PSYCH 301 1st Edition Lecture 11Outline of Last Lecture I. Statistical SignificanceII. ErrorsIII. Four Error ScenariosIV. QuestionsOutline of Current Lecture I. Calculating PowerII. What determines power?III. Effect SizeIV. QuestionsCurrent LectureI. Power = probability the study will give a significant result if the research hypothesis is truea. Gathering information about the new distribution (when you thing the your sample came froma different population and your data shows that they actually are from different populations)b. Power = 1 – βII. Steps for calculating power:1)Gather parameters for population 1 & population 2 (the comparison distribution)- Population 2 (where Hₒ is true): μ₂, σ₂- Population 1 (where HA is true): μ₁, assume σ₁ = σ₂2)Locate the raw-score cutoff point on the comparison distribution3)Find that cutoff point in the sampling distribution for population 1; turn it into a Zscore4)Find the probability of getting a score more extreme than that Zscore for the population (use a z table for this)These notes represent a detailed interpretation of the professor’s lecture. GradeBuddy is best used as a supplement to your own notes, not as a substitute.b. No matter how far your sample is from the population mean, there is still a chance that your sample belongs in the populationIII. Two ways to increase power:a. Increasing Sample size = smaller SEM = skinnier sampling distributions = mean differences between populations are now larger1)Example: Bob’s experiment on stress and job satisfaction ratings:- Not stressed workers: μ = 3.5, σ = 1- Highly Stressed workers: mean = 3.0, n = 42)Example: how the situation gets abused- Non-coffee drinkers’ IQ: μ = 100, σ = 15- Coffee-drinkers’ IQ: mean= 101, n = 25Type IIerrorCorrectDecisionType IerrorPowerDo NOT reject Hₒ, not enough statistical powerReject Hₒ3) The more people you test, the more likely you are to reject the null hypothesisb. Increasing effect size = depends on the difference in population means and variability1)Calculating effect size:μ₁ = mean for population 1 (your sample)μ₂ = mean for population 2 (comparison distribution)σ = standard deviation- Effect sizes are only meaningful when groups are significantly different- Only calculate effect size if you reject the null hypothesis2)How big is the effect size?IV. QuestionsDo NOT reject Hₒ, not enough statistical powerReject Hₒa. Out in the world, the null hypothesis is false. Based on your data, you do not reject the null hypothesis. Did you commit an error?1)Yes, a type II errorb. What is the definition of statistical power?1)A probability that the null hypothesis is false and you’re rejecting the nullc. What are two things you can so to increase power?1)Increase sample size2)Increase effect