Exploring Two–Variable Data

Study hours serve as a reliable predictor for GPA across all ranges of study time commitment.

The presence of homoscedasticity confirms a direct correlation between study times and GPAs.

Heteroscedasticity indicates that study habits do not contribute at all to predicting GPA accurately.

Variability in study hours may affect GPA prediction accuracy differently across groups.

Two-sample t-test

One-sample t-test

Paired t-test

ANOVA (Analysis of Variance)

The effect cannot be determined without additional information.

It stays the same.

It decreases.

It increases.

Population variance must be known.

The sampling method must be convenience sampling rather than random sampling.

The sample size must be greater than 30.

The population from which the sample is drawn must be approximately normally distributed.

Standard deviation because it measures spread irrespective of symmetry distribution shape.

Mode as it depicts only where values occur most often without symmetry relevance.

Range since it considers all values including potential outliers equally regardless of shape distribution.

Mean or median due to symmetry making them relatively equal.

It reduces variation by accounting for potential confounding variables.

It allows for testing multiple hypotheses simultaneously.

It ensures all experimental units receive each treatment level once.

It eliminates all sources of bias in response variables.

Standard deviation

Range

Mean

Median

Confusing synergistic effects from combined treatments as being separately effective when either drug alone may not surpass placebo effect significantly when tested independently due to inadequate power or sample size limitations inherent in subgroup analyses.

Underestimating natural variation, presuming long-term implications match short-term findings despite possibility chronic conditions vary over time requiring extended monitoring before generalization applies accurately.

Assuming that since all treatments appear successful, side-effects are uniformly non-significant across medication regimens even without specific safety efficacy assessment.

Misinterpreting high statistical significance levels as clinical relevance without taking into account magnitude effect size potentially leading overemphasis smaller practical impacts overshadowed larger picture patient outcomes perspective.

Kruskal-Wallis H test for comparing medians across more than two groups.

ANOVA to test mean differences across multiple groups.

Mann-Whitney U test (Wilcoxon rank sum test) for comparing medians between two unpaired groups.

Pearson correlation coefficient to measure strength and direction of linear relationships.

Event collection

Outcome set

Possibility range

Sample space