Please enable JavaScript.
Coggle requires JavaScript to display documents.
CH 10: The t Test for Two Independent Samples, The estimated standard…
CH 10: The t Test for Two Independent Samples
Definitions & Concepts
Formulas & Calculations
Hypothesis Testing & Interpretation
Applications & Assumptions
The estimated standard error of the mean difference is the denominator of t
Variance from both samples influences the standard error estimate
A pooled variance estimate integrates data from both samples
Larger samples are given more weight by pooled variance
The degrees of freedom are equal to (n₁ − 1) + (n₂ − 1)
The null hypothesis indicates that the population means are equal (μ₁ = μ₂).
The alternative hypothesis indicates the population means are not equal
The t statistic determines if the observed difference is greater than the anticipated sampling error
Two participant groups are compared using an independent-measures design
One sample's participant cannot be included in the second
Sample data is used to predict population differences
Sampling error causes some variation in sample means
Statistic evaluates the difference predicted by chance against the observed mean difference
The null hypothesis is rejected if computed t is greater than critical t
The degrees of freedom and alpha level ---> determine the critical t value
Less precision is shown by wider confidence intervals
The magnitude of an effect is not indicated by statistical significance
Homogeneity of variance assumes equal population variances
Statistical power is increased with a smaller standard error
One popular effect size metric for independent t tests --> Cohen's d
Research involving experiments and group comparisons often use independent sample t tests
Standard error decreases with larger sample sizes
The range of plausible population mean differences is estimated by confidence intervals
The goal is to determine whether two population means differ significantly