Inferential Statistics
- Inferential Statistics – Definition, Types, Examples, Formulas
- Observational Studies and Experiments
- Sample and Population
- Sampling Bias
- Sampling Methods
- Research Study Design
- Population Distribution, Sample Distribution and Sampling Distribution
- Central Limit Theorem
- Point Estimates
- Confidence Intervals
- Introduction to Bootstrapping
- Bootstrap Confidence Interval
- Paired Samples
- Impact of Sample Size on Confidence Intervals
- Introduction to Hypothesis Testing
- Writing Hypotheses
- Hypotheses Test Examples
- Randomization Procedures
- p-values
- Type I and Type II Errors
- P-value Significance Level
- Issues with Multiple Testing
- Confidence Intervals and Hypothesis Testing
- Inference for One Sample
- Inference for Two Samples
- One-Way ANOVA
- Two-Way ANOVA
- Chi-Square Tests
Confounding Variables
Confounding Variables
Randomized experiments are preferred over non-randomized studies because they offer greater control, reducing the likelihood of confounding variables. Confounding variables are factors that can affect the outcome being studied, and are related to both the independent and dependent variables.
Confounding variables are factors that are not the main independent variable of interest in a study, but can affect the outcome being studied. These variables may be related to both the independent variable and the outcome variable, which can make it difficult to determine whether the observed effect is due to the independent variable or the confounding variable.
In other words, confounding variables are extraneous variables that may be related to both the independent and dependent variables, which can create the appearance of a relationship between them when in fact no such relationship exists.
Controlling for confounding variables is important in research to ensure that the relationship between the independent and dependent variables is accurately measured. One way to control for confounding variables is to design the study in a way that allows for their effects to be measured and accounted for, such as through randomization or statistical analysis.
Example 1: Medication on blood pressure
For example, let’s say a study is being conducted to investigate the effect of a new medication on blood pressure. The independent variable is the medication, and the outcome variable is blood pressure. However, there may be other factors that could affect blood pressure, such as age, gender, diet, exercise, and other health conditions. If these factors are not controlled for in the study, they could be confounding variables that affect the results.
Example 2: Exercise and weight loss
For example, let’s say a researcher wants to study the relationship between exercise and weight loss. They randomly assign half of the participants to an exercise program and the other half to a control group. After several weeks, the exercise group shows a significant decrease in weight compared to the control group.
However, the researcher failed to account for the fact that the exercise group also had a significantly higher intake of protein compared to the control group, which could have been a confounding variable. The higher protein intake could have also contributed to weight loss, and not just the exercise program.
Therefore, the observed relationship between exercise and weight loss may be partially or entirely due to the confounding variable of protein intake. To minimize the impact of confounding variables, researchers can use various statistical techniques such as randomization, matching, and regression analysis to control for their effects.