Advanced quantitative data analysis

class: center, middle, inverse, title-slide

.title[
# Advanced quantitative data analysis
]
.subtitle[
## Introduction
]
.author[
### Mengni Chen
]
.institute[
### Department of Sociology, University of Copenhagen
]
.date[
### 31/08/2026
]

---

#Introduction
About me 
- Associate professor in Copenhagen University since 01/09/2026
- Assistant professor in Copenhagen University (01/05/2021-31/08/2026)
  - Teaching: sociology of family & population studies (BA), advanced quantitative data analysis (MA), basic statistics (BA)
  - Research: family formation and dissolution, fertility, ageing, population health, population policies
  
- Research Experience: Cologne University, Germany; Catholic University of Louvain, Belgium; Wittgenstein Centre for Demography, Austria

- Phd in University of Hong Kong (2013-2017)

- **My office hour**: During Week 36-38, 10:30-11:30 am, Room 16.1.48, Building 16

---
#Introduction
About you
- Name

- Year and department

---
#Introduction
About what you know

[I would like to know what you have known about R and statistics](https://docs.google.com/forms/d/e/1FAIpQLSdzkyzduRwtcJlCIIIajYbN_wXJhWMksap86UCWnBt2Q7fKPg/viewform?usp=publish-editor)

---
#[What's your expectation about the course](https://padlet.com/ucph/what-s-your-expectation-about-the-course-1e570op3ln3xrfd9)

---
#About the course
- Main content
  - Recapture of R: R basics (Mengni Chen)
  - Cross-sectional data analysis: OLS, OB-decomposition (Mengni Chen)
  - Panel data management (Fritz Troensegaard & Kasper Lykke Dausel)
  - Panel data analysis I: fixed effect (Stephan Dochow-Sondershaus)
  - Panel data analysis II: difference in difference (Stephan Dochow-Sondershaus)
  
- Learning approach
  - Lecture 
  - Small in-class exercise
  - Bring lap-top every class
  - Fagcafe after class: Wednesday 15:00-17:00; in Room CSS 4-1-24 for Week 36-49, in CSS 4-1-30 for Week 50
  -   Victoria Ritzau Pagh is our TA.
  
---
#Assessment
  - Portfolio of 3 assignments: select one of the three topics
    - P1: 3 pages for introduction, OLS regression, and OB-decomposition
      - Deadline is Oct.4, 23:59.
    - P2: 3 pages for describing what is fixed effect and applying fixed effect
      - Deadline is Nov.29, 23:59.
    - P3: 3 pages for  describing what is DID and applying DID
      - Deadline is Dec.13, 23:59.
    - Please submit your portfolio even if it is not 100% complete.
  - A final combined product for exam: max 10 pages for all 
    
  - Individual or group work: 10 + 5 pages per extra person
  - Final exam: 12 January 2027, upload in Digital Exam before 12:00 (noon, Danish Time) 
  - Re-exam: 15 February 2027, written exam with new questions

---
#Feedback
  - Portfolio 1: all get feedback from the teacher
  
  - Portfolio 2: randomly 50% get feedback from the teacher + peer feedback
  
  - Portfolio 3: the rest 50% get feedback from the teacher + peer feedback
  
  - Overall, everyone gets twice feedback from the teacher + twice peer feedback

---
#Three topics to choose
- Q1: How does first childbearing affect women’s subjective wellbeing of women?

- Q2: How does entry into a partnership affect people’s subjective wellbeing?

- Q3: How does partnership break-up affect people’s subjective wellbeing?

- **Choose one topic for all your 3 portfolios**
- The aim of 3 portfolios is to answer the chosen question using different methods.
---
#How to measure subjective wellbing
You can choose one of the following measurements as your outcome variable:

- Subjective satisfaction with work

- Subjective satisfaction with family

- General life satisfaction
---
#Access to the Pairfam data
- [What is pairfam](https://www.gesis.org/en/services/finding-and-accessing-data/selected-german-research-projects/pairfam)

The survey data is collected annually on a national level. The initial sample comprises more than 12,000 randomly selected individuals from the three birth cohorts 1971-73, 1981-83 and 1991-93. The multi-actor survey additionally includes the current partners of these young adults, up to three (step)parents and children at the age of 8 to 15 years. The resulting data base opens up new potentials for the analysis of partnership and family relations as they develop over time.

