Notes for week 4
Notes for week 4 Psych 2110
Popular in Elem Statistics Business
Popular in Psychology (PSYC)
verified elite notetaker
This 3 page Class Notes was uploaded by KhloNotes on Saturday September 24, 2016. The Class Notes belongs to Psych 2110 at University of Alabama - Tuscaloosa taught by Andre Souza in Fall 2016. Since its upload, it has received 4 views. For similar materials see Elem Statistics Business in Psychology (PSYC) at University of Alabama - Tuscaloosa.
Reviews for Notes for week 4
Report this Material
What is Karma?
Karma is the currency of StudySoup.
Date Created: 09/24/16
Key: Bold/Italic = important information, Highlighted = vital for test From 211 Elementary Statistical Methods (Psychology) ; Professor Souza Ideas Stressed/Restressed in Chapter Four: Statistics = random variables Inferential Statistics = procedures for drawing conclusion based on the scores collected in a research study but going beyond them; creates an educated guess for studies where it is impossible to question the entire population Memorize the Standard Distribution Variance Formula (Depicted below) Intro to Hypothesis Test A sample statistic is the best option for making an educated guess Summary of the five steps of the hypothesistesting procedure: 1. Restate the question as a research hypothesis and a null hypothesis about the populations. 2. Determine the characteristics of the comparison distribution. 3. Determine the cutoff sample score on the comparison distribution at which the null hypothesis should be rejected. 4. Determine the sample's score on the comparison distribution. 5. Decide whether to reject the null hypothesis. Ex: How conservative is Alabama? Five students ask a sample of fifty people with a scale (110) Thus, the head researcher has five guesses to average together = Central Limit Theorem Central Limit Theorem Given any population, the means of random samples together will converge to a normal distribution (mean of sample distribution & population are equal) Standard Distribution Variance (Standard Error of the Mean) Formula: σ √ n Standard Error = How much the sample mean would vary if we took several samples at the same size from the same sample More people = more precision Real World Application 1. Usually just grab one sample 2. Take the zscore for this sample (distance from the mean) Key: Bold/Italic = important information, Highlighted = vital for test Formula: z = x − M √nσ or z = x̄ − M SD Hypothesis Testing Cont. Objective = check whether a set of data agrees with a certain prediction Prediction = hypothesis → statement about a population, a guess that a parameter takes a particular numerical value or falls within a particular interval, systematic way to test claims or ideas about a population A significance test uses data to summarize the evidence about a hypothesis (comparing sample statistics or point estimates of parameters) to the values predicted by the hypothesis Ex: Are men paid more for the same job? Answer: ▯women < ▯men = if that #10, then ▯women must be less than 10 * if you see x̄women < x̄ men = sample mean (this is purposely the wrong answer, look for the ▯ symbol which is the correct form of the problem) Objective: To determine the likelihood that a sample statistic would be selected if the hypothesis regarding the population parameter were true Four Basic Steps (look in the book for more to review) State the hypothesis Set a criteria for decision (basically choose a αlevel) *Has to be done before observations Compute the test statistic (collect data) Compare and make a decision Ex: 1. Fabio is not gay. 2. How many gay bars he goes to → criteria = 3 bars 3. Stalk him to collect data 4. He goes to 4 → my hypothesis is wrong Explanation: Every hypothesis test has two hypothesis ● Null H →0tatement about value of a population parameter ● Alternative H or H → statement that the population parameter falls into an 1 a alternative range of values than those stated in null *Null usually states no effect Key: Bold/Italic = important information, Highlighted = vital for test Good v Bad ● A good hypothesis is one that can be rejected or falsified *the absence of evidence is not evidence of absence **null = crucial role → we can not prove something to be true, but we can prove it to be false → fail to reject it/rejection Ex: You get a call and a girl says pregnant! Null Hypothesis: Karen is n ot pregnant How to Test: Pregnancy Test Two Outcomes: K aren is, Karen isn’t Pregnancy Test Reality Not Preg Preg + False + Correct Correct False 2 possible errors of null hypothesis: ● Type 1 = We can reject the null (H ) when t0 null (H ) is true (0 se positive) ● Type 2 = Failure to reject the null (H ) when0 e null (H ) is false0 alse negative) *preventable with more research One Tailed v. Two Tailed Test Alternate ways of computing the statistical significance of a test statistic One Tailed: Only interested in one side of test, i.e.: Are girls are smarter than boys? You pay attention to the gender falling in the five percent zone. Two Tailed: Interested in both sides of the spectrum, i.e.: Does drinking lower your IQ? You pay attention to both sides.