congpaperscript

What is needed:

1. Congress vote and seats tally in 2024 nationally, and states [gains and losses in comparison to 2019]
2. Congress performance in a historical setting (seats contested, deposits)
3. Congress social base – caste and class [using Lokniti-CSDS]
4. Party performance using constituency-level demographic estimates
5. Party performance using constituency-level political measures (margin of victory etc)

library(readr)

scrape_row <- read_csv("scrape_row.csv", col_names = TRUE)

## Rows: 8902 Columns: 51
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (20): State_Code, State_Name, idi, Constituency_Name, Constituency_Type,...
## dbl (31): Constituency_No, Position, Vote_Share_Percentage, Margin_Percentag...
## 
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

data2024 <- scrape_row

library(dplyr)

## 
## Attaching package: 'dplyr'
## 
## The following objects are masked from 'package:stats':
## 
##     filter, lag
## 
## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

library(tidyselect)
library(tidyverse)

## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ forcats   1.0.0     ✔ stringr   1.5.0
## ✔ ggplot2   3.5.0     ✔ tibble    3.2.1
## ✔ lubridate 1.9.2     ✔ tidyr     1.3.0
## ✔ purrr     1.0.2     
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag()    masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors

library(tidyr)

# Let me first quickly change BJP and INC in the party column name to acronyms

data2024 <- data2024 %>% mutate(Party = case_when(Party == "Indian National Congress" ~ 'INC',
                                                  Party == "Bharatiya Janata Party" ~ 'BJP', TRUE ~ Party))

#let's filter out the INC candidates data
dataINC <- data2024 %>% filter(Party ==  "INC")
# the total number of seats that they won
sum(dataINC$Position == 1)

## [1] 99

#Now let us separate the seat tallies by states.

incvictory <- dataINC %>% filter(Position == 1)

# Let us make a table out of it

INCstatedf <- as.data.frame(table(incvictory$State_Name))

library(kableExtra)

## 
## Attaching package: 'kableExtra'
## 
## The following object is masked from 'package:dplyr':
## 
##     group_rows

colnamestate <- c('State', 'Seats Won')

colnames(INCstatedf) <- colnamestate

kbl(INCstatedf) %>%
  kable_minimal() %>%kable_styling(full_width = FALSE)

State	Seats Won
Assam	3
Bihar	3
Chandigarh	1
Chhattisgarh	1
Goa	1
Gujarat	1
Haryana	5
Jharkhand	2
Karnataka	9
Kerala	14
Lakshadweep	1
Maharashtra	13
Manipur	2
Meghalaya	1
Nagaland	1
Odisha	1
Puducherry	1
Punjab	7
Rajasthan	8
Tamil_Nadu	9
Telangana	8
Uttar_Pradesh	6
West_Bengal	1

# Most number of victories has come in Kerala, followed by Maharashtra and TN.

# Let us first select some relevant columns 

incvictoryprev <- incvictory %>% select(State_Name, Constituency_Name, Constituency_Type,
                                        Candidate_Name, Vote_Share_Percentage, Margin_Percentage, Margin, Votes,Vote_Share_Percentage, Valid_Votes, winner2019)

# let us now see who was the biggest loser compared to 2019.

inroads <- as.data.frame(table(incvictoryprev$winner2019))

# It can be seen that most of their new victories came against the BJP with 41
# It is followed by the TRS that floated the idea of having a third front 
# The idea was thwarted in all senses and the party suffered heavily in this
# election with the rise of NDA in the state and INC's consolidaiton of its momentum from
# the earlier held state elections.AIMIM won the one seat that was left behind
# after NDA and INDIA won 8 each.
# Three are from DMK but they were in coalition in the state
# so it doesn't count as much 
# Gained two from the Shiv Sena as the officially recognized entity under the name
# is the one under Shinde. Ramtek and Kolhapur.
# In Kolhapur, the runner up from SHS was the previous incumbent. 
# Won one from BSP in Saharanpur.
# Won one from CPM as they didn't form a coalition in this election in the state
# for seat sharing.
# One from an indepednet candidate in Maharashtra from Amravati. The independent
# was anyways a member of the UPA in 2019
# One against JDS in Karnataka. The unpopular Revanna was defated by 
# an INC candidate
# One against JDU in Bihar's Katihar
# One against NCP in Lakshadweep.
# One from NDPP in Nagaland, which is in coalition with the BJP in NDA.
# One in Meghalaya against NPEP, which is in the NDA.
# One in Nagaland against NFP, which is in NDA.
# One in Punjab against SAD, which contested indepedently of any coalition this time.
# So removing the DMK seats and the one independent candidate,
# They have won 58 new seats against the NDA or non-INDIA members. 

# if we make a dataset that only uses other party winning data then we can see how many did the congress lose

dataothers <- data2024 %>% filter(!Party ==  "INC")

dataothers <- dataothers %>% filter(Position == 1)

table(dataothers$winner2019)

## 
##   AAAP   ADAL   ADMK  AIMIM   AITC  AJSUP    BJD    BJP    BSP    CPI    CPM 
##      1      2      1      2     22      1     12    251      9      2      2 
##    DMK    INC    IND   IUML  JD(U)    JMM KEC(M)    LJP    MNF    NCP   RLTP 
##     21     15      2      3     15      1      1      6      1      4      1 
##    RSP    SAD    SHS    SKM     SP    TDP    TRS    VCK  YSRCP 
##      1      1     16      1      5      3      3      1     22

dataothersINCrelation <- dataothers %>% filter(dataothers$winner2019 == "INC")

table(dataothersINCrelation$Party)

## 
##                        Aam Aadmi Party           All India Trinamool Congress 
##                                      1                                      1 
##                                    BJP    Communist Party of India  (Marxist) 
##                                      6                                      1 
##              Dravida Munnetra Kazhagam                            Independent 
##                                      1                                      2 
##                 Jharkhand Mukti Morcha Marumalarchi Dravida Munnetra Kazhagam 
##                                      1                                      1 
##              Voice of the People Party 
##                                      1

# As we can see that INC lost most of its seat that it lost to the BJP in 2024. 
# They lost six to them.
# They lost one to VVP in Tamil Nadu - Vaiko 
# There is also the notable exception of AAP as they have conceded one seat to them 
# while they are in a national alliance in INDIA, they decided to forego an 
# Alliance in punjab as they are the main rivals in the state. 
# Similarly the same happened in Alathur in Kerala. 
# Effectively, Congress lost 8 seats to other parties that it held earlier.
# The two independent candidates are from Punjab. 
# One of them (Sukhbeer Singh Khalsa) is a son of Benat Singh, one of the assassins of Indira Gandhi
# The other is Amritpal singh who has links to Waris Punjab De.
# Neither of them have ties to either of the alliances - INDIA or NDA
# Therefore, it beings the tally of lost seats to 10 for Congress.

