Statistics: T or F? ...

True / False?

1. Holding other factors constant, the larger the value of the standard error for the difference between means the wider will be the 95% confidence interval.
2. Holding other factors constant, the larger the sample size the wider will be the 95% confidence interval.
3. Holding other factors constant, the 95% confidence interval will always be wider than the 99% confidence interval.
4. In designing an experiment, an investigator aims for as wide a confidence interval as possible.
5. Using a 95% confidence interval and a t-ratio with a significant level of .05 will always lead to the same decision as to whether the null hypothesis should be rejected or accepted.
6. The half-width of a 95% confidence interval (w) is the minimum difference needed to reject the null hypothesis at the .01 significant level.
7. An experiment using an independent-groups design has 12 in each condition. The degrees of freedom for the t-ratio for this experiment are 23.
8. The null model is a more parsimonious account of the data than the full model when the null hypothesis is false.
9. If the residuals are independently and normally distributed with equal variances, then the normal distribution rather than the t-distribution can be used to calculate a confidence interval for the difference between the means
10. An investigator calculates a 99% confidence interval, finds that it does not include the value of zero and so rejects the null hypothesis. Had the investigator used the data to calculate a 95% confidence interval instead, we can be certain that it too would not have included zero.
11. Increasing the precision of an experiment by increasing the sample size would increase the likelihood of committing a Type-I error.
12. Holding everything else constant, decreasing the Type-I error rate increases the likelihood of committing a Type-2 error.
13. If the null hypothesis is false, you can never commit a Type-2 error, no matter how small the sample.
14. In testing hypotheses about differences between two means, an experimenter mistakenly uses 36 instead of 20 degrees of freedom when looking up the table of the t-distribution to find the critical value. Assuming all other calculations are correct, the effect of this error will be to increase the probability of a Type-1 error.
15. An experimenter sets alpha = .05 as the significance level, obtains a t-statistic less than the critical value and therefore does not reject the null hypothesis. This statement means that the probability of the decision being erroneous is .05.
16. The 95% confidence interval for the difference between two means was found to be 1.2 ± 0.9. This results implies that using a Type-1 error rate of alpha = .05, we can reject the hypothesis that the true difference is zero.
17. In testing hypotheses about differences between two means, using the normal distribution as an approximation to the t-distribution will increase the likelihood of a Type-1 error.
18. The power of a statistical test can be described as the probability of not committing a Type-2 error.
19. In planning an experiment, an investigator first decides to use a Type-1 error rate of alpha = .01 but finally plans on using an alpha = .05. The effect of this change of plan would be to reduce the likelihood of a Type-2 error.