Skip to content

Commit

Permalink
created 2025 pre-meds assessment
Browse files Browse the repository at this point in the history
  • Loading branch information
@jamiecmontgomery
jamiecmontgomery committed Aug 13, 2024
1 parent 3ffa8e0 commit 45d8902
Show file tree
Hide file tree
Showing 101 changed files with 22,683 additions and 86 deletions.
52 changes: 31 additions & 21 deletions R/q18_prob_dist.R
View file
Original file line number Diff line number Diff line change
Expand Up @@ -81,6 +81,8 @@ clean_q18b_prob_dist_terms <- function(PLO_data_clean){
post_meds <- df |>
filter(timepoint == "Post-MEDS")

if("Post-MEDS" %in% df$timepoint){

#................add 0s where missing (post-MEDS)................
for (i in 1:length(options)){

Expand Down Expand Up @@ -122,6 +124,14 @@ clean_q18b_prob_dist_terms <- function(PLO_data_clean){
perc_label = paste0(percentage, "%")) |>
mutate(xvar = prob_dist_terms)


} else{
post_meds <- post_meds %>%
mutate(total_respondents = 0,
percentage = 0,
prob_dist_terms = NA)
}

##~~~~~~~~~~~~~~~~~~~~~~~
## ~ recombine dfs ----
##~~~~~~~~~~~~~~~~~~~~~~~
Expand Down Expand Up @@ -190,7 +200,7 @@ clean_q18b_FULLY_CORRECT <- function(PLO_data_clean){
##~~~~~~~~~~~~~~~~~~

#....total respondents that continued onto answer question 18b...
total_pre_resp <- meds2024_before_clean |>
total_pre_resp <- meds2025_before_clean |>
select(prob_dist) |>
group_by(prob_dist) |>
# filter(prob_dist != "1 (never heard of it)") |> # include this if you want % calculation to be out of only students who advanced to this question
Expand All @@ -216,32 +226,32 @@ clean_q18b_FULLY_CORRECT <- function(PLO_data_clean){
##~~~~~~~~~~~~~~~~~~~

#....total respondents that continued onto answer question 18b...
total_post_resp <- meds2024_after_clean |>
select(prob_dist) |>
group_by(prob_dist) |>
# filter(prob_dist != "1 (never heard of it)") |> # include this if you want % calculation to be out of only students who advanced to this question
count() |>
ungroup() |>
summarize(n = sum(n)) |>
pull()
# total_post_resp <- meds2025_after_clean |>
# select(prob_dist) |>
# group_by(prob_dist) |>
# # filter(prob_dist != "1 (never heard of it)") |> # include this if you want % calculation to be out of only students who advanced to this question
# count() |>
# ungroup() |>
# summarize(n = sum(n)) |>
# pull()

post_meds <- df |>
filter(timepoint == "Post-MEDS") |>
group_by(correctness) |>
count() |>
# summarize(total = sum(n)) |>
# ungroup() |>
mutate(timepoint = rep("Post-MEDS")) |>
mutate(total_respondents = total_post_resp,
percentage = round((n/total_respondents)*100, 1),
perc_label = paste0(percentage, "%")) |>
mutate(perc_label_long = paste0(perc_label, "\n(", n, "/", total_respondents, " respondents)"))
# post_meds <- df |>
# filter(timepoint == "Post-MEDS") |>
# group_by(correctness) |>
# count() |>
# # summarize(total = sum(n)) |>
# # ungroup() |>
# mutate(timepoint = rep("Post-MEDS")) |>
# mutate(total_respondents = total_post_resp,
# percentage = round((n/total_respondents)*100, 1),
# perc_label = paste0(percentage, "%")) |>
# mutate(perc_label_long = paste0(perc_label, "\n(", n, "/", total_respondents, " respondents)"))

##~~~~~~~~~~~~~~~~~~~~~~~
## ~ recombine dfs ----
##~~~~~~~~~~~~~~~~~~~~~~~

all_q18b_fully_correct <- rbind(pre_meds, post_meds) |>
all_q18b_fully_correct <- pre_meds |> #rbind(pre_meds, post_meds) |>
filter(correctness == "yes") |>
mutate(timepoint = fct_relevel(timepoint, c("Pre-MEDS", "Post-MEDS")))

Expand Down
68 changes: 39 additions & 29 deletions R/q21_spatial_data.R
View file
Original file line number Diff line number Diff line change
Expand Up @@ -81,6 +81,8 @@ clean_q21b_rep_spatial <- function(PLO_data_clean){
post_meds <- df |>
filter(timepoint == "Post-MEDS")

if("Post-MEDS" %in% df$timepoint){

#................add 0s where missing (post-MEDS)................
for (i in 1:length(options)){

Expand Down Expand Up @@ -122,6 +124,14 @@ clean_q21b_rep_spatial <- function(PLO_data_clean){
perc_label = paste0(percentage, "%")) |>
mutate(xvar = rep_spatial_data)

} else{
post_meds <- post_meds %>%
mutate(total_respondents = 0,
percentage = 0,
perc_label = NA,
rep_spatial_data = NA)
}

##~~~~~~~~~~~~~~~~~~~~~~~
## ~ recombine dfs ----
##~~~~~~~~~~~~~~~~~~~~~~~
Expand Down Expand Up @@ -193,7 +203,7 @@ clean_q21b_FULLY_CORRECT <- function(PLO_data_clean){
##~~~~~~~~~~~~~~~~~~