---
#How to use open AI in the course
  - UCPH policy for open AI: download, fill and attach the template in your final exam
  
  - Is it ok to use?
    - yes: check codes, ask questions, get some ideas 
    - no: just copy the codes without understanding what you are copying
    - mindful and critical

---
#Other info
- Absalon setup
  - Data: go to absalon → this course → files → Data
      - **Women data: can be used to answer Q1**
      - **50%data: can be used to answer the Q2 and Q3**
      - Codebook
      - Variables, pairfam Waves 1-13.xlsx
    
  - Reading for each session : go to absalon → this course → files → Reading
  
  - Example reading: literature related to the three topics
      - go to absalon → this course → files → Example reading
      
  - Assignment and exam examples
      - go to absalon → this course → files → Assignment and exam examples
    
---
#What is R
If you're mainly interested in using regression to analyze a specific survey, it is
1. ...a rather tedious stats programming language.

2. ...basically a calculator:

``` r
1+1
```

```
## [1] 2
```

``` r
x <- 1
x+1
```

```
## [1] 2
```

---
#What is R
If you enjoy coding and working with several data sources at once, it is

1. ... a great platform for (social) data science of

2. ... networks, text, geo, pictures...

3. ... an all in one solution.→  I coded these slides in R!

4. ... a huge, dynamic and cooperative community.

5. ... a very higher chance to get a job in Danish labor market!

[why you should use R](https://www.youtube.com/watch?v=9kYUGMg_14s&list=PLbh5g6tUnVSRwojrTf-ZBTXeCs-oYh9DV&index=94&t=3s)

---
#What you will learn in R

(Social) Data Science

![](https://d33wubrfki0l68.cloudfront.net/571b056757d68e6df81a3e3853f54d3c76ad6efc/32d37/diagrams/data-science.png)

.backgrnote[*Source*: Grolemund and Wickham (2017)]

<src="https://d33wubrfki0l68.cloudfront.net/571b056757d68e6df81a3e3853f54d3c76ad6efc/32d37/diagrams/data-science.png" width="70%" style="display: block; margin: auto;" >

---
#Install R and R
- [First, download R and install](https://www.rstudio.com/products/rstudio/download/#download)
- [Second, download R studio and install ](https://www.rstudio.com/products/rstudio/download/#download) 
- Video on how to install them in [Windows](https://www.youtube.com/watch?v=_2sewGCA0y4) and [Mac](https://www.youtube.com/watch?v=LanBozXJjOk)
- If you don't manage to install R and Rstudio, you can use [Rstudio cloud](https://posit.cloud/)
- [Video on how to use Rstudio cloud](https://www.youtube.com/watch?v=uK1Va_UWQFc)

---
#Interface in R
<img src="https://d33wubrfki0l68.cloudfront.net/8a64bb047429d7ae0e2acae35c40e421e6439bf6/80e5d/diagrams/rstudio-editor.png", width="550px" height="550px"
 style="position:absolute; top:120px;">

---
# R Studio workflow
1. Use **Projects** to manage all files (scripts, data, figures and tables) belonging to the same project.
2. Use **R scripts** to write down your codes, so you can reproduce your results by running the codes in the R script.

---
#Set up your own Rproject and Rscript
- Click on the Project button.
  - "New Directory"
  - "New Project"
  - Finally, a name for the folder and under "Browse" where you want that folder to be located.
  - Done! Now you should find an empty folder under the path your wanted it to be set up.
  - In the future, start RStudio by double clicking on the project icon in your folder.
- You can follow the procedures to create your R project and R script in [this website, click here](https://rpubs.com/fancycmn/1213843).

- [Video on creating a project: 3'38-7'32](https://www.youtube.com/watch?v=wqOme7xsZvs)
- [Video on openning an existing project: 9'35-10'29](https://www.youtube.com/watch?v=wqOme7xsZvs)

---
#Let us do a in-class quiz

- Go to absalon and find the quiz under Week 36.1 in the "Slides" module
---
#R scripts (wait and see)
OK, you are good to go!

Pro tip: Use the R Script to take notes during the lecture!

