(the symbol
is
often used instead of just a lower case s. This helps to avoid confusion
between the lower case s and the upper case S.)
Note:
(the symbol is
often used instead of just a lower case s. This helps to avoid confusion
between the lower case s and the upper case S.)
Distinguish between a between subjects design and a within subjects design. Recognize examples of each.
Recognize that the data from a between subjects design are treated like the data from two independent samples.
Recognize that the data from a within subjects design are treated like the data from two related samples.
Explain the implications of taking a convenience sample, rather than a random sample, for an experiment.
Explain the importance, in a between subjects experiment with two conditions, of randomly asssigning the subjects to the two conditions.
Explain the implications of sampling error for an experiment.
Recognize that the t test for two independent samples is sometimes simply called the two sample t test.
Recognize that the t test for two related samples is sometimes called the paired t test.
Define the sampling distribution of the difference between means (for independent or related samples), and describe its shape, mean, and standard deviation.
Define the standard error of the difference between means.
Explain how the standard error of the difference between means of related samples differs from that for independent samples.
In the test for independent samples, define the pooled variance estimate ( sp2 ) (what is it an estimate of?).
Determine the degrees of freedom for the two independent samples t test and the two related samples t test.
Calculate the pooled variance estimate either from raw scores or given the
unbiased standard deviations (
1 and
2 ) and sizes of both samples.
Calculate the estimated standard error of the difference between means for independent samples.
Calculate the estimated standard error of the mean difference for related samples, either from raw scores or given the unbiased standard deviation of the difference scores and sample size.
Conduct one-tailed and two-tailed tests of hypotheses about differences between two population means based on data from independent or related samples. Recognize correct and incorrect interpretations of the results and conclusions.
Recognize that the direction (upper or lower) of a one tailed test of the difference between two means can be arbitrarily determined by either subtracting the smaller mean from the larger mean (upper one tailed) or subtracting the larger mean from the smaller mean (lower one tailed), but that it is usually done as an upper one tailed test. Therefore, determining whether a treatment increased or decreased the dependent variable must be determined by looking at which mean is larger and which is smaller, not by considering the direction of the test.
Calculate confidence intervals for the difference between the means of two populations, based on data from independent or related samples. Recognize correct and incorrect conclusions and interpretations.
Recognize the relationship between a confidence interval for the difference between two means that either does or does not contain 0 (when the null hypothesis is that the difference is 0 ) and the significance of the associated hypothesis test.
State the assumptions underlying the t test for the significance of a difference between means (for independent and related samples). Describe effects of violating the assumptions.
Describe factors affecting the power of the t test. Given relevant information, decide which of two (or more) tests is more powerful.
Describe advantages and disadvantages of between subjects vs. within subjects designs.
Given some raw data appropriate for a two sample t test, with a description of how the data was collected and a research question it could answer, use SPSS to conduct the appropriate two sample t test, interpret the results correctly, and report the results.