click below
click below
Normal Size Small Size show me how
PSY2073
| Question | Answer |
|---|---|
| When do you use independent groups t-test? | Typically for random assignment, with clear distinction between groups and no overlap |
| When do you use related samples t-test? | Participants are their own controls (before-after) or when individuals are closely-matched between groups |
| When do you use one-sample t-test? | When one sample is being compared to the population (requires population mean) |
| What is the advantage of related samples t-test? | Related samples t-tests have more power than independent groups t-tests, because they require less people to reject the null. People serve as their own control. |
| What are the disadvantages of related samples t-test? | Carry-over effects where behavior may be influenced by previous conditions or tests. Subject-expectancy effects where behavior may be change due to learning about the experiment. |
| How does your choice of t-test affect sample size requirements? | Related samples t-test has more power than independent samples t-test |
| What are the assumptions of an independent groups t-test? | Independent groups, normal distribution, and homogeneity of variance |
| What are the assumptions of a related samples t-test? | Independent observations and normal distribution |
| How does increasing N affect power? | Increases t obtained and power |
| How does increasing standard deviation affect power? | Decreases t obtained and power |
| How does increasing the difference between means affect power? | Increases t obtained and power |
| How does t-distribution differ from normal distribution? | T-distribution gets fatter and flatter as degrees of freedom get smaller |
| What is an important assumption in the validity of a t-test? | A normal distribution with a small sample size |
| What is the first variable to find in a related samples t-test? | Sum of differences and sum of differences squared |
| What is the null hypothesis of a related samples t-test? | Mu diff = 0 |
| What is the first variable to find in an independence group t-test? | Sum of x's and sum of x's squared |
| What is the null hypothesis of an independent groups t-test? | Mu experimental - Mu control = 0 |
| What is the degrees of freedom formula for an independent groups t-test? | n1+n2-2 |
| What variable is required for a z-test? | Population standard deviation |
| What can be increased to increase power? | Difference between means, sample size, or alpha |
| What can be decreased to increase power? | Standard deviation |
| What are the r squared effect size estimates? | .01 (small), .09 (medium), .25 (large) |
| What is the formula for r^2? | t obtained ^2 / (t obtained ^2 + df) |