646 Obstetrics and Gynecology Board Review •••
❍ What are confounding variables?
Confounding variables are extraneous factors (variables) related to both the outcome of interest and the exposure
of interest, for example, age, sex, tobacco use, and history of tobacco use. Confounding variables can be addressed
through the study design or data analysis.
❍ What is an alpha (α) level of significance?
The alpha (α) level of significance, often set at 0.05, is the chance a researcher is willing to take that there is
actually no significant relationship between variables when the statistical test indicates that there is a relationship.
In other words, an alpha of 0.05 indicates the researcher is willing to take the chance that a “statistically significant”
result is incorrect 5% of the time; 95% of the time the relationship indicated by the statistical test (p-value) would
be correct.
❍ What is a p-value?
A p-value is the probability that the observed difference or relationship between variables occurred by chance
(“luck of the draw”). The higher the p-value, the lower the chance that an observed relationship actually exists. The
researcher sets the p-value less than or equal to the alpha (α) level chosen (see above). Typically, the p-value is set
at ≤0.05. If the researcher wants a lower probability of a statistical relationship occurring by chance, they may set
a significance (α) level ≤0.01 or ≤0.001. A p-value ≤0.001 indicates there is only a 0.1% chance that a statistically
significant relationship occurred by chance.
❍ What is the difference between clinical and statistical significance?
A clinically important finding is a conclusion that has possible implications for patient care. A statistically
significant finding is a conclusion that there is evidence against the null hypothesis.
❍ What is meant by a null hypothesis?
It represents the hypothesis being tested about a population. Null means “no difference,” and refers to a situation in
which no difference exists between variables (eg, no difference between the means in a treatment and control group).
❍ What is bias?
Bias is an error related to the way the targeted and sampled populations differ. Bias threatens the validity of a study.
❍ What is a type I error?
A type I error results if a true null hypothesis is rejected or if a difference is concluded when no actual difference
exists. A type I error is also known as an alpha (α) error.
❍ What is a type II error?
A type II error results if a false null hypothesis is not rejected or if a difference is not detected when a difference
exists. A type II error is also known as a beta (b) error.
❍ What are the two general types of study designs used in medical research?
The two general types of study designs are observational and experimental.
❍ What is an observational study?
An observational study is one that does not involve an intervention or manipulation. There are several types of
observational studies including cross-sectional, cohort, and case-control studies.