27 terms

sampling distribution

the distribution of sample means (or other statistics) based on repeatedly drawing simple random samples

Central Limit Theorem (CLT)

theorem states the conditions under which a sampling distribution will be approximately normal (and what parameters of the sampling distribution will be)

point estimate

the estimate of a parameter with a single value

unbiased estimator

a point estimate that does not consistently underpredict or overpredict the population parameter

interval estimate

an estimate of a parameter that gives a range of likely values

confidence interval

an interval that estimates a parameter with a given level of confidence

confidence level

the % of likelihood that the interval includes the true value of the population parameter

margin of error

the maximum distance from the point estimate at the center of the interval allowed given the specified level of confidence

z critical values

the values of the standard scores in the normal distribution that correspond to a given confidence (or significance) level

student T distribution

a symmetric distribution of a sampling distribution whose shape changes with sample size

t-critical value

the value of the T distribution for a particular degree of freedom that corresponds to a given confidence (or significance) level

T-distribution

distribution used for small sample sizes and/or unknown population standard deviation

normal distribution

the distribution used for large-sample proportions, and when the sample size is large (>40) AND the population standard deviation is known

hypothesis

an assumption or premise against which to test data

alternative hypothesis

the hypothesis in hypothesis testing that you want to prove

null hypothesis

the hypothesis in hypothesis testing that is the default assumption if you fail to prove your claim with the available data

test statistic

a statistic calculated on the sampling distribution that is used to test your hypothesis

statistically significant

data is described as this if the assumptions of the null hypothesis are sufficiently unlikely to produce the results shown in the data

level of significance

The level of unlikeliness established in advance of testing below which one can conclude the data is sufficiently unlikely to be produced by random chance from the situation described by the null hypothesis that you are willing to claim that the null hypothesis is likely to be false

reject the null hypothesis

conclusion reached if P-value is less than the level of significance

fail to reject the null hypothesis

conclusion reached if P-value is greater than the level of significance

Type I error

an error that occurs when the null hypothesis is really true, but we conclude it is probably not true

Type II error

an error that occurs when the null hypothesis is really false, but we are unable to conclude it is false from the data

P-value

the probability that the data you a testing was obtained from the assumptions of the null hypothesis (to be compared to the significance level)//The probability you could obtain the sample by random chance if the null hypothesis was true.

one-tailed test

a hypothesis test that considers only values greater than or less than the null hypothesis (not both)

two-tailed test

a hypothesis that tests for "differentness" from the null hypothesis and it could be above or below the assumed mean/proportion

rejection region

the value of the test statistic outside the range of the critical values in which the P-value is low enough to reject the null hypothesis