PSYC 203 1st Edition Lecture 13Outline of Last Lecture I. Monty Hall Problem II. Construct validity III. Reliability and Validity IV. Hypothesis V. NHST (null hypothesis significance testing)Outline of Current Lecture I.Review II.Answer monty hall problemIII.Null hypothesis IV.Alternative hypothesis V.Specifying null and research hypothesis VI.Practice problems Current LectureI.Review a. Construct Validity– Convergent Validity – Discriminant Validityb. Reliability needed for Validity– But being Consistent DOES NOT guarantee ValidityII.Answer monty hall problema.You should switch because the first foor has a 1/3 chance of winning but the second door has a 2/3 chance 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. http://www.stat.sc.edu/~west/javahtml/Lets MakeaDeal.html c.III.Null hypothesis a.Alternative outcomes IV.Alternative hypothesis a.Essentially, then null is the opposite of what you think is going to happen V.Specifying null and research hypothesis a. Unlike the null, the research hypothesis is at the level of the sample and directly linked to our method/measures i. Examples 1. Participants who take the pill will report less depression than participants who do not. 2. Men and women will report different levels of aggression on our measure. 3. There is an association between SAT scores and college GPA in our sample. b. This is what is stated in journal articles c.Example 1 is a directional hypothesis because “less depression” d.Example 2&3 is non-directionale. Non-directionali. Reflects a difference between groups, but the direction isnot specifiedii. Example: the mean for aggression will be different for men and womenf. Directional Research Hypothesisi. Direction of difference is specifiedii. Men will have higher scores on our aggression measure than womeng.directional hypothesis requires more precision because there are more ways you can be wrong. h. Process of Null Hypothesis Significance Testing (Differences between Two Groups)i. Decide the likelihood of observed difference if the NULL were TRUEj. If it is very unlikely – we say the difference was statistically meaningfulk. therefore, we REJECT the NULLl. We don’t normally prove something...– Just REJECT or FAIL TO REJECT the NULLm. Framed as a declaration not a question. i. Women will report more aggressive behaviorsthan men.ii. NOT: Will women report more aggressive behaviors thanmen?n. Specific.– Women will report more aggressive behaviors than men.o. NOT: Women and men will differ.p. Based on Scientific Literature – Science is cumulative– Research builds on the pastq. Concise– Brief and to the pointr. Testable– Could be evaluated with dataVI.Practice a. Research says that food supplements high in antioxidants can reduce or reverse age related declines in cognitive functioning in lab rats. You recruit a sample of older adults and give half of them these types of supplements and the other half a placebo, then later both groups complete a cognitive task.i.Null hypothesis: population of older adults given supplements equals the population of older adults given placebosii.Research hypothesis: the mean of the population given the treatment will be higher than the mean of the population given the placebos b. Participation in social activities with peers has been shown to improve self-esteem in adults. You recruit a group of children and measure their involvement in social activities and measure their self esteem.i.Null: There is no association in involvement in activity andself-esteem.ii.Research: there is a positive association between involvement and self-esteem. c. Research shows that activities such as listening to music use cognitive resources, leaving less attention available for other tasks. A class has an important test coming up. Some students study while listening to music, others study in silence.i. Null: Population students who study while listening to music will score the same as those who study in silence ii. Research: Students who study in silence will score higherthan those who study with music. 1. This is actually a personal hypothesis…some may hypothesize that those who study while listening to music will score higher than those who don’t. d. Research shows that social intelligence predicts academic performance. A dean begins measuring social intelligence in college applicants and examines their college grades.i.Null: there is no association between social intelligence and academic performance ii.Research: There is an association between social intelligence and academic performance. 1.Notice that this research hypothesis is