Delroy Paulhus, response bias is a common occurrence in the field of psychology, especially when it comes to self-reporting on:

• Personal traits
• Attitudes, like racism or sexism
• Behavior, like alcohol use or unusual sexual behaviors.

Questionnaire Format Issues:

Misleading questions can cause response bias; the wording of the question may influence the way a person responds. For example, a person may be asked about their satisfaction for a recent online purchase and may be presented with three options: very satisfied, satisfied, and dissatisfied. By being given only one option for dissatisfaction, the consumer may be less inclined to pick that option. In some cases, the entire questionnaire may result in response bias. For example, this study showed that patients who are more satisfied tend to respond to surveys in higher numbers than patients who were dissatisfied. This leads to the overestimation of satisfaction levels.

Other questionnaire format problems include:

• Unfamiliar content: the person may not have the background knowledge to fully understand the question.
• Fatigue: giving a survey when a person is tired or ill may affect their responses.
• Faulty recall: asking a person about an event that happened in the distant past may result in erroneous responses.

Many of the above issues can be averted by providing an opt-out choice like “undecided” or “not sure.”

An example of response bias:

In 1980 a Swedish psychologist asked 81 Americans to rate their driving ability among their peers. Of the 81 people who completed the survey, 93% of respondents said they were better than average. This is statistically improbable.

5. Observation bias

Observation bias is any kind of systematic discrepancy from the truth during the process of observing and recording information for a study by the observer. This is a type of detection bias and can affect assessment in many kinds of study including observational studies and intervention studies such as randomized trials.

It also occurs when the investigator is aware of the disease status, treatment group or outcome of the subject and their ability to interview the subject, collect or analyze the data in an unbiased manner is compromised.

An example of response bias:

One famous example of observer bias is the work of Cyril Burt, a psychologist best known for his work on the heritability of IQ. He thought that children from families with a low socioeconomic status (i.e. working class children) were also more likely to have lower intelligence, compared to children from higher socioeconomic statuses (Fancher, 1985). His “scientific” approach to intelligence testing was revolutionary, and “proved” that children from the working classes were in general, less intelligent. This led to the creation of a two-tier educational system in England in 1960s which sent middle and upper class children to elite schools and working class children to less desirable schools. Burt’s research was later debunked and it was concluded he falsified data. It is now accepted that intelligence is not hereditary.

6. Leading questions and wording bias

Leading questions are questions a researcher asks that may cause a respondent to answer in a biased, particular way. Leading questions are a vital issue of survey and question design, and should be avoided when conducting either quantitative or qualitative survey research. A leading question often aligns with the goals of the researcher, and thus can prime the respondent to unknowingly support the views held by the researcher. Also, including leading questions in survey design can harm both the results and credibility of the research. In addition, using questions that lead respondents can negatively affect objectivity and ethics of both the researcher and the study. Bad questions lead to bad results, and leading questions are among the most common types of bad questions.

It is also a type of question which forces the respondent to answer in a particular way. Often implying or encouraging a certain answer by manipulating the respondent. They are undesirable as they produce inaccurate information. Leading questions can often be answered with a yes or no. (Though not all yes-no questions are leading).

Wording bias is a type of bias where respondents are given questions that influence them in a particular direction.

5 main characteristics that define leading questions:

• They are intentionally framed to cultivate bias in respondents so that the answers are according to the survey creators plan.
• The questions have an element of conjecture and assumption.
• Leading questions thrive on a respondent’s personal input.
• These questions are often asked to understand the consequences of a situation.
• They tend to be forceful in terms of obtaining feedback.

An example of leading questions and wording bias:

Do you have any problems with your boss?

This question prompts the person to question their employment relationship. In a subtle way, it raises the prospect that there are problems.

Tell me about your relationship with your boss?

This question does not seek any judgment and there is less implication that there might be something wrong with the relationship.

7. Sponsor bias

Funding bias refers to when a study’s outcome is more likely to support the interests of the organization funding the study. A correlation has been found by multiple sources between funding source and outcome of research, for example tobacco research and chemical toxicity studies or nutrition studies and processed foods.

When companies pay for research into their products, a conflict of interest can happen. Studies funded by for-profit industries are more likely to yield biased results than those funded by non-profits, including governmental agencies like the National Science Foundation and National Institutes of Health.

Sources of Funding Bias

Funding bias usually happens when a backer has a financial stake in the outcome. It may be due to:

• Testing methods, when paid for by for-profits, may have an inherent bias towards producing positive outcomes for the company.
• Products may have gone through internal testing, weeding out ineffective or dangerous products before they get to the external testing stage.

An example of Sponsor bias:

A Sugar Association funded study by Duke University researchers found numerous adverse health effects from the consumption of Splenda (sucralose). Note that the sponsor for the study had a vested interest in scaring consumers away from artificial sweeteners. Research by many organizations, including the World Health Organization, the FDA and Johnson & Johnson subsidiary McNeil Nutritionals LLC — which sponsored its own research — all indicate sucralose is safe.