Quantitative Research methods

Amine Hadji

Leiden University

February 8, 2022
 Who are we?

Who am I?
 Who are we?

Who am I?

Who are you?

 Why this course?
Content:
• Introduction to basic statistics
• Conceptual understanding of statistical tools

Skills:
• Critical analysis of various types of data and appropriate use of statistical
methods
• Data description and analysis using R
• Communication and presentation of statistical results
 What you need to know
Course Material:
• (Mind on) Statistics 4th (International) Edition MoS
• Extra readings on Brightspace (BS)

Grades:
Participation 10%
R Homework (+ 1 Quizz) 15%
Mid-term 15%
Final exam 30%
Research Project 30%
 Statistics, data

Statistics: A branch of mathematics dealing with the study of data, which

covers its collection, organization, analysis, interpretation and presentation.

Data:
1 Factual information used as a basis for reasoning, discussion, or calculation.

2 Information output by a sensing device or organ that includes both useful

and irrelevant or redundant information and must be processed to be
meaningful.
3 Information in numerical form that can be digitally transmitted or processed.
 Statistical data analysis
Seat belt legislation

• Introduced in western countries between 1960-1980

• Controversy:
• pro-: reduced number of injuries/deaths on the road
• anti-: risk compensation and risk homeostasis advanced by researchers John
Adams and Gerald Wilde. This theory states that, if the risk of death or injury
from a car crash is reduced by the wearing of seat belts, drivers will respond
by reducing the precautions they take against crashes (source: wikipedia).
• Both arguments seem somehow right
• Controversy can only be solved by quantitative/statistical analysis.
 Responsible use of statistics I

• Sentenced to life imprisonment in 2003 for

7 murders
• Henk Elffers (court expert of statistics):
1:342 million (no correction for multiple
testing, wrong assumptions, bad
interpretation of the result)
• Richard D. Gill and Piet Groeneboom
recalculated: 1:25 (that a nurse can
experience a series of events like this)
• Exonerated in April 2010
causes lower blood pressure.
When groups are compared in an observational study, the groups usu-
ally differ in many important ways that may contribute to the observed rela-
Responsible use of statistics II

case study 1.6 Does Aspirin Reduce Heart Attack Rates?

• In 1988, a five-year randomized experiment
conducted using 22,071 male physicians Table 1.1 The Effect of Aspirin on Heart Attacks
Read the original source on your CD.
between the ages of 40 and 84
In •
1988, the Steering Committee of the Physicians’ Health Heart Doctors in Attacks per
Does aspirin reduce the risk ofStudy Research
a heart Treatment Attacks Group 1000 Doctors
Group attack?
released the results of a five-year randomized experiment conducted
Aspirin 104 11,037 9.42
using 22,071 male physicians between the ages of 40 and 84. The purpose
Placebo 189 11,034 17.13
of •
the There
experiment were two treatment
was to determine groups
whether taking aspirin reduces (aspirin
the risk
of a heart
or attack. The physicians had been randomly assigned to one of the
placebo) Because the men in this experiment were randomly assigned to the
two treatment groups. One group took an ordinary aspirin tablet every other conditions, other important risk factors such as age, amount of exercise,
day, while the other group took a placebo. None of the physicians knew dietary habits should have been similar for the two groups. The only im
whether he was taking the actual aspirin or the placebo. tant difference between the two groups should have been whether they t
The results, shown in Table 1.1, support the conclusion that taking as- aspirin or a placebo. This makes it possible to conclude that taking asp
pirin does indeed help to reduce the risk of having a heart attack. The rate actually caused the lower rate of heart attacks for that group. In a later ch
of heart attacks in the group taking aspirin was only about half the rate of ter, you will learn how to determine that the difference seen in this sam
 Further important/responsible use of statistics

• What are your examples?

 Further important/responsible use of statistics

• What are your examples?

• Does smoking increase the risk for lung cancer?
• Is acupuncture effective?
• Does a specific drug cure a certain disease?
• Who will win the next election in the US?
• Does the Higgs boson exist?
 Many solutions

Raphael Silberzahn and Eric Uhlmann (2015), in Nature.

Big Data
 Privacy

• Data is more valuable, in many instances automatically recorded

• Abuse of private data (e.g. Cambridge Analytics)
• How to protect data? Better legislation?