This is part 3 of a 4 part series, covering bias in data collection: what bias is, who data bias can affect, the importance of awareness of data bias, and ways in which we (as analysts and consultants) can attempt to mitigate bias in the collection and analysis phases.
Affected Groups of Biased Data Collection
Bias in data collection can affect different groups in different ways, depending on the nature of the bias and the characteristics of the groups in question. However, certain groups are more likely to be impacted by bias than others. Here are some examples:
Marginalized groups: People of colour, people with disabilities, and members of the LGBTQ+ community are more likely to be underrepresented in datasets and studies. Leading to biased results that do not accurately reflect the experiences or needs of these groups. For example, imagine a study that seeks to understand the prevalence of a certain health condition among different racial and ethnic groups. If the study only includes a small number of participants from certain racial or ethnic groups, the results may not accurately reflect the actual prevalence of the condition in those groups. This can lead to flawed decision-making in areas such as public health policy or resource allocation.
Women: Women are often underrepresented in studies, particularly in fields such as science, technology, engineering, and mathematics (STEM). This can lead to biased results that do not accurately reflect the experiences or needs of women.
Low-income individuals: Low-income individuals may be more difficult to reach and include in studies, particularly if the studies are conducted online or require access to certain technology. This can lead to biased results that do not accurately reflect the experiences or needs of low-income individuals. In addition, bias in data collection can impact the ability of different groups to access resources or services. For example, if a study only includes participants who have access to certain technologies or resources, it may not accurately reflect the experiences or needs of individuals who do not have access to those resources. This can lead to flawed decision-making in areas such as the design of programs to support individuals in low-income or rural areas.
Rural populations: Rural populations, similarly to the above, may be underrepresented in studies, particularly if the studies are conducted in urban areas or require access to certain resources or technologies.
Elderly populations: Elderly populations may be underrepresented in studies, particularly if the studies are conducted online or require access to certain technologies.
So now we understand who data bias can affect, and the importance of mitigating data bias, it is now important to explore the ways in which we can adjust for data bias which can be read here.