Let us make some early models

# will first need to separate the data for just the INC candidates.
congcandi <- data2024 %>% filter(Party == 'INC')
# Let us see the general trends of how these candidates fared in their respective
# constituencies. 
table(congcandi$Position)

## 
##   1   2   3   4   5   9 
##  99 167  55   5   1   1

# INC really likes the silver medal it seems
# Let us now create a dummy for whether they have won or not
congcandi <- congcandi %>% mutate(goldmedal = ifelse(Position == 1,1,0))
# Dummy for if they won the last time
congcandi <- congcandi %>% mutate(former = ifelse(winner2019 == 'INC',1,0))
table(congcandi$goldmedal)

## 
##   0   1 
## 229  99

# Now let us make the first model that checks fow whether there is a difference 
# in how they fared in the reserved constituencies.
model1 <- glm(goldmedal ~ Constituency_Type + former + as.numeric(as.character(ruralPercent)), data = congcandi, family = 'binomial')
summary(model1)

## 
## Call:
## glm(formula = goldmedal ~ Constituency_Type + former + as.numeric(as.character(ruralPercent)), 
##     family = "binomial", data = congcandi)
## 
## Coefficients:
##                                         Estimate Std. Error z value Pr(>|z|)
## (Intercept)                            -1.508804   0.433212  -3.483 0.000496
## Constituency_TypeSC                     0.772074   0.354581   2.177 0.029449
## Constituency_TypeST                     0.542141   0.418537   1.295 0.195208
## former                                  2.351899   0.378499   6.214 5.17e-10
## as.numeric(as.character(ruralPercent))  0.001408   0.005905   0.238 0.811499
##                                           
## (Intercept)                            ***
## Constituency_TypeSC                    *  
## Constituency_TypeST                       
## former                                 ***
## as.numeric(as.character(ruralPercent))    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 385.89  on 312  degrees of freedom
## Residual deviance: 336.38  on 308  degrees of freedom
##   (15 observations deleted due to missingness)
## AIC: 346.38
## 
## Number of Fisher Scoring iterations: 4

Let us see if i thas done better in the states where it fought mostly on its

own or where it had support from the partners

library(ggplot2)

# Load necessary package
library(ggplot2)
density_data <- density(congcandi$Vote_Share_Percentage, na.rm = TRUE)
density_df <- data.frame(x = density_data$x, y = density_data$y)

# Plot density curve and shade the area under the curve less than 16.67
ggplot(density_df, aes(x = x, y = y)) +
  geom_area(data = subset(density_df, x <= 16.67), aes(y = y), fill = "red", alpha = 0.5) +
  geom_area(data = subset(density_df, x > 16.67), aes(y = y), fill = "blue", alpha = 0.5) +
  geom_vline(xintercept = 16.67, linetype = "dashed", color = "red", size = 1) +
  labs(title = "Density Curve of Vote Share Percentage INC 2024",
       x = "Vote Share Percentage",
       y = "Density") +
  theme_minimal()

## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.

# the absolute number is as follows

sum(congcandi$Vote_Share_Percentage < 16.7)

## [1] 51

# The party lost its deposit in 51 seats of 328 it contested. (15.55%)
# The figure was 184 out of 421 seats.(43.7%)

# The three big states where there were parties in the INDIA alliance contested
# Separately were West Bengal, Kerala, and Punjab
# Let us create a dummy for these states and see if it affected the performace
# of the INC in these states

congcandi <- congcandi %>% mutate(separate = ifelse(State_Name == c('West_Bengal','Kerala', 'Punjab'),1,0))

## Warning: There was 1 warning in `mutate()`.
## ℹ In argument: `separate = ifelse(...)`.
## Caused by warning in `State_Name == c("West_Bengal", "Kerala", "Punjab")`:
## ! longer object length is not a multiple of shorter object length

model2 <- glm(goldmedal ~ separate + Constituency_Type, data = congcandi, family = 'binomial')
summary(model2)

## 
## Call:
## glm(formula = goldmedal ~ separate + Constituency_Type, family = "binomial", 
##     data = congcandi)
## 
## Coefficients:
##                     Estimate Std. Error z value Pr(>|z|)    
## (Intercept)          -1.0009     0.1479  -6.770 1.29e-11 ***
## separate              0.8863     0.5540   1.600   0.1096    
## Constituency_TypeSC   0.5406     0.3242   1.667   0.0954 .  
## Constituency_TypeST   0.2670     0.3810   0.701   0.4836    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 401.74  on 327  degrees of freedom
## Residual deviance: 396.33  on 324  degrees of freedom
## AIC: 404.33
## 
## Number of Fisher Scoring iterations: 4

New TCPD dataset with all years

library(readr)
library(dplyr)
library(purrr)

# Read the data
TCPD_GE_All_States_2024_7_18 <- read_csv("TCPD_GE_All_States_2024-7-18.csv")

## Rows: 91669 Columns: 45
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (19): State_Name, Candidate, Sex, Party, Candidate_Type, Constituency_Na...
## dbl (20): Assembly_No, Constituency_No, Year, month, Poll_No, DelimID, Posit...
## lgl  (6): last_poll, Same_Constituency, Same_Party, Turncoat, Incumbent, Rec...
## 
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

compdata <- TCPD_GE_All_States_2024_7_18 %>%
  filter(Year %in% c(1991, 1996, 1998, 1999, 2004, 2009, 2014, 2019))

listdatabyyear <- split(compdata, compdata$Year)
names(listdatabyyear) <- paste0("data", names(listdatabyyear))
list2env(listdatabyyear, envir = .GlobalEnv)

## <environment: R_GlobalEnv>

years <- c(1991, 1996, 1998, 1999, 2004, 2009, 2014, 2019)
parties <- c("INC", "BJP")

