Hypothesis Testing Principles and Methods
Foundations of Hypothesis Testing
Purpose: Evaluates sample data to assess the relative plausibility of competing claims regarding unknown population parameters, determining whether observed differences reflect true effects or random chance.
Null Hypothesis (): Represents the default claim of no effect, no difference, or status quo; always contains an equality condition (e.g., , ) and is presumed true until sufficient sample evidence indicates otherwise.
Alternative Hypothesis (): Represents the operational research question asserting an effect, difference, or parameter shift.
Two-Sided (Two-Tailed): Detects any non-directional difference ().
Right-Tailed (Upper-Tailed): Detects an increase or value higher than the null parameter ().
Left-Tailed (Lower-Tailed): Detects a decrease or value lower than the null parameter ().
Hypothesis Testing Procedure

Step 1: State the Hypotheses: Specify and , identifying directionality (left-, right-, or two-tailed).
Step 2: Select Significance Level (): Predetermine the acceptable probability of committing a Type I error (standard values: , ).
Step 3: Calculate Test Statistic: Standardize sample findings relative to null expectations using or .
Step 4: Determine the -Value or Rejection Region:
Rejection Region (Critical Value) Method: Compare the test statistic directly to critical threshold values established by and sample size.
-Value Method: Calculate the probability of obtaining sample results as extreme as, or more extreme than, those observed given that is true.
Step 5: Make a Statistical Decision:
If the test statistic falls into the critical rejection region (or ), reject .
If the test statistic falls in the region of no significance (or ), fail to reject . Correct terminology requires stating "fail to reject " or "do not reject "; never conclude "accept " or that is proven.
Step 6: State Conclusion in Context: Express findings in relation to the original research scenario, explicitly reporting the test statistic, , and -value.
Decision Errors and Statistical Power

Type I Error (): Occurs when a true null hypothesis is rejected (false-positive conclusion). The probability is set by the investigator prior to sampling: .
Type II Error (): Occurs when a false null hypothesis is not rejected (false-negative conclusion): .
Statistical Power (): The probability of correctly rejecting a false null hypothesis, confirming the validity of a true alternative hypothesis.
Error Trade-off: Lowering increases , and lowering increases .
-Value Interpretation Guidelines:
: Results not statistically significant.
: Trend toward statistical significance.
: Statistically significant.
: Highly significant.
: Very highly significant.
One-Sample Tests for Continuous Outcomes
Large Sample () — -Test:
Formula:
Two-sided critical value at is .
Framingham Heart Study Total Cholesterol Example: , , , , . Because , reject (); participants differ significantly from the national average.
Small Sample () — -Test:
Formula:
Degrees of freedom: .
Diabetic Diastolic BP Example: , , , , , . Two-sided critical value at is . Because , reject ; mean diastolic BP differs significantly from the general population.
One-Sample Tests for Dichotomous Outcomes
Sample Proportion: where is positive occurrences and is total subjects.
Test Statistic (-Test for Proportions):
Directional Critical Values ():
Lower-tailed: .
Upper-tailed: .
Framingham Heart Study Smoking Example (Lower-Tailed):
, , , , , .
Calculated test statistic: .
Decision: ; reject (). Smoking prevalence is significantly lower.
Newborn Gender Example (Upper-Tailed):
, , , , , .
Calculated test statistic: .
Decision: ; reject . Male birth prevalence is significantly higher than female birth prevalence.
Two Matched Samples for Continuous Outcomes
Design: Paired pre- and post-intervention evaluations on single subjects ( pairs), yielding mean difference and standard deviation of difference .
Hypotheses: versus (or directional: , ).
Test Statistic:
For :
For ():
Framingham Offspring Study Systolic BP Example:
, , , , , , .
Critical values: .
Calculated statistic:
Decision: ; fail to reject . There is no statistically significant difference in systolic blood pressures over time.