Free online courseData Analysis for Social Scientists

Conteúdo do Curso

• Video class: Lecture 01: Introduction to 14.310x Data Analysis for Social Scientists

  1h00m

• Exercise: What is one potential issue with using correlation to infer causation when analyzing data related to social sciences?

• Video class: Lecture 02: Fundamentals of Probability

  1h07m

• Exercise: What is the probability that a randomly chosen event A, contained within the sample space S, will have an exhaustive relationship with event B when their union is equal to the sample space?

• Video class: Lecture 03: Random Variables, Distributions, and Joint Distributions

  1h12m

• Exercise: What is the primary distinction between discrete and continuous random variables?

• Video class: Lecture 04: Gathering and Collecting Data

  1h23m

• Exercise: Which of the following statements is TRUE about methods to collect data for research purposes?

• Video class: Lecture 05: Summarizing and Describing Data

  1h08m

• Exercise: What is one key advantage of using the kernel density estimator over histograms for analyzing distributions of data?

• Video class: Lecture 06: Joint, Marginal, and Conditional Distributions

  0h59m

• Exercise: In the context of probability and statistics, what is meant by the 'support' of a distribution?

• Video class: Lecture 07: Functions of Random Variables

  1h20m

• Video class: Lecture 08: Moments of Distribution

  1h18m

• Exercise: What does the probability integral transformation allow us to do when we want to simulate random draws from a distribution?

• Video class: Lecture 09: Expectation, Variance, and Introduction to Regression

  1h08m

• Exercise: In the context of probability theory, why is it important to calculate the expected utility of a game rather than just the expected monetary winnings?

• Video class: Lecture 10: Special Distributions

  1h15m

• Exercise: What principle is NOT one of the three important principles outlined by the Belmont Report regarding human subjects in research?

• Video class: Lecture 11: Special Distributions, continued. The Sample Mean, Central Limit Theorem, and Estimation

  1h13m

• Exercise: What is an unbiased estimator for a parameter θ?

• Video class: Lecture 12: Assessing and Deriving Estimators

  1h06m

• Exercise: In the context of data estimation, which of the following statements is true regarding consistent estimators?

• Video class: Lecture 13. Confidence Intervals, Hypothesis Testing, and Power Calculations

  1h16m

• Exercise: What does the 'standard error' of an estimator represent in the context of data analysis?

• Video class: Lecture 14: Causality

  1h15m

• Exercise: In the context of causal inference, which of the following describes the Stable Unit Treatment Value Assumption (SUTVA)?

• Video class: Lecture 15: Analyzing Randomized Experiments

  1h19m

• Exercise: What is one method used for analyzing completely randomized experiments as mentioned in the lecture?

• Video class: Lecture 16: (More) Explanatory Data Analysis: Nonparametric Comparisons and Regressions

  1h22m

• Exercise: What is one of the primary issues faced in the interpretation of experimental results in pharmaceutical trials due to financial incentives?

• Video class: Lecture 17: The Linear Model

  1h20m

• Exercise: In the context of estimating parameters of joint distributions in social science, if we replace a categorical coin flip treatment variable with a continuous random variable in a linear regression model, what concept does this illustrate?

• Video class: Lecture 18: The Multivariate Model

  0h41m

• Exercise: In the context of the multivariate linear model discussed, what is a key assumption made to ensure that the model can be estimated properly?

• Video class: Lecture 19: Practical Issues in Running Regressions

  1h20m

• Exercise: What is a key distinction between using a t-test and an F-test in the context of regression analysis?

• Video class: Lecture 20: Omitted Variable Bias

  1h20m

• Exercise: When transforming a variable for linear regression, why might you choose to transform the independent variable instead of the dependent variable?