PSYC 295 1st EditionExam # 2 Study Guide Lectures: 7-12Lecture 7 (September 23)Define Reliability.Reliability: consistency, always getting the same result back in an experiment.-if it is consistent then you have high re-test and is measured with correlation.Describe Spearman’s Rho and Cronbach’s alpha.Spearman’s Rho: two things correlate when they vary together; asses how well the relationship between two variables can be described using a monotonic function.Cronbach’s alpha: how closely related a set of items are as a group; measure of internal consistency. Lecture 8 (September 25) List the advantages and disadvantages of the Types of Measures.1. Self Report: a. Advantages: fast, simple, straight forward b. Disadvantages: easily biased 2. Observational (behavioral)- voluntarilya. Advantages: quantifications (put a # on it); connections to behavior, they are objective b. Disadvantages: quantification (some behavior exists on a continuum and does not fit into a category) 3. Physiological (biological)- involuntarilya. Advantages: precise quantification b. Disadvantages: connection to behavior psychological constructs. *combination of them is best*List and describe the types of scales.-Categorical: hair color, gender, field of study, college attended, political affiliation- Ordinal: ranking -Quantitative: it’s a number value; height, age, GPA, salary, temperature- Interval: distance between; ex: 100 degrees is twice as hot as 50 degrees which is twice as hot as 25 degrees.- Ratio: interval scales with a true zero point- Continuous: type of quantitative with a decimal List and describe validities. -Face validity: reasonable (subjective; have to have empirical evidence with lots of studies) -Content validity: complete (are you missing some in your validity?)-Predictive validity: model=data -Convergent validity: similar -Discriminant: different What the survey issues? Explain each.1. Leading questions: lack of neutrality 2. Double Barreled Questions: 2 questions at the same time 3. Double Negatives: confusing questions 4. Question order: bias from previous questions 5. Response sets: shortcuts 6. Social Desirability: yay/nay; fence setters (those who are neutral) Lecture 9 (September 30) Define Kurtosis and describe rank and transform data. Kurtosis- when a curve is problematic (the more spread out curve is the more problematic than the less spread out curve) 1. Rank  non-problematic (highest or lowest number in order) 2. Transform data  (pretty common but hard to imagine) use raw data when making graphs What does ‘Mu’ and ‘Omega’ stand for?-‘Mu’= populationmean; often used in hypothesis testing -‘Omega’= standard deviation of population Describe null and alternate hypothesis. -Null- (H not; H0) generally null hypothesis is bad -alternate hypothesis- (H alt; H1)Lecture 10 (October 2)Define correlation. Correlation: association between 2 variables *Correlation is not causation Give an example of a causal claim. -Eating chocolate increases one’s self esteem.Lecture 11 (October 7)Describe how regression, t-test, and ANOVA are reported.-Regression: like a table, although the APA does not explicitly say this  example: social support significantly predicted depression scores b= -.34, t(225)=6.53 p<.001 social support also explained a significant proportion of varience in depression scores, R2=12, F(1,225)=42.64 p<.001.-T-test: reported like chi-squares, but only the degrees of freedom are in parenthesis. Following that, report the t statistics (rounded to 2 decimal places) and significance level.  example: There was a significant effect for gender, t(54)=5.43, p<.001 with men receiving higher scores than that of women. -ANOVA: reported like the t-test, but there are 2 degrees-of-freedom numbers to report.  example: there was a significant main effectfor treatment, F(1,145=5.43 p=.02, and a significant interaction, F(2,145)=3.24, p=.04. Be able to know the table about t-tests, ANOVA, and regression.Define Type I and Type II errors.Type I: the null is true but you reject it Type II: the null is false but you accept it.Lecture 12 (October 9) Define residual.Residual: the difference between the observed value of the dependent variable (y) and the predicted variable (y hat) is called residual (e) each data point has one residual.*residual= observed variable-predicted variableList and describe the 3 causal claims.1. covariance: 2 variables; one goes up the other either goes up or down 2. temporal precedence: has to happen before 3. lack of confounds: the main reason we do experiments Define multicollinearity (collinearity). Multicollinearity: a statistical phenomena in which two or more predictor variables in a multiple regression model are highly correlated meaning that one can be linearly predicted form the others with a non-trivial degree of accuracy. What are 2 terms for variables in regression analysis?1. Predictor variables 2. Criterion variables What are 2 reasons we use regression? 1. Sometimes thing cannot be randomly assigned (for ethical or physical reasons)2. Good starting point Define longitudinal studies. Name two disadvantages of longitudinal studies. Longitudinal study: a study with repeated observations over a period of time. -Disadvantages: a lot of time spent on them and resources; low