Exam 3 Review
Home Up Syllabus Lecture Schedule Lab Schedule Teaching Assistants Handouts

 

 

Chapters 12-16

CONCEPTUAL QUESTIONS 

In what way(s) are estimation and hypothesis testing similar? How do they differ?
Distinguish between point and interval estimates. What are the advantages of each?
What are confidence intervals? What does the width of the confidence interval indicate?
How is the width of a confidence interval influenced by sample size, variability, and the level of confidence?
When should we use ANOVA instead of t-tests?
When would you use an independent samples design? A repeated measures design? A matched participants design?
What are the advantages and disadvantages of each type of design?
What is the logic behind ANOVA (e.g., why should F = 1.00 when Ho is true? How do between group variability and within group variability influence F?).
Why shouldn’t we do a bunch of t-tests instead of an analysis of variance (see box 13.1)?
Why is the critical value for the F ratio positive?
What does a statistically significant F-test allow you to conclude?
What are post-hoc tests? Why and when do we use them?
What are planned comparisons? Why and when do we use them?
What is the relationship between independent samples F and t?
What is homogeneity of variance and why is it important?
What is the statistical difference between the independent samples and repeated measures ANOVA?
What is the advantage (both conceptually and computationally) of repeated measures ANOVA?
What is the advantage of using a factorial design (rather than conducting multiple single factor experiments)?
What three “effects” are tested in a two factor ANOVA?
What are the assumptions for each type of ANOVA?
What are the three characteristics of a pearson correlation?
When and why do we use correlations? 
What approximate values of r are considered weak, moderate, and strong?
What is the relationship between r and z?
Does correlation indicate causation?
How does restricted range influence r?  How do outliers influence r?
What is the coefficient of determination (r2) and what does it tell us?
What is regression to the mean? In what way does regression to the mean complicate interpretation of correlations?
In what other ways (besides pearson correlation) can the relationship between variables be measured? When is each type of correlation used?
What is the goal of a regression analysis?
Why do we compute the standard error of the estimate?
What is the relationship between the standard error and the correlation?

THINGS YOU MAY BE ASKED TO DO 

Compute and interpret point and interval estimates (confidence intervals) for both z and t.
Read the description of a study and identify all of the following: the IV and levels of the IV, the DV, operational definitions of IV(s) and DV(s), the scale of measurement, the type of research design, and the specific statistical analysis that should be used to analyze the data.
Distinguish independent samples from repeated measures designs.
Translate a research question into statistical hypotheses (H0 and H1).
Find the rejection region and critical value(s) for a given alpha level and use this information to formulate a decision rule.
Determine whether or not to reject H0.
Interpret the results of a statistical test.
Compute an F-ratio.
Construct an ANOVA summary table and/or complete an ANOVA summary table if given partial information.
Conduct Tukey’s HSD and Scheffe’s F tests.
Determine coefficients for a planned contrast.
Use Hartley’s F-max test to assess homogeneity of variance.
Identify, describe, and interpret main effects and interactions (through hypothesis testing, by simply looking at the pattern of means, and by using ANOVA).
Graph the results of a factorial experiment and/or interpret results of a factorial experiment from a graph.
Identify and describe the direction, form, and degree (strength) of a correlation.
Test hypotheses with the pearson and spearman correlations.
Compute each type of correlation introduced in chapter 16.
Interpret correlations and understand why we can’t infer causality from correlations.
Compute a regression equation.
Use a regression equation to find a predicted value of Y.
Interpret SPSS output for ANOVA, correlation, and simple regression.

KEY TERMS & CONCEPTS 

estimation
point estimate
interval estimate
confidence intervals
independent variable (IV) = factor
dependent variable
single factor v. factorial designs
independent samples designs
repeated measures designs
analysis of variance (ANOVA)
F-ratio
variability between groups
variability within groups
treatment effect   
experimental error
MS (mean square)
SS (sum of squares)
df (degrees of freedom)
F-distribution
ANOVA summary table
error term
post-hoc tests
experimentwise alpha level
Tukey’s HSD
studentized range statistic (q)
Scheffe’s F test
planned comparisons/contrasts
individual differences
main effect     
interaction
simple main effect
correlation
scatterplot
positive v. negative correlation
linear v. non-linear relationships
prediction
validity
reliability
theory verification
pearson correlation coefficient ( r )
covariability
sum of products (SP)
restricted range
coefficient of determination (r2)
outliers (outriders)
regression to the mean
spearman correlation (rs)
point-biserial correlation
phi-coefficient
linear equation
regression
regression line
Y intercept
least squared error
regression equation for Y
standard error of the estimate (SEE)

 

This review sheet is intended to help you prepare for the exam. While it is designed to be fairly comprehensive in scope, it is not necessarily an exhaustive list of all possible exam material. All material from the texts, lectures, and labs is fair game for the exam.