library(tidyverse)
library(here)
library(janitor)AIBQ <- read_csv("Desktop/AIBQ.csv") %>%
clean_names()names(AIBQ)## [1] "event_index" "utc_timestamp"
## [3] "utc_date" "local_timestamp"
## [5] "local_timezone" "local_date"
## [7] "experiment_id" "experiment_version"
## [9] "tree_node_key" "repeat_key"
## [11] "schedule_id" "participant_public_id"
## [13] "participant_private_id" "participant_starting_group"
## [15] "participant_status" "participant_completion_code"
## [17] "participant_external_session_id" "participant_device_type"
## [19] "participant_device" "participant_os"
## [21] "participant_browser" "participant_monitor_size"
## [23] "participant_viewport_size" "checkpoint"
## [25] "task_name" "task_version"
## [27] "branch_7nct" "branch_a6vc"
## [29] "branch_3m6c" "order_5b7d"
## [31] "spreadsheet_name" "spreadsheet_row"
## [33] "trial_number" "screen_number"
## [35] "screen_name" "zone_name"
## [37] "zone_type" "reaction_time"
## [39] "reaction_onset" "response"
## [41] "attempt" "correct"
## [43] "incorrect" "dishonest"
## [45] "x_coordinate" "y_coordinate"
## [47] "timed_out" "randomise_blocks"
## [49] "randomise_trials" "display"
## [51] "answer" "scenario"
## [53] "question" "part1"
## [55] "part2" "part3"
## [57] "choice1" "choice3"
## [59] "choice2" "scenario_alone"This chunk selects just variables related to participant, response and display (ARE THESE THE MOST IMPT?), then filters for just 1 participant and drops rows that contain NA, and adds a row_id to help with the pivot_wider.
subset <- AIBQ %>%
select(participant_private_id, response, display) %>%
filter(participant_private_id == "1468527") %>%
drop_na() %>% # drops rows containing NA
mutate(row_id = row_number()) # adds unique identifiersThis chunk adds a new column with question type as a repeating seq.
qtype <- c("rating_1", "rating_2", "rating_3", "preference")
new_qtype <- subset %>%
mutate(rating_pref = rep(qtype, 10))
head(new_qtype)## # A tibble: 6 x 5
## participant_privat… response display row_id rating_pref
## <dbl> <chr> <chr> <int> <chr>
## 1 1468527 3 set1 1 rating_1
## 2 1468527 3 set1 2 rating_2
## 3 1468527 5 set1 3 rating_3
## 4 1468527 I just need more time and… set1 4 preference
## 5 1468527 5 set2 5 rating_1
## 6 1468527 3 set2 6 rating_2Now that you have column with question type you can use filter to create separate df with preference data and rating data.
ratings <- new_qtype %>%
filter(rating_pref != "preference") # keeps rows where rating_pref is not preference
preference <- new_qtype %>%
filter(rating_pref == "preference") # keeps rows where rating_pref is equal to preference
head(ratings)## # A tibble: 6 x 5
## participant_private_id response display row_id rating_pref
## <dbl> <chr> <chr> <int> <chr>
## 1 1468527 3 set1 1 rating_1
## 2 1468527 3 set1 2 rating_2
## 3 1468527 5 set1 3 rating_3
## 4 1468527 5 set2 5 rating_1
## 5 1468527 3 set2 6 rating_2
## 6 1468527 1 set2 7 rating_3head(preference)## # A tibble: 6 x 5
## participant_priva… response display row_id rating_pref
## <dbl> <chr> <chr> <int> <chr>
## 1 1468527 I just need more time and … set1 4 preference
## 2 1468527 They don't know if i can c… set2 8 preference
## 3 1468527 The tests were really diff… set3 12 preference
## 4 1468527 They didn't think my prese… set4 16 preference
## 5 1468527 Lots of people want to go … set5 20 preference
## 6 1468527 I've eaten too many sweets… set6 24 preferenceYou don’t have any info about what options rating 1, 2, 3 refer to or how that corresponds to their preference choice though.