{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

LN1_Introduction+to+RA

LN1_Introduction+to+RA - Lecture Note 1 Introduction to...

Info iconThis preview shows pages 1–4. Sign up to view the full content.

View Full Document Right Arrow Icon
Lecture Note 1 Introduction to Regression Analysis Empirical Methods II (API202A) — Spring 2009 Harvard Kennedy School 1 How Can Regression Analysis Help Us Shape Policy? 1.1 Definition Regression analysis (RA) is a widespread statistical tool used to study the empirical rela- tionship between variables. The phrase ‘the empirical relationship between two variables’ has two meanings: The extent to which two variables move together in the data, and therefore, the predictive power that one variable has over the other – called the association or correlation between two variables. The degree to which changing one variable affects the outcome of another variable – called causality or causal effect of one variable over another. Example: Sleeping with the light on and myopia. A study published in Nature in 1999, and broadly covered in the popular press, stated that “Young children who sleep with the light on are much more likely to develop myopia in later life.” 1
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
LN1—API202A Spring 2009 Harvard Kennedy School 1.2 How Can Regression Analysis Help Us Shape Policy? RA can help us to identify the causal effect in the association that we observe between variables in the data. Why do we want to know the causal effect? To develop sound policies that deliver the expected outcome. Example: Back to sleeping with the light on and myopia. “Young children who sleep with the light on are much more likely to develop myopia in later life.” Based on this evidence, and to reduce myopia in the population, a government pro- gram was suggested to decrease the number of children sleeping with the light on. Would we characterize this program as good or bad policy? Our goal in this course is to understand RA and how it can help us address policy ques- tions about how one variable affects another. Ultimately, we are interesting in answering: Under what conditions can we infer a causal effect from the association that we observe between variables? How can RA help us to achieve these conditions? Are we using sound empirical evidence to shape policy? 2
Background image of page 2
LN1—API202A Spring 2009 Harvard Kennedy School Examples of policy questions that we can answer with RA. i) Questions that deal with the effect of one variable on another: Does maternal smoking affect newborns’ weights? If so, by how much? Does a higher minimum wage increase unemployment? If so... Example of an answer obtained using regression analysis: A 1 % increase in state-level minimum wages reduces employment of young blacks and hispanics by 0.5 to 0.6%. (Neumark and Wascher 2007) ii) Questions that specifically deal with the effectiveness of a government program (pro- gram evaluation): What is the effect of conditional cash transfers on health and educational outcomes?
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 4
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}