
Already know some of this? A 2-minute test skips the lessons you know.
Take the placement testA free 25-lesson course built from numiqo's 4h 55m of YouTube video. 3 days at 90 minutes a day, with study notes and a quiz in every lesson. A 2-minute placement test skips what you already know.
Taught by numiqo on YouTube. Videos play in YouTube's own player, so the creator keeps every view.
Calculate and interpret measures of central tendency (mean, median, mode) and dispersion (standard deviation, variance, range, IQR) to summarize a dataset.
Create and interpret frequency tables, contingency tables, and visualizations (bar charts, histograms, box plots) for categorical and metric variables.
Define null/alternative hypotheses, p-values, significance levels, and Type I/II errors to structure statistical inference.
Classify variables into four measurement levels and determine which statistical analyses and visualizations each level supports.
Select and apply the appropriate t-test (one-sample, independent, paired) to compare means, checking assumptions of normality and equal variance.
Conduct a one-way ANOVA to test for mean differences across three or more independent groups, including post-hoc pairwise comparisons.
Perform a two-way ANOVA to assess main effects of two factors and their interaction effect on a metric dependent variable.
Apply repeated measures ANOVA to detect changes in a dependent variable across three or more time points or conditions for the same subjects.
Combine between-subject and within-subject factors in a mixed-model ANOVA to test main effects and interaction with dependent and independent groups.
Decide between parametric and non-parametric tests based on normality, measurement level, and statistical power considerations.
Evaluate normality using Shapiro-Wilk, Kolmogorov-Smirnov, Q-Q plots, and understand sample-size effects on analytical tests.
Check the homogeneity of variance assumption for t-tests and ANOVA using Levene's test and interpret its p-value.
Perform the Mann-Whitney U test to compare two independent groups when data are not normally distributed, using rank sums instead of means.
Apply the Wilcoxon signed-rank test to compare two dependent samples without assuming normality, using signed ranks of paired differences.
Use the Kruskal-Wallis test to compare three or more independent groups on ordinal or non-normal metric data via rank sums.
Conduct the Friedman test for three or more dependent samples when normality is violated, using within-subject ranking across conditions.
Test for a relationship between two nominal variables using the Chi-Square test of independence with contingency tables.
Measure and test the strength and direction of association between two variables using the appropriate correlation coefficient for the data type.
Fit a simple linear regression model, interpret slope and intercept, check assumptions (linearity, independence, homoscedasticity, normality), and test significance.
Extend regression to multiple predictors, interpret standardized coefficients, detect multicollinearity via VIF, and validate regression assumptions.
Model binary outcomes with logistic regression, interpret coefficients as log-odds, calculate odds ratios, and evaluate model fit and significance.
Model binary outcomes with logistic regression, interpret coefficients as log-odds, calculate odds ratios, and evaluate model fit and significance.
Model binary outcomes with logistic regression, interpret coefficients as log-odds, calculate odds ratios, and evaluate model fit and significance.
Apply k-means clustering to partition data into k groups, iterate centroids to convergence, and select optimal k using the elbow method.
Interpret confidence intervals as long-run coverage probabilities of the estimation method, not as probabilities for a fixed parameter, and compute them for the mean.