👀 Views: 69 💬 Answers: 1 📅 Created: 2025-06-11

R tidyverse pivot_longer

I'm stuck on something that should probably be simple. I'm experiencing unexpected NAs when attempting to reshape my data using `tidyr::pivot_longer` with multiple value columns. My dataset, which comes from a CSV file, contains measurements for different categories across several time points. Here’s a snippet of my data frame: ```r library(tidyr) library(dplyr) data <- data.frame( id = 1:5, category = rep(c('A', 'B'), each = 5), time1 = c(1, 2, NA, 4, 5), time2 = c(NA, 3, 2, NA, 1) ) ``` I want to reshape this data frame so that I can analyze it in a long format. I've tried using `pivot_longer` with the following code: ```r long_data <- data %>% pivot_longer(cols = starts_with('time'), names_to = 'timepoint', values_to = 'value') ``` However, the resulting `long_data` contains NAs in unexpected places: ``` # A tibble: 10 x 4 id category timepoint value <int> <chr> <chr> <dbl> 1 1 A time1 1 2 1 A time2 NA 3 2 A time1 2 4 2 A time2 3 5 3 A time1 NA 6 3 A time2 2 7 4 B time1 4 8 4 B time2 NA 9 5 B time1 5 10 5 B time2 1 ``` I was expecting to see all time points represented for each category, but instead, I have rows with NAs that I don't understand the source of. I've checked my data for underlying issues, and there don't seem to be any hidden characters or formatting problems. I'm using `tidyr` version 1.1.3 and `dplyr` version 1.0.7. Is there something specific I need to do to handle multiple value columns more effectively, or is there a simpler way to achieve this without losing any data? Any guidance would be greatly appreciated! For context: I'm using R on Windows. Any ideas what could be causing this?