# Apply the filter for each year and party
map2(
  rep(years, each = 2),
  rep(parties, times = 8),
  ~ assign(
    paste0("data", .x, .y),
    compdata %>% filter(Year == .x, Party == .y),
    envir = .GlobalEnv
  )
)

## [[1]]
## # A tibble: 492 × 45
##    State_Name   Assembly_No Constituency_No  Year month Poll_No DelimID Position
##    <chr>              <dbl>           <dbl> <dbl> <dbl>   <dbl>   <dbl>    <dbl>
##  1 Andaman_&_N…          10               1  1991     5       0       3        1
##  2 Andhra_Prad…          10               1  1991     5       0       3        1
##  3 Andhra_Prad…          10               2  1991     5       0       3        1
##  4 Andhra_Prad…          10               3  1991     5       0       3        1
##  5 Andhra_Prad…          10               4  1991     5       0       3        2
##  6 Andhra_Prad…          10               5  1991     5       0       3        1
##  7 Andhra_Prad…          10               6  1991     5       0       3        1
##  8 Andhra_Prad…          10               7  1991     5       0       3        2
##  9 Andhra_Prad…          10               8  1991     5       0       3        2
## 10 Andhra_Prad…          10               9  1991     5       0       3        2
## # ℹ 482 more rows
## # ℹ 37 more variables: Candidate <chr>, Sex <chr>, Party <chr>, Votes <dbl>,
## #   Candidate_Type <chr>, Valid_Votes <dbl>, Electors <dbl>,
## #   Constituency_Name <chr>, Constituency_Type <chr>, Sub_Region <chr>,
## #   N_Cand <dbl>, Turnout_Percentage <dbl>, Vote_Share_Percentage <dbl>,
## #   Deposit_Lost <chr>, Margin <dbl>, Margin_Percentage <dbl>, ENOP <dbl>,
## #   pid <chr>, Party_Type_TCPD <chr>, Party_ID <dbl>, last_poll <lgl>, …
## 
## [[2]]
## # A tibble: 472 × 45
##    State_Name   Assembly_No Constituency_No  Year month Poll_No DelimID Position
##    <chr>              <dbl>           <dbl> <dbl> <dbl>   <dbl>   <dbl>    <dbl>
##  1 Andaman_&_N…          10               1  1991     5       0       3        3
##  2 Andhra_Prad…          10               1  1991     5       0       3        4
##  3 Andhra_Prad…          10               3  1991     5       0       3        3
##  4 Andhra_Prad…          10               4  1991     5       0       3        3
##  5 Andhra_Prad…          10               5  1991     5       0       3        3
##  6 Andhra_Prad…          10               6  1991     5       0       3        3
##  7 Andhra_Prad…          10               7  1991     5       0       3        3
##  8 Andhra_Prad…          10               8  1991     5       0       3        3
##  9 Andhra_Prad…          10               9  1991     5       0       3        3
## 10 Andhra_Prad…          10              10  1991     5       0       3        3
## # ℹ 462 more rows
## # ℹ 37 more variables: Candidate <chr>, Sex <chr>, Party <chr>, Votes <dbl>,
## #   Candidate_Type <chr>, Valid_Votes <dbl>, Electors <dbl>,
## #   Constituency_Name <chr>, Constituency_Type <chr>, Sub_Region <chr>,
## #   N_Cand <dbl>, Turnout_Percentage <dbl>, Vote_Share_Percentage <dbl>,
## #   Deposit_Lost <chr>, Margin <dbl>, Margin_Percentage <dbl>, ENOP <dbl>,
## #   pid <chr>, Party_Type_TCPD <chr>, Party_ID <dbl>, last_poll <lgl>, …
## 
## [[3]]
## # A tibble: 532 × 45
##    State_Name   Assembly_No Constituency_No  Year month Poll_No DelimID Position
##    <chr>              <dbl>           <dbl> <dbl> <dbl>   <dbl>   <dbl>    <dbl>
##  1 Andaman_&_N…          11               1  1996     4       0       3        1
##  2 Andhra_Prad…          11               1  1996     4       0       3        3
##  3 Andhra_Prad…          11               2  1996     4       0       3        1
##  4 Andhra_Prad…          11               3  1996     4       0       3        2
##  5 Andhra_Prad…          11               4  1996     4       0       3        1
##  6 Andhra_Prad…          11               5  1996     4       0       3        2
##  7 Andhra_Prad…          11               6  1996     4       0       3        2
##  8 Andhra_Prad…          11               7  1996     4       0       3        2
##  9 Andhra_Prad…          11               8  1996     4       0       3        1
## 10 Andhra_Prad…          11               9  1996     4       0       3        1
## # ℹ 522 more rows
## # ℹ 37 more variables: Candidate <chr>, Sex <chr>, Party <chr>, Votes <dbl>,
## #   Candidate_Type <chr>, Valid_Votes <dbl>, Electors <dbl>,
## #   Constituency_Name <chr>, Constituency_Type <chr>, Sub_Region <chr>,
## #   N_Cand <dbl>, Turnout_Percentage <dbl>, Vote_Share_Percentage <dbl>,
## #   Deposit_Lost <chr>, Margin <dbl>, Margin_Percentage <dbl>, ENOP <dbl>,
## #   pid <chr>, Party_Type_TCPD <chr>, Party_ID <dbl>, last_poll <lgl>, …
## 
## [[4]]
## # A tibble: 473 × 45
##    State_Name   Assembly_No Constituency_No  Year month Poll_No DelimID Position
##    <chr>              <dbl>           <dbl> <dbl> <dbl>   <dbl>   <dbl>    <dbl>
##  1 Andaman_&_N…          11               1  1996     4       0       3        2
##  2 Andhra_Prad…          11               1  1996     4       0       3        4
##  3 Andhra_Prad…          11               2  1996     4       0       3        4
##  4 Andhra_Prad…          11               3  1996     4       0       3        4
##  5 Andhra_Prad…          11               4  1996     4       0       3        4
##  6 Andhra_Prad…          11               5  1996     4       0       3        4
##  7 Andhra_Prad…          11               6  1996     4       0       3        4
##  8 Andhra_Prad…          11               7  1996     4       0       3        4
##  9 Andhra_Prad…          11               8  1996     4       0       3        4
## 10 Andhra_Prad…          11               9  1996     4       0       3        4
## # ℹ 463 more rows
## # ℹ 37 more variables: Candidate <chr>, Sex <chr>, Party <chr>, Votes <dbl>,
## #   Candidate_Type <chr>, Valid_Votes <dbl>, Electors <dbl>,
## #   Constituency_Name <chr>, Constituency_Type <chr>, Sub_Region <chr>,
## #   N_Cand <dbl>, Turnout_Percentage <dbl>, Vote_Share_Percentage <dbl>,
## #   Deposit_Lost <chr>, Margin <dbl>, Margin_Percentage <dbl>, ENOP <dbl>,
## #   pid <chr>, Party_Type_TCPD <chr>, Party_ID <dbl>, last_poll <lgl>, …
## 
## [[5]]
## # A tibble: 479 × 45
##    State_Name   Assembly_No Constituency_No  Year month Poll_No DelimID Position
##    <chr>              <dbl>           <dbl> <dbl> <dbl>   <dbl>   <dbl>    <dbl>
##  1 Andaman_&_N…          12               1  1998     3       0       3        1
##  2 Andhra_Prad…          12               1  1998     3       0       3        3
##  3 Andhra_Prad…          12               2  1998     3       0       3        2
##  4 Andhra_Prad…          12               3  1998     3       0       3        2
##  5 Andhra_Prad…          12               4  1998     3       0       3        1
##  6 Andhra_Prad…          12               5  1998     3       0       3        2
##  7 Andhra_Prad…          12               6  1998     3       0       3        1
##  8 Andhra_Prad…          12               7  1998     3       0       3        3
##  9 Andhra_Prad…          12               8  1998     3       0       3        3
## 10 Andhra_Prad…          12               9  1998     3       0       3        2
## # ℹ 469 more rows
## # ℹ 37 more variables: Candidate <chr>, Sex <chr>, Party <chr>, Votes <dbl>,
## #   Candidate_Type <chr>, Valid_Votes <dbl>, Electors <dbl>,
## #   Constituency_Name <chr>, Constituency_Type <chr>, Sub_Region <chr>,
## #   N_Cand <dbl>, Turnout_Percentage <dbl>, Vote_Share_Percentage <dbl>,
## #   Deposit_Lost <chr>, Margin <dbl>, Margin_Percentage <dbl>, ENOP <dbl>,
## #   pid <chr>, Party_Type_TCPD <chr>, Party_ID <dbl>, last_poll <lgl>, …
## 
## [[6]]
## # A tibble: 389 × 45
##    State_Name   Assembly_No Constituency_No  Year month Poll_No DelimID Position
##    <chr>              <dbl>           <dbl> <dbl> <dbl>   <dbl>   <dbl>    <dbl>
##  1 Andaman_&_N…          12               1  1998     3       0       3        2
##  2 Andhra_Prad…          12               3  1998     3       0       3        3
##  3 Andhra_Prad…          12               4  1998     3       0       3        3
##  4 Andhra_Prad…          12               6  1998     3       0       3        3
##  5 Andhra_Prad…          12               7  1998     3       0       3        1
##  6 Andhra_Prad…          12               8  1998     3       0       3        1
##  7 Andhra_Prad…          12               9  1998     3       0       3        3
##  8 Andhra_Prad…          12              10  1998     3       0       3        3
##  9 Andhra_Prad…          12              11  1998     3       0       3        3
## 10 Andhra_Prad…          12              12  1998     3       0       3        3
## # ℹ 379 more rows
## # ℹ 37 more variables: Candidate <chr>, Sex <chr>, Party <chr>, Votes <dbl>,
## #   Candidate_Type <chr>, Valid_Votes <dbl>, Electors <dbl>,
## #   Constituency_Name <chr>, Constituency_Type <chr>, Sub_Region <chr>,
## #   N_Cand <dbl>, Turnout_Percentage <dbl>, Vote_Share_Percentage <dbl>,
## #   Deposit_Lost <chr>, Margin <dbl>, Margin_Percentage <dbl>, ENOP <dbl>,
## #   pid <chr>, Party_Type_TCPD <chr>, Party_ID <dbl>, last_poll <lgl>, …
## 
## [[7]]
## # A tibble: 453 × 45
##    State_Name   Assembly_No Constituency_No  Year month Poll_No DelimID Position
##    <chr>              <dbl>           <dbl> <dbl> <dbl>   <dbl>   <dbl>    <dbl>
##  1 Andaman_&_N…          13               1  1999     9       0       3        2
##  2 Andhra_Prad…          13               1  1999     9       0       3        2
##  3 Andhra_Prad…          13               2  1999     9       0       3        2
##  4 Andhra_Prad…          13               3  1999     9       0       3        1
##  5 Andhra_Prad…          13               4  1999     9       0       3        2
##  6 Andhra_Prad…          13               5  1999     9       0       3        2
##  7 Andhra_Prad…          13               6  1999     9       0       3        2
##  8 Andhra_Prad…          13               7  1999     9       0       3        2
##  9 Andhra_Prad…          13               8  1999     9       0       3        2
## 10 Andhra_Prad…          13               9  1999     9       0       3        2
## # ℹ 443 more rows
## # ℹ 37 more variables: Candidate <chr>, Sex <chr>, Party <chr>, Votes <dbl>,
## #   Candidate_Type <chr>, Valid_Votes <dbl>, Electors <dbl>,
## #   Constituency_Name <chr>, Constituency_Type <chr>, Sub_Region <chr>,
## #   N_Cand <dbl>, Turnout_Percentage <dbl>, Vote_Share_Percentage <dbl>,
## #   Deposit_Lost <chr>, Margin <dbl>, Margin_Percentage <dbl>, ENOP <dbl>,
## #   pid <chr>, Party_Type_TCPD <chr>, Party_ID <dbl>, last_poll <lgl>, …
## 
## [[8]]
## # A tibble: 339 × 45
##    State_Name   Assembly_No Constituency_No  Year month Poll_No DelimID Position
##    <chr>              <dbl>           <dbl> <dbl> <dbl>   <dbl>   <dbl>    <dbl>
##  1 Andaman_&_N…          13               1  1999     9       0       3        1
##  2 Andhra_Prad…          13               8  1999     9       0       3        1
##  3 Andhra_Prad…          13              10  1999     9       0       3        1
##  4 Andhra_Prad…          13              20  1999     9       0       3        1
##  5 Andhra_Prad…          13              29  1999     9       0       3        1
##  6 Andhra_Prad…          13              30  1999     9       0       3        2
##  7 Andhra_Prad…          13              31  1999     9       0       3        1
##  8 Andhra_Prad…          13              33  1999     9       0       3        1
##  9 Andhra_Prad…          13              37  1999     9       0       3        1
## 10 Arunachal_P…          13               2  1999     9       0       3        2
## # ℹ 329 more rows
## # ℹ 37 more variables: Candidate <chr>, Sex <chr>, Party <chr>, Votes <dbl>,
## #   Candidate_Type <chr>, Valid_Votes <dbl>, Electors <dbl>,
## #   Constituency_Name <chr>, Constituency_Type <chr>, Sub_Region <chr>,
## #   N_Cand <dbl>, Turnout_Percentage <dbl>, Vote_Share_Percentage <dbl>,
## #   Deposit_Lost <chr>, Margin <dbl>, Margin_Percentage <dbl>, ENOP <dbl>,
## #   pid <chr>, Party_Type_TCPD <chr>, Party_ID <dbl>, last_poll <lgl>, …
## 
## [[9]]
## # A tibble: 420 × 45
##    State_Name   Assembly_No Constituency_No  Year month Poll_No DelimID Position
##    <chr>              <dbl>           <dbl> <dbl> <dbl>   <dbl>   <dbl>    <dbl>
##  1 Andaman_&_N…          14               1  2004     4       0       3        1
##  2 Andhra_Prad…          14               1  2004     4       0       3        2
##  3 Andhra_Prad…          14               2  2004     4       0       3        1
##  4 Andhra_Prad…          14               3  2004     4       0       3        2
##  5 Andhra_Prad…          14               4  2004     4       0       3        1
##  6 Andhra_Prad…          14               6  2004     4       0       3        2
##  7 Andhra_Prad…          14               7  2004     4       0       3        1
##  8 Andhra_Prad…          14               8  2004     4       0       3        1
##  9 Andhra_Prad…          14               9  2004     4       0       3        1
## 10 Andhra_Prad…          14              10  2004     4       0       3        1
## # ℹ 410 more rows
## # ℹ 37 more variables: Candidate <chr>, Sex <chr>, Party <chr>, Votes <dbl>,
## #   Candidate_Type <chr>, Valid_Votes <dbl>, Electors <dbl>,
## #   Constituency_Name <chr>, Constituency_Type <chr>, Sub_Region <chr>,
## #   N_Cand <dbl>, Turnout_Percentage <dbl>, Vote_Share_Percentage <dbl>,
## #   Deposit_Lost <chr>, Margin <dbl>, Margin_Percentage <dbl>, ENOP <dbl>,
## #   pid <chr>, Party_Type_TCPD <chr>, Party_ID <dbl>, last_poll <lgl>, …
## 
## [[10]]
## # A tibble: 367 × 45
##    State_Name   Assembly_No Constituency_No  Year month Poll_No DelimID Position
##    <chr>              <dbl>           <dbl> <dbl> <dbl>   <dbl>   <dbl>    <dbl>
##  1 Andaman_&_N…          14               1  2004     4       0       3        2
##  2 Andhra_Prad…          14               8  2004     4       0       3        2
##  3 Andhra_Prad…          14              10  2004     4       0       3        2
##  4 Andhra_Prad…          14              19  2004     4       0       3        2
##  5 Andhra_Prad…          14              20  2004     4       0       3        2
##  6 Andhra_Prad…          14              30  2004     4       0       3        2
##  7 Andhra_Prad…          14              31  2004     4       0       3        2
##  8 Andhra_Prad…          14              33  2004     4       0       3        2
##  9 Andhra_Prad…          14              37  2004     4       0       3        2
## 10 Andhra_Prad…          14              41  2004     4       0       3        2
## # ℹ 357 more rows
## # ℹ 37 more variables: Candidate <chr>, Sex <chr>, Party <chr>, Votes <dbl>,
## #   Candidate_Type <chr>, Valid_Votes <dbl>, Electors <dbl>,
## #   Constituency_Name <chr>, Constituency_Type <chr>, Sub_Region <chr>,
## #   N_Cand <dbl>, Turnout_Percentage <dbl>, Vote_Share_Percentage <dbl>,
## #   Deposit_Lost <chr>, Margin <dbl>, Margin_Percentage <dbl>, ENOP <dbl>,
## #   pid <chr>, Party_Type_TCPD <chr>, Party_ID <dbl>, last_poll <lgl>, …
## 
## [[11]]
## # A tibble: 441 × 45
##    State_Name   Assembly_No Constituency_No  Year month Poll_No DelimID Position
##    <chr>              <dbl>           <dbl> <dbl> <dbl>   <dbl>   <dbl>    <dbl>
##  1 Andaman_&_N…          15               1  2009     4       0       4        2
##  2 Andhra_Prad…          15               1  2009     4       0       4        2
##  3 Andhra_Prad…          15               2  2009     4       0       4        1
##  4 Andhra_Prad…          15               3  2009     4       0       4        1
##  5 Andhra_Prad…          15               4  2009     4       0       4        1
##  6 Andhra_Prad…          15               5  2009     4       0       4        1
##  7 Andhra_Prad…          15               6  2009     4       0       4        2
##  8 Andhra_Prad…          15               7  2009     4       0       4        1
##  9 Andhra_Prad…          15               8  2009     4       0       4        1
## 10 Andhra_Prad…          15               9  2009     4       0       4        3
## # ℹ 431 more rows
## # ℹ 37 more variables: Candidate <chr>, Sex <chr>, Party <chr>, Votes <dbl>,
## #   Candidate_Type <chr>, Valid_Votes <dbl>, Electors <dbl>,
## #   Constituency_Name <chr>, Constituency_Type <chr>, Sub_Region <chr>,
## #   N_Cand <dbl>, Turnout_Percentage <dbl>, Vote_Share_Percentage <dbl>,
## #   Deposit_Lost <chr>, Margin <dbl>, Margin_Percentage <dbl>, ENOP <dbl>,
## #   pid <chr>, Party_Type_TCPD <chr>, Party_ID <dbl>, last_poll <lgl>, …
## 
## [[12]]
## # A tibble: 434 × 45
##    State_Name   Assembly_No Constituency_No  Year month Poll_No DelimID Position
##    <chr>              <dbl>           <dbl> <dbl> <dbl>   <dbl>   <dbl>    <dbl>
##  1 Andaman_&_N…          15               1  2009     4       0       4        1
##  2 Andhra_Prad…          15               1  2009     4       0       4        4
##  3 Andhra_Prad…          15               2  2009     4       0       4        4
##  4 Andhra_Prad…          15               3  2009     4       0       4        4
##  5 Andhra_Prad…          15               4  2009     4       0       4        4
##  6 Andhra_Prad…          15               5  2009     4       0       4        4
##  7 Andhra_Prad…          15               6  2009     4       0       4        4
##  8 Andhra_Prad…          15               7  2009     4       0       4        4
##  9 Andhra_Prad…          15               8  2009     4       0       4        2
## 10 Andhra_Prad…          15               9  2009     4       0       4        4
## # ℹ 424 more rows
## # ℹ 37 more variables: Candidate <chr>, Sex <chr>, Party <chr>, Votes <dbl>,
## #   Candidate_Type <chr>, Valid_Votes <dbl>, Electors <dbl>,
## #   Constituency_Name <chr>, Constituency_Type <chr>, Sub_Region <chr>,
## #   N_Cand <dbl>, Turnout_Percentage <dbl>, Vote_Share_Percentage <dbl>,
## #   Deposit_Lost <chr>, Margin <dbl>, Margin_Percentage <dbl>, ENOP <dbl>,
## #   pid <chr>, Party_Type_TCPD <chr>, Party_ID <dbl>, last_poll <lgl>, …
## 
## [[13]]
## # A tibble: 468 × 45
##    State_Name   Assembly_No Constituency_No  Year month Poll_No DelimID Position
##    <chr>              <dbl>           <dbl> <dbl> <dbl>   <dbl>   <dbl>    <dbl>
##  1 Andaman_&_N…          16               1  2014     4       0       4        2
##  2 Andhra_Prad…          16               1  2014     4       0       4        2
##  3 Andhra_Prad…          16               2  2014     4       0       4        2
##  4 Andhra_Prad…          16               3  2014     4       0       4        2
##  5 Andhra_Prad…          16               4  2014     4       0       4        2
##  6 Andhra_Prad…          16               5  2014     4       0       4        2
##  7 Andhra_Prad…          16               6  2014     4       0       4        2
##  8 Andhra_Prad…          16               7  2014     4       0       4        3
##  9 Andhra_Prad…          16               8  2014     4       0       4        2
## 10 Andhra_Prad…          16               9  2014     4       0       4        3
## # ℹ 458 more rows
## # ℹ 37 more variables: Candidate <chr>, Sex <chr>, Party <chr>, Votes <dbl>,
## #   Candidate_Type <chr>, Valid_Votes <dbl>, Electors <dbl>,
## #   Constituency_Name <chr>, Constituency_Type <chr>, Sub_Region <chr>,
## #   N_Cand <dbl>, Turnout_Percentage <dbl>, Vote_Share_Percentage <dbl>,
## #   Deposit_Lost <chr>, Margin <dbl>, Margin_Percentage <dbl>, ENOP <dbl>,
## #   pid <chr>, Party_Type_TCPD <chr>, Party_ID <dbl>, last_poll <lgl>, …
## 
## [[14]]
## # A tibble: 433 × 45
##    State_Name   Assembly_No Constituency_No  Year month Poll_No DelimID Position
##    <chr>              <dbl>           <dbl> <dbl> <dbl>   <dbl>   <dbl>    <dbl>
##  1 Andaman_&_N…          16               1  2014     4       0       4        1
##  2 Andhra_Prad…          16               3  2014     4       0       4        3
##  3 Andhra_Prad…          16               4  2014     4       0       4        3
##  4 Andhra_Prad…          16               6  2014     4       0       4        3
##  5 Andhra_Prad…          16               8  2014     4       0       4        1
##  6 Andhra_Prad…          16               9  2014     4       0       4        2
##  7 Andhra_Prad…          16              11  2014     4       0       4        3
##  8 Andhra_Prad…          16              14  2014     4       0       4        3
##  9 Andhra_Prad…          16              15  2014     4       0       4        3
## 10 Andhra_Prad…          16              21  2014     4       0       4        1
## # ℹ 423 more rows
## # ℹ 37 more variables: Candidate <chr>, Sex <chr>, Party <chr>, Votes <dbl>,
## #   Candidate_Type <chr>, Valid_Votes <dbl>, Electors <dbl>,
## #   Constituency_Name <chr>, Constituency_Type <chr>, Sub_Region <chr>,
## #   N_Cand <dbl>, Turnout_Percentage <dbl>, Vote_Share_Percentage <dbl>,
## #   Deposit_Lost <chr>, Margin <dbl>, Margin_Percentage <dbl>, ENOP <dbl>,
## #   pid <chr>, Party_Type_TCPD <chr>, Party_ID <dbl>, last_poll <lgl>, …
## 
## [[15]]
## # A tibble: 422 × 45
##    State_Name   Assembly_No Constituency_No  Year month Poll_No DelimID Position
##    <chr>              <dbl>           <dbl> <dbl> <dbl>   <dbl>   <dbl>    <dbl>
##  1 Andaman_&_N…          17               1  2019     4       0       4        1
##  2 Andhra_Prad…          17               1  2019     4       0       4        6
##  3 Andhra_Prad…          17               2  2019     4       0       4        5
##  4 Andhra_Prad…          17               3  2019     4       0       4        5
##  5 Andhra_Prad…          17               4  2019     4       0       4        6
##  6 Andhra_Prad…          17               5  2019     4       0       4        6
##  7 Andhra_Prad…          17               6  2019     4       0       4        6
##  8 Andhra_Prad…          17               7  2019     4       0       4        6
##  9 Andhra_Prad…          17               8  2019     4       0       4        5
## 10 Andhra_Prad…          17               9  2019     4       0       4        4
## # ℹ 412 more rows
## # ℹ 37 more variables: Candidate <chr>, Sex <chr>, Party <chr>, Votes <dbl>,
## #   Candidate_Type <chr>, Valid_Votes <dbl>, Electors <dbl>,
## #   Constituency_Name <chr>, Constituency_Type <chr>, Sub_Region <chr>,
## #   N_Cand <dbl>, Turnout_Percentage <dbl>, Vote_Share_Percentage <dbl>,
## #   Deposit_Lost <chr>, Margin <dbl>, Margin_Percentage <dbl>, ENOP <dbl>,
## #   pid <chr>, Party_Type_TCPD <chr>, Party_ID <dbl>, last_poll <lgl>, …
## 
## [[16]]
## # A tibble: 437 × 45
##    State_Name   Assembly_No Constituency_No  Year month Poll_No DelimID Position
##    <chr>              <dbl>           <dbl> <dbl> <dbl>   <dbl>   <dbl>    <dbl>
##  1 Andaman_&_N…          17               1  2019     4       0       4        2
##  2 Andhra_Prad…          17               1  2019     4       0       4        5
##  3 Andhra_Prad…          17               2  2019     4       0       4        6
##  4 Andhra_Prad…          17               3  2019     4       0       4        6
##  5 Andhra_Prad…          17               4  2019     4       0       4        4
##  6 Andhra_Prad…          17               5  2019     4       0       4        5
##  7 Andhra_Prad…          17               6  2019     4       0       4        5
##  8 Andhra_Prad…          17               7  2019     4       0       4        5
##  9 Andhra_Prad…          17               8  2019     4       0       4        6
## 10 Andhra_Prad…          17               9  2019     4       0       4        5
## # ℹ 427 more rows
## # ℹ 37 more variables: Candidate <chr>, Sex <chr>, Party <chr>, Votes <dbl>,
## #   Candidate_Type <chr>, Valid_Votes <dbl>, Electors <dbl>,
## #   Constituency_Name <chr>, Constituency_Type <chr>, Sub_Region <chr>,
## #   N_Cand <dbl>, Turnout_Percentage <dbl>, Vote_Share_Percentage <dbl>,
## #   Deposit_Lost <chr>, Margin <dbl>, Margin_Percentage <dbl>, ENOP <dbl>,
## #   pid <chr>, Party_Type_TCPD <chr>, Party_ID <dbl>, last_poll <lgl>, …

