One samples t test
The sum of the squared deviations of each raw score from the sample mean (otherwise known as the sum of squares, or SS) is the numerator of the formula for estimated variance.
True
If you had a sample size of n = 57 and you were going to perform a one-sample t test using alpha = .05, you would have to choose between the critical values at df = 40 or df = 60. You SHOULD use df = 40 because
it is more conservative than using 60
If your treatment does not have an effect,
large values of t-obtained are unlikely to occur.
If your treatment has no effect at all,
the sample mean (X-bar or M) will be exactly the same as μ more frequently than it is any other number, making it your best guess as to what μ is (and therefore what is called the expected value of μ). 2.sampling error will usually keep the sample mean (X-bar) from being exactly the same as μ
The correlation (r) between depression and anxiety in one of my studies was r = .90. What proportion of variance in depression scores is explained by one's level of anxiety?
.81
Which of the following statements accurately describes the difference between a one-sample t test and a z test?
1.You must calculate the estimated variance (or standard deviation) for the t test but not for the z test. 2.You must know the population variance (or standard deviation) for the z test but not for the t test. 3.You must know the degrees of freedom to find the critical t but not the critical z value.
f you had a sample size of n = 38, and you were going to perform a one-sample t test using alpha = .05, you would have to choose between the critical values at df = 30 or df = 40. You SHOULD use df =
30 because it decreases the likelihood of making a Type I error.
In a hypothesis test using a t statistic, what is the influence of using a large sample?
A larger sample tends to increase the likelihood of rejecting the null hypothesis.
As sample size increases ____
All of the other options are true as sample size increases.
Standard deviation equals variance squared.
False
Which of the following assumptions must be met in order to use the one-sample t test?
The population of dependent scores, which must be interval or ratio scores, forms a normal distribution.
Estimated standard deviation (lowercase s) can have a subscript of X or not have it; it means the same thing with or without the X.
True
You have a sample size of n = 20.
You will perform a one-tailed t test with alpha = .05. You have looked up the critical value and it is -1.73. You have calculated t-obtained and it equals -1.41. Which of the following is correct to do?
Which of the following will give you a larger value for t-obtained?
increasing the degrees of freedom in your experiment
When using the sample variance to estimate the population variance, we divide the numerator by
n - 1 because dividing by n would underestimate the population variance
If a researcher conducts a one-sample t test and reports a t statistic with df = 20, how many individuals were in the sample?
n = 21
A sample is selected from a population with µ = 80. If the sample has a mean of 85, which of the following will produce the largest value for t-obtained?
n = 25 and s2 = 4
The proportion of variance in the dependent scores that is explained by the IV equals
r-squared
The p value indicates ____, and effect size provides information about _____.
statistical significance; clinical significance
You have a sample size of n = 12. You will perform a one-tailed t test with alpha = .05. You have looked up the critical value and it is -1.34. You have calculated t-obtained and it equals -1.41. Which of the following statistical results is written correctly?
t(11) = -1.41, p < .05
For BOTH the z test and the one-sample t test, we must know
μ under a condition of the IV other than the one in our sample
A population is known to have a mean of μ = 150. A treatment is expected to increase scores for individuals in this population. If the treatment is evaluated using a one-tailed hypothesis, then the NULL hypothesis would state __________.
μ ≤150