Term
|
Definition
| the interval to be constructed that specifies the range of values that we are relatively confidents the population mean falls within; stated in terms of a probability or percentage; the one used most commonly by researchers in behavioral sciences is the 95% confidence interval |
|
|
Term
|
Definition
| define the boundaries of the confidence interval |
|
|
Term
| Confidence Interval when the standard deviation of a sampling distribution is known |
|
Definition
| uses z-scores; uses the standard normal distribution as a model; |
|
|
Term
| Confidence Intervals when the standard deviation of a sample distribution is unknown |
|
Definition
| uses the t-distribution; used also because the standard normal distribution is not always the best model |
|
|
Term
| When is the standard normal distribution a good fit? |
|
Definition
a. when N is sufficiently large
b. when the scores are randomly selected from a population that is normally distributed |
|
|
Term
| When does a t-distribution provide a better model than does the standard normal distribution? |
|
Definition
| when o` is unknown; which is virtually always |
|
|
Term
|
Definition
| the number of independent pieces of information needed to determine a statistic; (N-1) |
|
|
Term
| Two Goals of Inferential Statistics |
|
Definition
1. make unbiased estimates
2. assess error made when making estimates |
|
|
Term
|
Definition
| the standard deviation of a sampling distribution |
|
|
Term
| Standard Error of the Mean |
|
Definition
| the standard deviation of a sampling distribution for an estimate of the mean |
|
|
Term
| How can we calculate unbiased estimators of o`2 and o`? |
|
Definition
| by dividing SS by its degrees of freedom |
|
|
Term
|
Definition
| underestimate; thus it is an unbiased estimator |
|
|
Term
| So how do we get an unbiased estimator of o`? |
|
Definition
| By dividing SS by its degrees of freedom (n-1) |
|
|
Term
| What does the standard error of the mean estimate? |
|
Definition
| It estimates the typical amount of error made when using X` to estimate u |
|
|
Term
| The greater the absolute value of the hypothesis testing statistic... |
|
Definition
| the greater the size of the apparent effect relative to sampling error |
|
|