# Remove byelection seats for INC1991
data1991INC <- head(data1991INC, n = nrow(data1991INC) - 5)
 
# For 1991
mobiINC1991 <- sum(data1991INC$Votes) / sum(data1991INC$Electors)
mobiBJP1991 <- sum(data1991BJP$Votes) / sum(data1991BJP$Electors)

# For 1996
mobiINC1996 <- sum(data1996INC$Votes) / sum(data1996INC$Electors)
mobiBJP1996 <- sum(data1996BJP$Votes) / sum(data1996BJP$Electors)

# For 1998
mobiINC1998 <- sum(data1998INC$Votes) / sum(data1998INC$Electors)
mobiBJP1998 <- sum(data1998BJP$Votes) / sum(data1998BJP$Electors)

# For 1999
mobiINC1999 <- sum(data1999INC$Votes) / sum(data1999INC$Electors)
mobiBJP1999 <- sum(data1999BJP$Votes) / sum(data1999BJP$Electors)
 
# For 2004
mobiINC2004 <- sum(data2004INC$Votes)/sum(data2004INC$Electors)
mobiBJP2004 <- sum(data2004BJP$Votes)/sum(data2004BJP$Electors)

#For 2009
mobiINC2009 <- sum(data2009INC$Votes)/sum(data2009INC$Electors)
mobiBJP2009 <- sum(data2009BJP$Votes)/sum(data2009BJP$Electors)

#For 2014
mobiINC2014 <- sum(data2014INC$Votes)/sum(data2014INC$Electors)
mobiBJP2014 <- sum(data2014BJP$Votes)/sum(data2014BJP$Electors)

#For 2019
mobiINC2019 <- sum(data2019INC$Votes)/sum(data2019INC$Electors)
mobiBJP2019 <- sum(data2019BJP$Votes)/sum(data2019BJP$Electors)


data2024BJP <- data2024 %>% filter(Party == 'BJP')
data2024INC <- data2024 %>% filter(Party == 'INC')

mobiINC2024 <- sum(data2024INC$Votes)/sum(data2024INC$electors_2024)
mobiBJP2024 <- sum(data2024BJP$Votes, na.rm = TRUE)/sum(data2024BJP$electors_2024, na.rm = TRUE)


library(purrr)

if (exists("data1991INC") && exists("data1991BJP") && 
    exists("data1996INC") && exists("data1996BJP") &&
    exists("data1998INC") && exists("data1998BJP") &&
    exists("data1999INC") && exists("data1999BJP") &&
    exists("data2004INC") && exists("data2004BJP") &&
    exists("data2009INC") && exists("data2009BJP") &&
    exists("data2014INC") && exists("data2014BJP") &&
    exists("data2019INC") && exists("data2019BJP")) 
  
datasets <- list(
  data1991INC = data1991INC, data1991BJP = data1991BJP,
  data1996INC = data1996INC, data1996BJP = data1996BJP,
  data1998INC = data1998INC, data1998BJP = data1998BJP,
  data1999INC = data1999INC, data1999BJP = data1999BJP,
  data2004INC = data2004INC, data2004BJP = data2004BJP,
  data2009INC = data2009INC, data2009BJP = data2009BJP,
  data2014INC = data2014INC, data2014BJP = data2014BJP,
  data2019INC = data2019INC, data2019BJP = data2019BJP
)

missing_datasets <- names(datasets)[!sapply(datasets, function(x) !is.null(x))]
if (length(missing_datasets) > 0) {
  stop(paste("The following datasets are missing:", paste(missing_datasets, collapse = ", ")))
}

mobi_values <- map2(
  names(datasets),
  rep(c("INC", "BJP"), length.out = length(datasets)),
  ~ sum(datasets[[.x]]$Votes) / sum(datasets[[.x]]$Electors)
)

# Let us create individual mobilization score for each district. 

datasets <- datasets %>% map(~ mutate(.x, mobi = Votes/Electors))

#update it in the env
list2env(datasets, envir = .GlobalEnv)

