Data

df <- read_dta("survey_final.dta")
head(df)

## # A tibble: 6 × 226
##   responseid datesubmitted         lastpage startlanguage       seed datestarted
##        <dbl> <chr>                    <dbl> <chr>              <dbl> <chr>      
## 1         53 "2025-04-18 15:49:18"       53 it             673068623 2025-04-18…
## 2         54 ""                          29 it            2050084500 2025-04-18…
## 3         55 "2025-04-18 23:13:11"       53 en            1981381329 2025-04-18…
## 4         56 ""                          15 it              77409783 2025-04-18…
## 5         57 "2025-04-18 23:35:07"       53 it            1009827004 2025-04-18…
## 6         58 "2025-04-19 00:00:15"       53 it              60550412 2025-04-18…
## # ℹ 220 more variables: datelastaction <chr>, gender <chr>, gender_other <chr>,
## #   birth_year <dbl>, whattypeofhighschooldidyougradua <chr>, v12 <chr>,
## #   erasmus_status <chr>, course <chr>, course_other <chr>,
## #   migrant_student <chr>, migrant_mother <chr>, migrant_father <chr>,
## #   mother_edu <chr>, father_edu <chr>, howwouldyourateyouracademicresul <chr>,
## #   siblings <chr>, doyoulivewithyourparents <chr>, religious <chr>,
## #   pleaseforeachofthefollowingstate <chr>, v28 <chr>, v29 <chr>, …

# convert to date-time format
df$datelastaction <- as.POSIXct(df$datelastaction, format="%Y-%m-%d %H:%M:%S")
df <- df |>
  filter(datelastaction >= "2025-04-18 17:10:00") # when the survey was sent out officially

df <- df |>
  filter(!course %in% c("Research fellow", "Professor"))# drop non-students

Response summary

• Students (Target): 1434
  

• PhD students (Target): 23
  

• Total target population: 1457
  

• Responses collected: 231 (total obs 236, but 5 dropped: responses before the official launch or reported as research fellow and professor)

• Response rate: 16% (very low…)

• Spoiler: some clearly ghosted us harder than others

Let’s meet the 16% who took time

Let’s see when the waves of responses hit

• April 18: Survey officially launched at 17:10
• April 22: ? Dunno
• April 24: TA shared it with one SRS Bachelor class
• April 25: TA circulated it among SRS PhD students
• May 04: ? Dunno

How much of the survey did they complete?

Of all respondents, 180 (78%) actually hit “submit”, completing our survey.

## 
##   0   1 
##  51 180

Which language did they choose?

Let’s take a quick look at who preferred Italian vs. English.

## 
##  en  it 
##  19 212

• No surprise: Italian dominates

Gender profile

Let’s look at how gender is distributed among respondents.

## 
##            Female              Male             Other Prefer not to say 
##               161                53                 2                 7

Political orientations

table(df$self_report_LR, useNA = "always") # Self-reported left (1)-(10) right

## 
##    1    2    3    4    5    6    7    8    9   10 <NA> 
##   27   40   55   32   18    7    4    3    1    1   43

table(df$social_LR, useNA = "always") # Society: 1 = participation and expression, 10 = order and stability

## 
##    1    2    3    4    5    6    7    8   10 <NA> 
##   26   33   46   23   36    6   10    3    2   46

table(df$econ_LR, useNA = "always") # Economy: 1 = government control, 10 = free market (original scale reversed)

## 
##    1    2    3    4    5    6    7    8    9   10 <NA> 
##    2    3   13   26   27   47   26   28    8    6   45

Political compass

To prevent overlapping points in the plot, random noise within ±0.25 units is added using geom_jitter()

Political compass by color gradient

• Common profile of SRS students: Both economic left and right libertarians
• Fuzzy identity labels: Self-reported left-right positions don’t always align with respondents’ concrete views on economic and social dimensions

Also, by fertility intentions for curiosity

## 
##                   Yes                    No I haven't decided yet 
##                    80                    36                    64 
##  Prefer not to answer               Missing 
##                     6                    45

What have we learned up to this point?

• Overall response rate was very low (16%)
• Completion rates were fine (78%), but still visible drop-off
• Strong Italian-speaking environment
• Gender unbalance
• Political profile
• …

This quick intro provides an overall context for the presentations we are about to see.

Over to u

Now that we’ve seen who responded, when, and how…

It’s your turn to dive into the data and tackle our two defined project topics: political outcomes and fertility intentions.