China
Africa
Explore chapters and articles related to this topic
Basic Research Design:
Published in Lynne M. Bianchi, Research during Medical Residency, 2022
Lynne M. Bianchi, Luke J. Rosielle, Justin Puller, Kristin Juhasz
Although you choose your independent and dependent variables, there are usually other variables that influence your outcomes. A confounding variable, or confounder, is an extraneous variable that interacts with the independent variable or affects the outcome (dependent variable). An often-cited example is the erroneous association of lung cancer, the primary outcome measure, with coffee drinking. Based on data it appears that coffee drinkers have higher rates of lung cancer than non-coffee drinkers. However, most coffee drinkers who had lung cancer also smoked cigarettes. The two habits tended to go together; the individuals often smoked while drinking coffee. In this example, coffee drinking is a confounding variable. Smoking cigarettes is the independent variable (exposure) associated with lung cancer (outcome).
Model Estimation and Evaluation
Published in Douglas D. Gunzler, Adam T. Perzynski, Adam C. Carle, Structural Equation Modeling for Health and Medicine, 2021
Douglas D. Gunzler, Adam T. Perzynski, Adam C. Carle
A confounder is an extraneous variable whose presence affects the relationship between variables being studied [39]. A confounder is associated with both an independent variable and an outcome being examined. The results may not reflect the actual relationship between variables being studied if confounders of that relationship are not controlled for in the statistical model. For example, consider a study of the relationship between alcohol consumption and lung cancer. Smoking is a confounder of this relationship since smoking is associated with alcohol consumption and a cause of lung cancer. After accounting for smoking, there is no significant association between alcohol consumption and lung cancer [40].
Clinical Development Plan and Clinical Trial Design
Published in Mark Chang, John Balser, Jim Roach, Robin Bliss, Innovative Strategies, Statistical Solutions and Simulations for Modern Clinical Trials, 2019
Mark Chang, John Balser, Jim Roach, Robin Bliss
In statistics, a confounding factor (also confounder, hidden variable, or lurking variable) is an extraneous variable that correlates, positively or negatively, with both the dependent variable and the independent variable. Such a relationship between two observed variables is termed a spurious relationship.
Procedural Integrity Reporting in the Journal of Organizational Behavior Management (2000–2020)
Published in Journal of Organizational Behavior Management, 2022
Daniel Cymbal, David A. Wilder, Nelmar Cruz, Grant Ingraham, Mary Llinas, Ronald Clark, Marissa Kamlowsky
One of the hallmarks of applied behavior analysis (ABA) in general and organizational behavior management (OBM) in particular is the conduct of socially significant research (Baer et al., 1968). The existence of this research enables practitioners to recommend and implement interventions with confidence. Thus, empirical studies published in behavior analytic journals must be sound. That is, the dependent variables must be measured accurately, the independent variables must be clearly described, and the conclusions derived from the results of the study must be based on the observed functional relations between these variables. These features strengthen the internal validity of a study, or the extent to which the independent variable, and not an extraneous variable, is responsible for changes in the dependent variable (Johnston & Pennypacker, 2020, p. 143).
Powerlessness in the moral self: a social cognitive perspective on drug users
Published in Journal of Social Work Practice in the Addictions, 2021
The data were analyzed by IBM SPSS Statistics version 25. Correlations were tested to see if drug users’ drug abuse severity is a plausible extraneous variable that influences research variables (i.e., internalization, symbolization, and moral judgment on drug use). For the tests of research hypotheses, the study first compared the three research variables between drug users and non-drug users. A Mann-Whitney U test was used to compare moral identity variables (internalization and symbolization) between groups (Hypothesis 1). An independent samples t-test was conducted to compare moral judgment on drug use between groups (Hypothesis 2). Lastly, the study carried out multiple regression analyses to compare the relationship between moral identity and moral judgment on drug use between groups (Hypothesis 3).
A Bibliometric Analysis of the Quantitative Mental Health Literature in Occupational Therapy
Published in Occupational Therapy in Mental Health, 2018
Twenty-one (30.88%) articles of the 68 were studies in which an intervention was assessed using a one-group pre-test–post-test design. These types of research designs are common in clinical settings in which participants are selected from a convenience sample and a control group is not feasible. One-group pre-test–post-test designs are categorized as Level III evidence in AOTA’s Evidence Level Scale because they lack both internal and external validity—that is, the findings from these types of studies cannot be generalized to the larger population and, without a control group, it is difficult to determine if the intervention, or some extraneous variable, caused observed changes in the dependent variable (i.e., the outcome measure). One-group pre-test–post-test designs; however, are often used as pilot studies to determine if larger, more rigorous studies are possible. These designs can also be more easily implemented in real-life clinical settings where it is often not possible to recruit large sample sizes and/or randomly assign participants to intervention and control groups (Portney & Watkins, 2015).