**# take notes**

``` r
#take notes
x <- 1
x+1
```

```
## [1] 2
```
---
#First steps with R
- Creating objects
 - R is an object-oriented programming language. We create objects with the assignment operator
 
- The assignment operator may be used in both directions: <-, or ->. This can help you write more legible code.
 - Shorthand in Windows, press **alt** and **-** together
 - Shorthand in Mac, press **Option** and **-** together
 
- Run code
 - For Windows, press **Ctrl** and **enter** together
 - For Mac, press **cmd** and **enter** together

---
#First steps with R
- Creating objects
  - Objects can contain all kinds of information.

``` r
# Create a new object "test_object" and assign to it the result of 5 + 2
test_object <- 5 + 2
#printing the result: type test_object in the "Console" and see what you get
test_object
```

```
## [1] 7
```

``` r
# text can also be an object
text_object <- "I love this course so much!"
#pring the results: type text_object in the "Console" and see what you get
text_object
```

```
## [1] "I love this course so much!"
```

---
#First steps with R
compare the two sets of codes

``` r
test_object1 <- 5+2
test_object1 + 3
test_object1
```

``` r
test_object2 <- 5+2
test_object2 <- test_object2 + 3 
test_object2 
```
--
If you don't use the assignment operator (<-), R will simply print the result of an operation without changing that object (i.e., without any further consequences).

If you actually want to create or modify objects, you need to use the assignment operator and explicitly assign the result of your operation to the object you want to change.

---
#First steps with R: Use functions
function_name(argument1 = value1, argument2 = value2, ...)

Example 1:  The class() function tells you what class/type of object your are dealing with:
--

``` r
class(test_object1) # test_object1 contails whole numbers, so it is:
```

```
## [1] "numeric"
```
--

``` r
class(text_object) # "text_object" contains a sentence, so it is:
```

```
## [1] "character"
```
--

``` r
class(table) #table is a function in R
```

```
## [1] "function"
```

---
#First steps with R: Use functions
Example 2: The **seq()** function, for instance, produces a sequence of numbers and has (among others) the arguments from and to.

``` r
yet_another_object <- seq(from = 1, to = 5)
yet_another_object
```

```
## [1] 1 2 3 4 5
```
--

``` r
#Try to run
alphabet <- seq(from = "a", to = "z")
```
--
Functions will throw an error message, if you feed them with an object of a class it cannot handle.

---
#First steps with R: operators
Arithmetic operators: plus=+, multiply=*, power=^, square root=sqrt()

``` r
2 + 3 
```

```
## [1] 5
```

``` r
2 * 3 
```

```
## [1] 6
```

``` r
2^3   
```

```
## [1] 8
```

``` r
sqrt(36) #square root as a function
```

```
## [1] 6
```
---
#First steps with R: operators
Logical operator

- and = **&**;
- or = **|**;
- not= **!**;
- belong = **%in%**;
- equal = **==**;
- not equal= **!=**;

---
#Logical operator in action

``` r
x <- 2
x==3
```

```
## [1] FALSE
```
--

``` r
x!=3
```

```
## [1] TRUE
```
--

``` r
x > 5 & 6 # greater than 5 and greater than 6
```

```
## [1] FALSE
```
--

``` r
x > 5|1 # greater than 5 or greater than 1
```

```
## [1] TRUE
```
---
#Other basics
R has a steep learning curve and is, at times, simply frustrating. The documentation of a specific function can be accessed via typing **?<Name of function>**

``` r
?seq
?sqrt
```

Clean the objects or data you create

``` r
rm() #rm stands for remove
rm(x)
```

Clean the output shown in the console

``` r
press ctrl and l
```

---
#Take home
1. R is an object-oriented programming language. Objects are basically containers and they can contain anything: data, functions, other objects, results, figures, and more ... Objects can be called by their name, which needs to start with a letter. If you just type the name of an object, R prints the object.

2. Functions do stuff with objects, which we assign as values to a function's arguments. What exactly a function does, depends on the class of an object. Functions themselves are objects too.

3. Important code
- "?": View a function's help file.
- "#": defines a comment in your code. Comment as much as possible!
- "<-": is the assignment operator of R.
- +, -, *, ^ : Arithmetic operators
- & means "and", | means "or", ! means "not", == means testing whether the two are equal 
- class(): tells you the class of an object

**Reference**: Grolemund, G. and H. Wickham (2017). R for Data Science. O'Reilly.

---
class: center, middle
#[Exercise: 1-6 necessary; 7-9 optional](https://rpubs.com/fancycmn/1214111)