CK020 - Statistical Tests for Linear Regression Flashcards
What does the ‘One-sample t-test’ test?
Tests if the mean of a single sample is equal to a known value or population mean.
When do you use the ‘One-sample t-test’ test?
Use when the data is approximately normally distributed, and you want to compare the sample mean to a specific value.
What does the ‘One-sample Wilcoxon signed rank test’ test?
Tests if the median of a single sample is equal to a known value.
When do you use the ‘One-sample Wilcoxon signed rank test’ test?
Use when the data is NOT normally distributed, and you want to compare the sample median to a specific value.
What does the ‘Two-sample t-test (independent)’ test?
Tests if the means of two independent groups are equal.
When do you use the ‘Two-sample t-test (independent)’ test?
Use when comparing the means of two separate groups and the data is approximately normal.
What does the ‘Two-sample Wilcoxon rank sum test’ test?
Tests if the distributions of two independent groups are the same.
When do you use the ‘Two-sample Wilcoxon rank sum test’ test?
Use when comparing two independent groups where the data is NOT normally distributed.
What does the ‘Two-sample t-test (dependent)’ test?
Tests if the means of two related groups are equal.
When do you use the ‘Two-sample t-test (dependent)’ test?
Use when comparing paired data (e.g., before-and-after measurements) that is approximately normal.
What does the ‘Two-sample Wilcoxon signed rank test’ test?
Tests if the medians of two related groups are equal.
When do you use the ‘Two-sample Wilcoxon signed rank test’ test?
Use for paired data when the differences between pairs are NOT normally distributed.
What does the ‘ANOVA test’ test?
Tests if the means of three or more groups are equal.
When do you use the ‘ANOVA test’ test?
Use when comparing means across multiple groups with normally distributed data and equal variances.
What does the ‘M-sample Kruskal-Wallis test’ test?
Tests if the distributions of three or more groups are the same.
When do you use the ‘M-sample Kruskal-Wallis test’ test?
Use when comparing multiple groups where the data is NOT normally distributed or variances are UNEQUAL.
What does the ‘Pearson correlation test’ test?
Tests the strength and direction of a linear relationship between two variables.
When do you use the ‘Pearson correlation test’ test?
Use when assessing linear correlation between two continuous variables with normal distributions.
What does the ‘Spearman correlation test’ test?
Tests the strength and direction of a monotonic relationship between two variables.
When do you use the ‘Spearman correlation test’ test?
Use when assessing the relationship between two continuous or ordinal variables that are NOT normally distributed.
What are the assumptions of the ‘One-sample t-test’ ?
- Dependent variable must be continuous
- Observations are independent
- Dependent variable is approximately normally distributed
- Dependent variable does NOT contain any outliers
What are the assumptions of the ‘One-sample Wilcoxon signed rank test’ ?
- Population distribution is symmetric
- Observations are independent
What are the assumptions of the ‘Two-sample t-test (independent)’ ?
- Dependent variable must be continuous
- Observations are independent
- Dependent variable is approximately normally distributed
- Dependent variable does NOT contain any outliers
What are the assumptions of the ‘Two-sample Wilcoxon rank sum test’ ?
- Population distribution is symmetric
- Observations are independent
What are the assumptions of the ‘Two-sample t-test (dependent)’ ?
- Dependent variable must be continuous
- Observations are dependent
- Distribution of the differences in variables between correlated groups are normally distributed
- Dependent variable does NOT contain any outliers
What are the assumptions of the ‘Two-sample Wilcoxon signed rank test’ ?
- Population distribution of the differences is symmetric
- Observations are dependent
What are the assumptions of the ‘ANOVA test’ ?
- Variables are continuous
- Samples are independent
- Homogeneity of variance
- Data are normally distributed
What are the assumptions of the ‘M-sample Kruskal-Wallis test’ ?
Within & between group observations are independent of eachother
What are the assumptions of the ‘Pearson correlation test’ ?
- Variables must be contiuous
- There is a linear relationship between the 2 variables
- Data has homoscedasticity
- Variables are approximately normally distributed
- Variables represent paired observations
- Variables do NOT contain any outliers
What are the assumptions of the ‘Spearman correlation test’ ?
- Variables must be continuous
- There is a monotonic relationship between the 2 variables
- Variables represent paired observations