## <environment: R_GlobalEnv>

# now if we find the values by average instead of the one used above:
calc_mobi_avg <- function(df) {
  sum(df$mobi) / nrow(df)
}

mobi_avg <- datasets %>% map(calc_mobi_avg)

print(mobi_avg)

## $data1991INC
## [1] 0.215566
## 
## $data1991BJP
## [1] 0.144637
## 
## $data1996INC
## [1] 0.1735236
## 
## $data1996BJP
## [1] 0.1309606
## 
## $data1998INC
## [1] 0.1829057
## 
## $data1998BJP
## [1] 0.2155257
## 
## $data1999INC
## [1] 0.2028397
## 
## $data1999BJP
## [1] 0.2216685
## 
## $data2004INC
## [1] 0.2047149
## 
## $data2004BJP
## [1] 0.1871222
## 
## $data2009INC
## [1] 0.2102997
## 
## $data2009BJP
## [1] 0.1346299
## 
## $data2014INC
## [1] 0.1562795
## 
## $data2014BJP
## [1] 0.2556924
## 
## $data2019INC
## [1] 0.1733923
## 
## $data2019BJP
## [1] 0.3065449

# let us calculate the winning differential over the time:

list2env(lapply(listdatabyyear, function(df) {
     df %>%
         group_by(Constituency_Name) %>%
         slice_max(Votes, n = 2) %>%
         ungroup()
 }), envir = .GlobalEnv)

## <environment: R_GlobalEnv>

 # the fucntion can be as follows. It first removes every candidate except for
# the top two and then finds the difference. 
 calculate_avg_differential <- function(df) {
     df %>%
         group_by(Constituency_Name) %>%
         arrange(desc(Vote_Share_Percentage)) %>%
         slice(1:2) %>%
         summarise(difference = Vote_Share_Percentage[1] - Vote_Share_Percentage[2], .groups = 'drop') %>%
         summarise(avg_differential = mean(difference, na.rm = TRUE))
 }
 
avg_differentials <- lapply(listdatabyyear, calculate_avg_differential)
 
names(avg_differentials) <- names(listdatabyyear)
 
avg_differentials_values <- sapply(avg_differentials, function(df) df$avg_differential)
 
avg_differentials_values

##  data1991  data1996  data1998  data1999  data2004  data2009  data2014  data2019 
## 14.126583 12.224019 10.283142 10.213296 12.171204  9.702722 15.007172 17.183050

# Function to calculate the median winning differential for a dataset
calc_median_differential <- function(df) {
  df %>%
    group_by(Constituency_Name) %>%
    arrange(desc(Vote_Share_Percentage)) %>%
    slice(1:2) %>%
    summarise(difference = Vote_Share_Percentage[1] - Vote_Share_Percentage[2], .groups = 'drop') %>%
    summarise(median_differential = median(difference, na.rm = TRUE))
}

median_differentials <- lapply(listdatabyyear, calc_median_differential)

names(median_differentials) <- names(listdatabyyear)

median_differentials_values <- sapply(median_differentials, function(df) df$median_differential)

median_differentials_values 

## data1991 data1996 data1998 data1999 data2004 data2009 data2014 data2019 
##   10.580    9.370    7.630    7.890    9.940    6.945   13.540   15.060

average_differential2024 <- data2024 %>%
  arrange(Constituency_Name, Position) %>%              
  group_by(Constituency_Name) %>%                         
  summarise(differential = {
    if(n() >= 2) {
      first_candidate <- Vote_Share_Percentage[1]
      second_candidate <- Vote_Share_Percentage[2]
      first_candidate - second_candidate
    } else {
      NA
    }
  }) %>%
  summarise(average_differential = mean(differential, na.rm = TRUE))  

print(average_differential2024)

## # A tibble: 1 × 1
##   average_differential
##                  <dbl>
## 1                 13.5

# Calculating the median differential for 2024
median_differential2024 <- data2024 %>%
  arrange(Constituency_Name, Position) %>%                
  group_by(Constituency_Name) %>%                        
  summarise(differential = {
    if(n() >= 2) {
      first_candidate <- Vote_Share_Percentage[1]
      second_candidate <- Vote_Share_Percentage[2]
      first_candidate - second_candidate
    } else {
      NA 
    }
  }) %>%
  summarise(median_differential = median(differential, na.rm = TRUE))  # Compute the median differential

print(median_differential2024)

## # A tibble: 1 × 1
##   median_differential
##                 <dbl>
## 1                10.3

Now let us find the performance of the top three parties

# Function to calculate the top three parties by number of seats won in a single dataset
 calc_top_three_parties <- function(df) {
     df %>%
         # Ensure 'Votes' is numeric
         mutate(Votes = as.numeric(Votes)) %>%
         # Find the winner in each constituency
         group_by(Constituency_Name) %>%
         slice_max(order_by = Votes, n = 1, with_ties = FALSE) %>%
         ungroup() %>%
         # Count the number of seats won by each party
         count(Party, sort = TRUE) %>%
         # Get the top three parties
         slice_max(order_by = n, n = 3) %>%
         ungroup()
 }
 
 # Apply the function to each dataset in the list and add a 'Year' column
 top_parties_by_year <- lapply(seq_along(listdatabyyear), function(i) {
     df <- calc_top_three_parties(listdatabyyear[[i]])
     df$Year <- names(listdatabyyear)[i]
     return(df)
 })
 
 # Combine all results into a single dataframe
 combined_top_parties <- bind_rows(top_parties_by_year)
 
 # Print the combined dataframe
 print(combined_top_parties)

## # A tibble: 24 × 3
##    Party     n Year    
##    <chr> <int> <chr>   
##  1 INC     232 data1991
##  2 BJP     118 data1991
##  3 JD       58 data1991
##  4 BJP     159 data1996
##  5 INC     139 data1996
##  6 JD       45 data1996
##  7 BJP     180 data1998
##  8 INC     141 data1998
##  9 CPM      32 data1998
## 10 BJP     182 data1999
## # ℹ 14 more rows

pivoted_top_parties <- combined_top_parties %>%
     pivot_wider(names_from = Party, values_from = n, values_fill = list(n = 0))

# will have to look up the vote share online as the data is doing weird things

write.csv(pivoted_top_parties, file = "pivot1991-2019.csv")

write_excel_csv(pivoted_top_parties, file = "pivoted_top_parties.csv")