In research, bias can be defined as the process of introducing a systematic error into the sampling and thus encouraging one outcome over the other, (Arnold, 2010). This can be attributed to many reasons, for instance due to experimental error, where the researcher fails to take into account all the research variables. There are many types of biases and all these depend on the research design employed. The following are the possible research biases for both qualitative and quantitative research.
Bias in Qualitative Research
In qualitative research, bias is defined in terms of the validity and reliability of the research findings, (Polit & Beck, 2012). If the findings are not reliable and valid, then they shall be termed as biased. The main disadvantage of bias is that it leads to distortion of truth in addition to producing skewed data.
Bias from Moderator
The moderator is the individual who is responsible for collecting the research data. The moderator can be a source of bias based on his/her facial expressions, mode of dressing, tonal expressions and style of language among other factors. Although this type of bias is hard to eliminate, it can be minimized through maintenance of neutrality in tonal expression, mode of dressing and language.
The way in which questions are asked can influence the answers given by respondents. There is need for the interviewer to determine biased questions and rephrase them. A question forms the main basis in which information is collected and this can lead invalid findings if they are biased. The following are the major ways in which questions can lead to biased research findings.
The most common form of bias in asking of questions is the tendency of the field interviewer to ask questions that suggests possible answer, (Green, 2013). Leading questions give slanted answers from respondents. An example of a leading question can take the form of, “Doctors have found that sugars are actually responsible for excess fat in our bodies. What do you think?”
Misunderstood/ Unanswerable Questions
Another form of bias with regard to asking of questions is asking the respondent a question, which he/her cannot understand. The respondent will be forced give his/her answer based on the perceived understanding and this can give biased information.
Biased answers arise from statements that are generally untrue or partially true. The common occasions where such answers are obtained are when conduction interviews on focus groups. The presence of dominant respondents in focus groups may influence other respondents and this may create skewed answers. Additionally, another source of bias can arise from inconsistent answers especially in questions where one leads to the other question.
Sample refers to subgroup of the target population where research is conducted. If the sample is not screened well, one may interview wrong people. Interviewing respondents who do not form the subgroup constitutes bias. Another form of bias in sample selection is failure to use random sampling in the selection of the sample. Random sampling ensures that the study samples have equal chance of selection.
Sometimes, bias can arise from reporting of the results. This source of bias may arise because of personal beliefs, customs, attitude, culture and errors among many other factors. If the person reporting analyses the research information based on his/her beliefs other than the view perceived by the respondents, the findings shall be compromised and hence biased.
Bias in Quantitative Research
Whereas in qualitative research an effort is made to understand the source of bias, in quantitative research, the researcher tries to eliminate bias.
Design bias encompasses bias that may arise when conducting the experiments, as well as during analysis of the results. Design biases are always common, mostly due to the failure of the researchers to take into account the likely impact of the bias in the research they conduct.
Sampling biases occurs in quantitative research when a researcher compromises with the selection of the study subgroup. For instance, a researcher may decide to omit a certain group in the study sample, or include only a specific group. For example, a study on breast cancer that includes only male participants is said to be biased and its results cannot be extrapolated to cover entire study population that include females. Similarly, when a study is done outside a recreation centre on students in a psychological study is biased since it is not inclusive. The major source of sampling bias occurs in systematic and random sampling.
Random and Systematic Bias
Random and systematic sampling can be a source of bias if the researcher fails to select a representative sample. For, example, if a research is done on a population of around 10,000 students and the researcher takes a sample size of 40 students, then the research would be deemed biased since it is not representative. Additionally, if on selection process, the researcher picks specific group of students then it would give biased results since the chosen study sample should always possess an equal chance of being studied and this should be randomized.
ORDER UNIQUE ANSWER NOW