Dependent-Sample t-Tests Cheat Sheet
Dependent-sample t-tests compare means from the same group at different times or under different conditions to determine if a statistically significant difference exists, crucial for within-subjects research designs.
Core Principles
- Hypotheses: Always state a null (H0) and an alternative (H1) hypothesis to frame the test.
- Critical Region: Define the rejection zone based on alpha level and degrees of freedom to decide if results are significant.
- Test Statistic (t_obs): Calculate the observed t-value to measure the difference between sample means relative to variability.
- Decision Rule: Compare t_obs to the critical t-value; if t_obs falls in the critical region, reject H0.
- Effect Size: Quantify the magnitude of the difference (Cohen's d, r^2) beyond statistical significance.
- Confidence Intervals: Estimate the range within which the true population mean difference likely lies.
Action Steps
- Step 1: State Hypotheses (H0: no difference, H1: difference exists) to set up the statistical test.
- Step 2: Determine Critical Region (df, t_critical) to establish the threshold for statistical significance.
- Step 3: Calculate Test Statistic (M_D, s_MD, t_obs) to quantify the observed effect in the sample.
- Step 4: Make a Decision (Reject H0 or Fail to Reject H0) based on whether t_obs falls within the critical region.
- Step 5: State Conclusions in sentence form, interpreting the decision in the context of the research scenario.
- Step 6: Calculate Effect Size (Cohen's d, r^2) to understand the practical significance of the findings.
- Step 7: Calculate Confidence Interval (Lower Limit, Upper Limit) to estimate the range of the true population mean difference.
Key Terms
- Dependent-Sample t-Test: A statistical test used to compare the means of two related groups, typically from the same sample measured twice (e.g., before and after an intervention).
- Within-Subjects Design: An experimental design where the same participants are exposed to all conditions or are measured at all time points, leading to dependent samples.
- M_D: The mean of the differences between paired scores in the sample.
- s_MD: The standard deviation of the differences between paired scores in the sample.
- t_obs: The observed t-statistic calculated from sample data, indicating the size of the difference relative to variability.
- Cohen's d: A measure of effect size that quantifies the difference between two means in standard deviation units.
- r^2: A measure of effect size representing the proportion of variance in the dependent variable that is accounted for by the independent variable.
- Confidence Interval (CI): A range of values, derived from sample statistics, that is likely to contain the true value of an unknown population parameter (e.g., the mean difference).
Real World Examples
- A clinician studies ketamine's effect on depression by measuring anhedonia scores in 12 individuals before and after ketamine injection.: A dependent-sample t-test is used to determine if ketamine significantly reduces anhedonia scores (pleasure seeking).
- A researcher uses provided data (MD = 3.5, n = 30, SS = 1443) to conduct a dependent-sample t-test with alpha = 0.05, two-tailed.: Calculate t_obs, critical values, and a 95% CI to assess the significance and range of the mean difference.
- Another dataset (MD = 43, n = 16, s = 116.20) requires a dependent-sample t-test with alpha = 0.05, one-tailed, positive difference.: Determine if there is a significant positive difference and calculate the 95% CI for the mean difference.
Related cheat sheets
More like this