Download Hypothesis Testing and Statistical Significance: Key Terms and Concepts and more Quizzes Psychology in PDF only on Docsity! TERM 1 null hypothesis (2 definitions) DEFINITION 1 the results are due to chance alone or there is no systematic effect TERM 2 level of significance DEFINITION 2 the probability that the result is not due to chance alone TERM 3 critical value DEFINITION 3 the cutoff sample score to determine a results significance TERM 4 alpha DEFINITION 4 probabilityof making a Type I error TERM 5 beta DEFINITION 5 probability of making a Type II error TERM 6 marginal significance DEFINITION 6 when p is equal to predetermined cutoff TERM 7 standard error of the mean DEFINITION 7 standard deviation for the comparison distribution TERM 8 confidence interval DEFINITION 8 range of scores that is likely to include the true population mean TERM 9 margin of error DEFINITION 9 amount of random sampling error in a surveys results TERM 10 directional hypothesis DEFINITION 10 one-tailed hypothesis in which the variable is either higher or lower than the independent variable TERM 21 Write the two-by-two table for critical values of z scores as a function of significance level and number of tails. DEFINITION 21 One-Tailed,p < .05 - 1.64,p < .01 - 1.96,Two-Tailed,p < .05 - 2.33,p < .01 - 2.57 TERM 22 Dr. Stats, a statistically informed person, hears about a study that says cats are smarter than dogs. What is Dr. Stats first question? DEFINITION 22 Is this a real difference or could this have occurred by chance alone? TERM 23 63% of population like porn with a margin of error of +/- 3. What does this mean? DEFINITION 23 95% confident that between 60% and 66% of people in the population like porn TERM 24 Why is the variability of a population of means so much less than the variability of a population of individual scores? DEFINITION 24 The extremes in the population of means cancel each other out. Whereas in the population of individual scores the extremes are more varied TERM 25 Study has direc. hyp. and lev. of sig. = .05, why do we use 1.96 instead of 1.64 when calculating the C.I. around sample mean? DEFINITION 25 It is necessary to have two tails to find the confidence interval. Confidence Intervals need a low point and a high point. TERM 26 If a study is sig., What does sig. mean? What does sig. NOT mean? DEFINITION 26 Significant means that it is unlikely the results were not due to chance alone. Significant does not mean that it is 100% correct TERM 27 Why would a study need replication? DEFINITION 27 Replication minimizes the chance that the results were due to chance alone TERM 28 Compare and contrast Type I error and Type II error DEFINITION 28 Type I error is when the research hypothesis is supported by the experiment, but the null hypothesis is true for the real population. Type II Error occurs when the null hypothesis is supported by the experiment, but the research hypothesis is true for the real population. They are both decision errors TERM 29 What is the effect size, d, needed to say the study has an effect size of a. small, b. medium, c. large DEFINITION 29 a. 0.2, b. 0.5, c. 0.8 TERM 30 Compare and contrast significance and effect size. DEFINITION 30 They are both ways to evaluate the study. Significance is the probability the results are due to chance alone, and effect size how big or important the effects are TERM 31 Compare and contrast significance and power. DEFINITION 31 Power is looking at Type II error and significance is looking at Type I error. Both refer to the probability of making decision errors TERM 32 What is the relationship between power and beta? DEFINITION 32 They add up to 100% TERM 33 What five factors influence power? DEFINITION 33 Significance Level (Alpha),Amount of Tails,Hypothesis Testing Method,Effect Size, and Sample Size TERM 34 Name the eight ways to increase power in a study. DEFINITION 34 Increase Effect Size,One Tail Design,Decrease Standard Error,Hypothesis Method Used,Increase Sample Size,More Lenient Significance,Increase Difference Between Means TERM 35 When does a researcher calculate power? Explain why in your own words. DEFINITION 35 Before collecting data because we want to make sure we have the right amount of power for the experiment.