Intro to Statistics (Final exam) Questions and Answers 100% Solved

Intro to Statistics (Final exam) Questions and Answers 100% Solved p-value the probability, when the null hypothesis is true, of obtaining a statistic as extreme as (or more extreme than) the observed statistic smaller p-value stronger the statistical evidence is against the null hypothesis and in favor of the alternative randomized distribution 1. simulate many samples using a random process that matches the way the original data were collected and that assumes the null hypothesis is true 2. collect the values of a sample statistic for each sample to create a randomized distribution 3. asses the significance of the regional sample by determining where its sample statistic lies in the randomized sample reject the null hypothesis -if a statistic is unlikely to have occurred when the null hypothesis is true -if the p-value is less than the significant level do not reject the null hypothesis -if a statistic is not unlikely to have occurred when the null hypothesis is true -if the p-value is greater t Han or equal to the significant level significant level -a boundary below which we conclude that a p-value shows statistically significant evidence against the null hypothesis -represents the tolerable probability of making a Type I error common significant levels 1. 0.05 2. 0.01 3. 0.10 how do we decide on a significant level -many research area have a standard accepted values of significant levels for their field -researchers are allowed to choose a significant level based on what makes the most sense for their participation situation -important to choose the signifiant level for the experiment Errors: null hypothesis is true and not rejected type I error errors: null hypothesis is true and is not rejected no error error: null hypothesis is false and rejected no error error: null hypothesis is false and is not rejected type II error Strength of p values 1%: very strong 5%: moderate 10%: some/little generating randomized samples using the data hypothesis tests adress questions about a specific value and produce results with a level of strength, p-values confidence intervals adress questions of how much density curve a theoretical model for a distribution normal distribution follows a bell-shape curve central limit theorem must have random samples and a large sample size