Read: Goodman, S.N., Schneeweiss, S. and Baiocchi, M. (2017).Using design thinking to differentiate useful from misleading evidence in observational research(Links to an external site.).JAMA. February...

Research Studies        2
RESEARCH STUDIES
Table of Contents
Observational Biases    3
Design Thinking and Randomized Clinical Research    5
References    7
Observational Biases
    Bias in a research study can be defined as a type of error that can manipulate research results in a particular direction. In a research study, there is a multitude of observational biases. This article will define the biases one by one.
    Selection bias can be defined as the kind of error which occurs when the researcher decides the participants of the research or the population that will be studied with the help of the research (Goodman, Schneeweiss & Baiocchi, 2017). This kind of bias is especially common in research studies that do not implement randomized populations.
    Immortal time bias can be described as the period of follow-up, where there is a problem between the follow-up time and the treatment time in the cohort studies. Immortal time happens when the participants in the cohort study fail to experience the outcome for some period of follow-up time.
    Reverse causation bias can be described as the mistake in research studies where there is confusion between the causality of a particular study (Goodman, Schneeweiss & Baiocchi, 2017). That is to say when researchers believe that X causes Y, however, the reality is that Y causes X.
    Adjustment for immediate variables can be described as a control for a certain variable which increases the net bias or leads to the decrement of precision without affecting bias.
    Right model choice or choice supportive bias can be described as the tendency of the researchers to retroactively assign positive attributes to a certain option that one has selected, while simultaneously demoting the forgone options and choices.
    Survivorship bias can be described as the type of bias arising out of sample selection which tends to occur when a particular data set considers the observations that are existing or ‘surviving’ and fails to take into consideration the observations that have stopped to exist (Goodman, Schneeweiss & Baiocchi, 2017).
    The observer expectancy bias can be described as a certain form of reactivity where the researcher's cognitive bias leads the participants of the same experiment to be subsequently influenced by the same bias in a subconscious manner.
    Interviewer bias can be described as the way in which interviewers ask questions or respond to answers. Interviewer bias can lead to differences in the expected research outcomes, which will be...