Bias in surveys refers to systematic deviations from the proper population parameters that occur consistently in the same direction (Li, Higgins & Deeks, 2019). These biases can affect the accuracy and representativeness of survey results, leading to incorrect or misleading conclusions.
Types of bias:
- Sampling bias: occurs when the sample of respondents is not representative of the population as a whole. For example, if you only survey people in a particular city, my results will not be generalizable to people in other cities (Ketokivi, 2019).
- Non-response bias occurs when some people selected to participate in a survey do not respond. This can be due to various reasons, such as being too busy, not being interested in the topic, or not being able to understand the questions.
- Response bias: occurs when respondents answer questions in a way that is not truthful or accurate. This can be due to various reasons, such as wanting to please the interviewer, trying to appear in a certain way, or simply not understanding the question.
The bias in my research:
In my research, I will use mixed Quantitative and Qualitative data to analyze the problem of construction pollution. I will use mobile interviews, video calls, and online surveys sent via Google Forms during the data collection process. There are several ways to reduce bias during this process, such as:
1. Using a standardized interview protocol: means using the same questions and procedures for all interviews, regardless of the respondent. This helps to ensure that all respondents are treated equally and that the interviewer's personal biases do not influence their responses (Simundic, 2013).
2. Using a random sample of respondents: means randomly selecting respondents from the population. This helps to ensure that the sample is representative of the population as a whole and that the study results are not biased by any particular group or demographic (Quantitative, Qualitative, and Mixed Research, 2020).
3. Using various data collection methods: means using a combination of telephone or video interviews, online surveys, and other data collection methods. This helps reduce the risk of bias by ensuring the data is collected from various sources.
4. Being aware of the potential for bias: means being aware of your personal biases and how they might influence your study. It also means being open to feedback from others and being willing to change your research design if necessary (Nie, Tian, Taylor& Zou, 2018).
Conclusion
During the sampling process, bias sometimes occurs in data collection. Systematic errors in the sample survey procedures mainly cause this bias. So it can be seen in the sampling planning process (Li, Higgins & Deeks, 2019). For example, in my research, bias occurs if a specific group is targeted to answer the questions. This group is known to the researcher. Its answer will be in the direction of proving the research question and its results.
References
Ketokivi, M. (2019). Avoiding bias and fallacy in survey research: A behavioral multilevel approach. Journal of Operations Management, 65(4), 380-402.
Li, T., Higgins, J. P., & Deeks, J. J. (2019). Collecting data. Cochrane handbook for systematic reviews of interventions, 109-141.
Simundic, A. M. (2013). Bias in research. Biochemia medica, 23(1), 12-15.
Nie, X., Tian, X., Taylor, J., & Zou, J. (2018, March). Why adaptively collected data have negative bias and how to correct for it. In International Conference
on Artificial Intelligence and Statistics (pp. 1261-1269). PMLR.
Quantitative, Qualitative and Mixed Research (2020). Retrieved from https://www.sagepub.com/sites/default/files/upm-binaries/38123_Chapter2.pdf