LESSON 1

1. What is the definition of applied research?

2. What is applied research investigation aimed at?
3. What’s the purpose behind developing models?
4. What are the reasons why we use a model-based approach?
5. What’s the difference between dependency and interdependency
models?
6. What’s the difference between a dependent, independent and
confounding variable?
7. What do we mean for intervention?
8. How do we measure the result of an intervention on the dependent
variable?
9. How do we measure complex concepts with variables?
10. What do we mean by structure in the data?
11. What are the steps in the research process (6)?
12. What are the common elements of problem definition (5)?
13. What does the research design describe?
14. Which statistical techniques do we study for the exam? Can we
combine them?
15. What are some of the appropriate methods for dependency
models only?
16. How do we measure an outcome variable when an intervention
takes place?
17. What are the most famous examples of intervention experiments?
18. What’s the difference between cross-sectional, longitudinal studies,
and panels?

LESSON 2

19. What’s the difference between primary and secondary data?

20. How can we gather primary data?
21. What are the disadvantages and the advantages of primary data?
22. What are the disadvantages and the advantages of secondary
data?
23. What are characteristics data should have to be considered quality
data?
24. What do we mean by valid data?
25. When is it that data are reliable?
26. With respect to the visual representation of observations close to a
central target, what does the three different dispositions of the points
around the center mean?
27. When is data accurate?
28. When is data precise?
29. What are the possible ways of collecting data?
30. When do we need to sample?
 31. What do we mean for representative sample? Probability vs. non-
probability?
32. Distinguish between data collection issues between primary and
secondary data?
33. What do we mean by data processing?
34. How do we prevent errors? (primary vs. secondary)
35. What is missing values analysis about?
36. What actions can be taken to handle missing values?
37. What decisions should be made when deciding to drop or impute
cases? (pros vs. cons)
38. What do we mean by pairwise and listwise deletion?
39. Which aspects should you inspect data upon?
40. How do we detect outliers?
41. How do we inspect data for nonnormality?
42. What are the most common statistical tools (descriptive analysis
and graphs)?
43. What techniques do we employ in bivariate data screening?

LESSON 3

44. What types of variables are required to perform an ANOVA?

45. How does hypothesis testing generally work?
46. What’s the difference between one-way and n-way ANOVA?
47. What’s the definition of a one-way ANOVA?
48. What are the assumptions for an ANOVA?
49. What two hypotheses are being tested in an ANOVA?
50. What is the graphical representation of the two hypotheses?
51. Explain how variability affects the results of an ANOVA and the
visual representation?
52. How do we partition the variance?
53. What do mean by variation and how do we compute it?
54. What’s the visual representation of SST, SSG, and SSW?
55. How do we fill a one-way ANOVA table?
56. What’s the test statistic in the case of ANOVA? Which distribution
does it follow?
57. What’s the typical relationship between degrees of freedom?
58. What’s the rejection rule?
59. Why do we need a post-hoc test? What does it tell us?

LESSON 4

60. What should we start with when running a regression, or, in

general, to quantitatively analyze data?
61. Explain which three parts are computed when performing a
regression.
62. What does the variable coefficients tell us?
 63. What is the STATA command to explore linearity?
64. How do we interpret confidence intervals in a regression?
65. Why do we need a multiple regression?
66. What are the advantages of a multiple regression?
67. Write the general multiple linear regression equation.
68. How is a two variable model represented?
69. How do we assess coefficients as we have more than one
independent variables?
70. Are beta coefficients completely independent (i.e., are they
influenced by one another?)
71. How can we mathematically consider these coefficients? What’s its
formal interpretation?
72. When describing the effect of the coefficients on the variable what
two key pieces of information should we always include?
73. Is the interception coefficient always relevant to evaluate?
74. How can we say if X1 has a greater impact on Y than X2? Explain
the meaning behind the results and how we compute this measure.
75. How can we interpret dummy variables in a regression model? How
about Y prediction with dummy variables?

LESSON 5

76. How can we evaluate if a model is good? Explain what the three
tools mean.
77. How can we decompose the variation in our model?
78. How can we mathematically compute the R-squared?
79. What are the issues with R-squared? Explain the effect on degrees
of freedom.
80. Explain the relationship between R-squared and adjusted R-
squared.
81. What are the most important things when comparing two models?
82. What is the model error variance? Is it relevant? What’s its
relationship with R-squared?
83. What’s the purpose of the F-test? How is it computed? What
conclusions can be drawn if you fail to reject the null hypothesis?
84. What’s the difference and relationship between the F-test and the t-
tests?

LESSON 6

85. What are the assumptions of a regression model (6)?

86. What are the properties of ordinary least squares (Theorem)?
87. Which assumptions are needed to sustain the OLS estimators’
characteristics?
88. What happens when normality is not respected?
89. Why do we need assumptions?
 90. What are residuals?
91. What do we mean by diagnostic? What should we pay attention to
in graphs?
92. When is it that a model is linear? How can we assess linearity?
93. What are our expectations for linearity scatters?
94. How can we solve the issue of linearity?
95. What considerations on the slope should be done in case of a non-
linear regression?
96. How do we read the impact of a change in X in case of a non-linear
regression?
97. What do we mean by i.i.d.? What happens in case of positive
correlation?
98. How do we check for residuals correlation? Should we worry all the
time?
99. What does homoscedasticity mean and what does it imply?
100. How do we detect heteroskedasticity?
101. What are the remedies for heteroskedasticity (4)?
102. What do we mean by endogeneity? How do we detect it?
103. Remedies to endogeneity?

LESSON 7

104. What is multicollinearity? Does it cause problems with OLS

estimators?
105. Which problems does multicollinearity cause?
106. What is VIF?
107. How do we compute VIF?
108. What is the idea behind a test on a subset of coefficients?
109. What conclusions can we draw from a Chow Test (partial F-test)?

LESSON 8

110. How do we interpret coefficients for dummy variables?

111. What’s Simpson’s Paradox about?
112. What is a slope dummy? When do we use it? How do we model
interaction?
113. How do we avoid the dummy variable trap?
114. What do we mean by multinomial categorical covariate?
115. How and why do we transform a variable into logarithmic form?
116. How do we interpret results on the coefficient depending on how
the logarithmic transformation is done?

LESSON 9

LESSON 10

What’s the difference between cluster analysis and factor analysis?

What’s factor analysis?

What’s the relationship between observed variables and factors?

What do factor loadings express?

What’s correlation?

What are the characteristics of the correlation coefficient? How is this different
from covariance?

Can variables be correlated with more than on factor? In what kind of variance
are we interested in?

When is it that variables do not form any factor?

LESSON 11

Why is factor analysis useful?

What is the difference between exploratory and confirmatory factor analysis?

What are the assumptions behind factor analysis?

What is communality and uniqueness?

What do factor loadings represent?

What are factor scores?

What are the steps of factor analysis?