# The Effect

## Ch 1: Designing Research

Key point: the ability to answer a question is a quality of the data (not of statistical methods), and research design determines the data


### Brief overview of concepts

“Well-designed research is research capable of answering the question it’s trying to answer”

Talk about with chapter 2, instead: 
A good research question is 
* well-defined
* answerable
* understandable

“Quantitative empirical research is just empirical research that uses quantitative measurements (numbers, usually). More data sets, fewer interviews.”
* Misrepresents the issue. Interviews generate quantitative data. Qualitative data are not substantively different from quantitative. 
* What people often mean by “qualitative” data is data that they have left unstructured. These could be translated into quantities, but the researcher has chosen to not do that. 
* Also, interviews very often yield quantitative data
* Qualitative researchers purposely leave data unstructured. This concerns me. They do not clearly define concepts or measures. Their hypothesis tests are ad hoc. They rely on intuition rather than systematic reasoning. I wonder if I am misunderstanding something (probably).


Example research question: “does adding an additional highway lane reduce traffic?”

Is “traffic” a well-defined concept?

### Correlation vs causation

Section we should spend a lot of time on in class:
“...we might want to study the impact of additional lanes by comparing two-lane highways to three-lane highways. But we probably aren’t actually interested in how much traffic there is on three-lane highways and on two-lane highways. We’re probably interested in whether we can make traffic go down by turning a two-lane highway into a three-lane highway! But as much as we want them to, the numbers we have don’t actually tell us that right away. All we have are two-lane highways and three-lane highways. We don’t have a “what if” highway that tells us how much traffic there would have been if we’d made that two-lane highway one lane wider.”

* Key idea that only gets brief treatment in 1.2
* Gets more coverage in 1.3
* Discussion ideas
	- Have class develop explanation of this paragraph
	- Suppose 2-lane and 3-lane highways have same congestion. Does that mean that adding a lane to a 2-lane highway will not change congestion?
	- What else can explain the patterns?
* Other examples
	- Reverse causality:
		+ Police and crime
		+ Chairs in auditorium and number of people standing
	- 3rd variable
		+ Schooling and earnings
		+ Polo shirts and schooling
		+ Alcohol as confounder of effect of sleep on alertness the next day
		+ Houselessness and drug use
		+ Fertility and career aspirations
		+ Exercise and blood pressure
		+ Exercise and vegetable consumption
		+ Diabetes and religiosity
		+ Fertility spike in 1970 and draft deferrment/exemption
		+ Suicide and pumpkin spice lattes
	- Both
		+ The pill and pregnancy rate over time
			* Pill -> lower risk of pregnancy each time have sex
			* Lower risk of pregnancy -> more sex
			* More sex -> more pregnancy
			* If we want to know about efficacy of the pill, data on proportion who get pregnant within a year do not answer that question even if the pill was randomly assigned. But they do let us answer a similar question.
		+ Violent media causes people to become more violent
		+ Being in the military increases the probability of later unemployment or incarceration
* Activity:
	- Choose 3 vars you think are related
	- Draw diagram of relationship
	- Have class try to come up with other relationships between the vars
* Activity
	- Read meerkat mothering article (popular press)
	- Discuss what we take away/learn from it
	- Look at original research article
	- Does the research answer the questions the pop press article implies it does?
* Activity: decide if research designs (that I write up) are experimental or observational