R: Finding Out Which Students are Improving Their Grade

I am working with the R programming language.

Suppose I have the following dataset of student grades:

my_data = data.frame(id = c(1,1,1,1,1,2,2,2,3,3,3,3), year = c(2010,2011,2012,2013, 2014, 2008, 2009, 2010, 2018, 2019, 2020, 2021), grade = c(55, 56, 61, 61, 62, 90,89,89, 67, 87, 51, 65))

> my_data
   id year grade
1   1 2010    55
2   1 2011    56
3   1 2012    61
4   1 2013    61
5   1 2014    62
6   2 2008    90
7   2 2009    89
8   2 2010    89
9   3 2018    67
10  3 2019    87
11  3 2020    51
12  3 2021    65

My Question: I want to find out which students improved their grades (or kept the same grade) from year to year, and which students got worse grades from year to year.

Using the idea of "grouped window functions", I tried to write the following functions :

check_grades_improvement <- function(grades){
  for(i in 2:length(grades)){
    if(grades[i] < grades[i-1]){
      return(FALSE)
    }
  }
  return(TRUE)
}

check_grades_decline <- function(grades){
  for(i in 2:length(grades)){
    if(grades[i] > grades[i-1]){
      return(FALSE)
    }
  }
  return(TRUE)
}

Then, I tried to apply these functions to my dataset:

  improving_students <- my_data %>% group_by(id) %>% 
  filter(check_grades_improvement(grade)) %>% 
  select(id) %>% 
  unique()


worse_students <- my_data %>% 
  group_by(id) %>% 
  filter(check_grades_decline(grade)) %>% 
  select(id) %>% 
  unique()

But I am getting empty results

Can someone please show me what I am doing wrong and how I can fix this?

Thanks!

Something like this:

library(dplyr)

my_data %>% 
  group_by(id) %>% 
  mutate(x = grade-lag(grade, default = grade[1])) %>% 
  mutate(peformance = case_when(x == 0 ~ "kept_same",
                                x > 0 ~ "improved",
                                x < 0 ~ "got_worse",
                                TRUE ~ NA_character_), .keep="unused")

     id  year grade peformance
   <dbl> <dbl> <dbl> <chr>     
 1     1  2010    55 kept_same 
 2     1  2011    56 improved  
 3     1  2012    61 improved  
 4     1  2013    61 kept_same 
 5     1  2014    62 improved  
 6     2  2008    90 kept_same 
 7     2  2009    89 got_worse 
 8     2  2010    89 kept_same 
 9     3  2018    67 kept_same 
10     3  2019    87 improved  
11     3  2020    51 got_worse 
12     3  2021    65 improved 

If we want to break the function at the first instance

check_grades_improvement <- function(grades){
  for(i in 2:length(grades)){
    if(grades[i] < grades[i-1]){
      return(FALSE)
      break
    }
  }
  return(TRUE)
}

check_grades_decline <- function(grades){
  for(i in 2:length(grades)){
    if(grades[i] > grades[i-1]){
      return(FALSE)
      break
    }
  }
  return(TRUE)
}

-testing

 my_data %>%
   group_by(id) %>% 
   filter(check_grades_improvement(grade)) %>% 
   ungroup %>%
   select(id) %>%
   unique()
# A tibble: 1 × 1
     id
  <dbl>
1     1


my_data %>% 
   group_by(id) %>% 
   filter(check_grades_decline(grade)) %>%
   ungroup() %>%
   select(id) %>% 
   unique()
# A tibble: 1 × 1
     id
  <dbl>
1     2

Or if it is for all instances

my_data %>% 
  arrange(id, year) %>%
  group_by(id) %>% 
  filter(c(FALSE, diff(grade) > 0)) %>%
  ungroup %>%
  select(id) %>%
  unique
# A tibble: 2 × 1
     id
  <dbl>
1     1
2     3
my_data %>% 
  arrange(id, year) %>% 
  group_by(id) %>% 
  filter(c(FALSE, diff(grade) < 0)) %>%
  ungroup %>%
  select(id) %>%
  unique
# A tibble: 2 × 1
     id
  <dbl>
1     2
2     3

</details>