Ch. 11: Statistical Inference: One-Sample Confidence Interval

lower and upper endpoints

-confidence limits -the values that mark the beginning and end of the confidence interval

Hedges estimator of d

-g -represents the size of the effect that a researcher has obtained in units of the sample standard deviation -g=.2--small effect, g=.5--medium, g=.8--large

cohen's measure of effect size d can be estimated with

-hedges g and student's d

effect magnitude statistics

-measures of effect size and measures of strength association -enables the researcher to aggregate the results of many studies in procedure called meta-analysis

open vs closed CI

-open--neither endpoint is included in the interval

The confidence coefficient corresponding to a significance level of .01 is

.99

What determines the size of a confidence Interval?

1. confidence coefficient 2. size of the sample 3. the size of the sample estimate of the population standard deviation 4. whether the interval is one sided or two sided

four common criticisms of hypothesis testing

1. does not indicate whether the effect is large or small 2. is a trivial exercise 3.Turns a continuum of uncertainty into a reject-do-not reject decision. 4. Answers the wrong question--does not address if the results are important or useful

Advantages of confidence interval

1. more precise information 2. brackets the possible values for a population mean 3. can be used to test any null hypothesis simply by looking at the interval

Null hypothesis significant testing

1. turns a continuum of uncertainty into a dichotomous reject/do not reject decision 2. distracts researchers from the main business of science 3. tells a researcher the conditional probability of obtaining a particular set of data given that the null hypothesis is true

What does 95% confidence interval mean?

95% of the sample intervals obtained will contain mew

relationship between cohen's d and hedges g

Hedges is a sample estimator of Cohen's

according to John Turkey, null hypothesis testing is

a trivial exercise

how are practical statistics related to effect magnitude statistics

effect magnitude statistics can assist a researcher in deciding whether results are practically significant

as the confidence coefficient increases from .95 to .99, the confidence interval _____

increases

Assumptions:

same as t test 1. random sample of n observations 2. population is normally distributed 3. population standard deviation is unknown

a null hypothesis significance test tells a researcher

the conditional probability of obtaining a particular set of data given that the null hypothesis is true

confidence coefficient

the probability that (1-sig. level)

confidence interval

using a random sample to specify a segment or interval on the number line such that the population mean has a high probability of lying on the segment

open interval

when neither endpoints are included in the interval