Term
| .05 level of significance (p.147) |
|
Definition
| The minimum standard. if we decide to reject H0 we want to to commit a type 1 error, this threshold frames the inferential fate of H0 and Ha: If the answer is "more than 5 out of 100" we do not reject the null hypothesis. |
|
|
Term
| asymetric measure of association(p. 160) |
|
Definition
| by contrast, models the independent variables as the casual variable and the dependent variable as the effect. Because these measures are better suited to the enterprise of testing a casual hypthesis, they are preffered to symmetric measures, which are agnostic on the question of cause and effect. |
|
|
Term
| chi-square test of significance (p.154) |
|
Definition
| determines whether the observed dispersal of cases depart significantly from what we would expect to find if the null hypothesis were correct. |
|
|
Term
|
Definition
| a pair of observations that is consistent with a negative relationship. |
|
|
Term
| confidence interval approach (p.149) |
|
Definition
| the investigator uses the standard error to determine the smallest plausible mean of difference in the population. |
|
|
Term
|
Definition
| based on chi-squared, takes a value of between 0, no relatiohship, and 1, a perfect realtionship. used instead of lambda. |
|
|
Term
|
Definition
| marks the upper plausible boundry for random error and so defines H0's limit. |
|
|
Term
|
Definition
| a pair of observations that is consistent with a negative relationship. |
|
|
Term
|
Definition
| is designed to measure the strength of the relationship between two catagorial variables, at least one of which is nominal-level. |
|
|
Term
| measure of association(p.146) |
|
Definition
| tells the researcher how well the independent variable works in explaining the dependent variable |
|
|
Term
|
Definition
| The skeptical assumption. Plays a vital role in hypothesis testing. States that in the population there is no relationship between the independent variable and the dependent variable. |
|
|
Term
| one-tailed test of significance (p.150) |
|
Definition
| we do not divide .05 in two and find upper and lower confidence boundries. Rather, we place the entire rejection region in null hypothesis territory, the region of the curve containing a population difference equal to 0. |
|
|
Term
proportional reduction in error (p.159)
(PRE) |
|
Definition
| is a prediction-based metric that varies in madnitude between 0 and 1. The precise value of the measure tells you how much better you can predict the dependent variable by knowing the independent variable than by not knowing the independent variable. |
|
|
Term
|
Definition
| the researcher determines the exact probability of obtaining the observed sample differenc, under the assumption that the null hypothesis is correct. if the probability value is less than or equal to .05 the the null hypothesis can be rejected. |
|
|
Term
|
Definition
| is appropriate for gauging the strength of ordinal-level relationships. |
|
|
Term
| standard error of difference (p. 148) |
|
Definition
| to see whether or not the statistical evidence supports the prima facie case against H0, a more formal derivation of the standard error of the mean difference is required. this is.. |
|
|
Term
| symmetric measure of assosiation (p. 160) |
|
Definition
| takes on the same value, regardless of whether the independent variable is used to predict the dependent variable or the dependent variable is used to predict the independent variable. |
|
|
Term
| test of statistical significance (p.146) |
|
Definition
| helps you decide whether an observed relationship between an independent variable and a dependent variable really exsists or whether it could have happened by chance when the sample was drawn. |
|
|
Term
|
Definition
tells us exactly how many standard errors separate the sample difference from zero, the difference claimed by H0. The general formula for this is:
(Ha-H0)/Standard error of the difference |
|
|
Term
|
Definition
| has different values on the independent variable but the same values on the dependent variable. |
|
|
Term
|
Definition
For large samples, the normal distribution and the t-distribution are virtually identical. Even so, computer programs usually report the P-values that are associated with the students t-distribution statistic. this statistic is called..., is calculated just like Z
(Ha-H0) / standard error of the difference
(4.6 - 0) / 1.4
=3.3 |
|
|
Term
| two-tailed test of statistical significance (p.149) |
|
Definition
| it divides the .05 rejection region in half, reporting the value above which .025 of the curve falls and the value below which .025 of the curve falls. |
|
|
Term
|
Definition
| occurs when the researcher concludes that there is a relationship in the population when in fact, there is none. |
|
|
Term
|
Definition
| occurs when the researcher infers that there is no relationship in the population when, in fact there is. |
|
|