#....total respondents that continued onto answer question 18b...
total_pre_resp <- meds2024_before_clean |>
total_pre_resp <- meds2025_before_clean |>
select(rep_spatial_data) |>
group_by(rep_spatial_data) |>
filter(rep_spatial_data != "1 (never worked with it before)") |>
Expand All @@ -217,30 +227,30 @@ clean_q21b_FULLY_CORRECT <- function(PLO_data_clean){
##~~~~~~~~~~~~~~~~~~~

#....total respondents that continued onto answer question 18b...
total_post_resp <- meds2024_after_clean |>
select(rep_spatial_data) |>
group_by(rep_spatial_data) |>
filter(rep_spatial_data != "1 (never worked with it before)") |>
count() |>
ungroup() |>
summarize(n = sum(n)) |>
pull()

post_meds <- df |>
filter(timepoint == "Post-MEDS") |>
group_by(correctness) |>
count() |>
ungroup() |>
mutate(timepoint = rep("Post-MEDS")) |>
mutate(total_respondents = sum(n), #total_post_resp,
percentage = round((n/total_respondents)*100, 1),
perc_label = paste0(percentage, "%"))
# total_post_resp <- meds2025_after_clean |>
# select(rep_spatial_data) |>
# group_by(rep_spatial_data) |>
# filter(rep_spatial_data != "1 (never worked with it before)") |>
# count() |>
# ungroup() |>
# summarize(n = sum(n)) |>
# pull()
#
# post_meds <- df |>
# filter(timepoint == "Post-MEDS") |>
# group_by(correctness) |>
# count() |>
# ungroup() |>
# mutate(timepoint = rep("Post-MEDS")) |>
# mutate(total_respondents = sum(n), #total_post_resp,
# percentage = round((n/total_respondents)*100, 1),
# perc_label = paste0(percentage, "%"))

##~~~~~~~~~~~~~~~~~~~~~~~
## ~ recombine dfs ----
##~~~~~~~~~~~~~~~~~~~~~~~

all_q21b_fully_correct <- rbind(pre_meds, post_meds) |>
all_q21b_fully_correct <- pre_meds |> #rbind(pre_meds, post_meds) |>
filter(correctness == "yes") |>
mutate(timepoint = fct_relevel(timepoint, c("Pre-MEDS", "Post-MEDS"))) |>
mutate(perc_label_long = paste0(perc_label, "\n(", n, "/", total_respondents, " respondents)"))
Expand Down Expand Up @@ -314,20 +324,20 @@ clean_q21c_vec_ras <- function(PLO_data_clean){
##~~~~~~~~~~~~~~~~~~~

#........separate post-MEDS (to add 0s for missing cats).........
post_meds <- df |>
filter(timepoint == "Post-MEDS") |>
# drop_na() |> # include this if you want % calculation to be out of only students who advanced to this question
mutate(total_respondents = sum(n),
percentage = round((n/total_respondents)*100, 1),
perc_label = paste0(percentage, "%")) |>
mutate(xvar = vec_or_ras) |>
mutate(perc_label_long = paste0(perc_label, "\n(", n, "/", total_respondents, " respondents)"))
# post_meds <- df |>
# filter(timepoint == "Post-MEDS") |>
# # drop_na() |> # include this if you want % calculation to be out of only students who advanced to this question
# mutate(total_respondents = sum(n),
# percentage = round((n/total_respondents)*100, 1),
# perc_label = paste0(percentage, "%")) |>
# mutate(xvar = vec_or_ras) |>
# mutate(perc_label_long = paste0(perc_label, "\n(", n, "/", total_respondents, " respondents)"))

#~~~~~~~~~~~~~~~~~~~~~~~
## ~ recombine dfs ----
##~~~~~~~~~~~~~~~~~~~~~~~

all_q21c_data <- rbind(pre_meds, post_meds) |>
all_q21c_data <- pre_meds |> #rbind(pre_meds, post_meds) |>
filter(vec_or_ras == "vector")

return(all_q21c_data)
Expand Down
55 changes: 33 additions & 22 deletions R/q26_TrainValidSplit.R
View file
Original file line number Diff line number Diff line change
Expand Up @@ -85,6 +85,8 @@ clean_q26c_mod_perf <- function(PLO_data_clean){
post_meds <- df |>
filter(timepoint == "Post-MEDS")

if("Post-MEDS" %in% df$timepoint){

#................add 0s where missing (post-MEDS)................
for (i in 1:length(options)){

Expand Down Expand Up @@ -126,6 +128,15 @@ clean_q26c_mod_perf <- function(PLO_data_clean){
perc_label = paste0(percentage, "%")) |>
mutate(xvar = learning_from_model)

} else{
post_meds <- post_meds %>%
mutate(total_respondents = 0,
percentage = 0,
perc_label = NA,
learning_from_model = NA)
}


##~~~~~~~~~~~~~~~~~~~~~~~
## ~ recombine dfs ----
##~~~~~~~~~~~~~~~~~~~~~~~
Expand Down Expand Up @@ -203,7 +214,7 @@ clean_q26c_FULLY_CORRECT <- function(PLO_data_clean){
##~~~~~~~~~~~~~~~~~~

#....total respondents that continued onto answer question 18b...
total_pre_resp <- meds2024_before_clean |>
total_pre_resp <- meds2025_before_clean |>
select(learning_from_model) |>
group_by(learning_from_model) |>
# filter(learning_from_model != "1 (never heard of it)") |> # include this if you want % calculation to be out of only students who advanced to this question
Expand All @@ -229,32 +240,32 @@ clean_q26c_FULLY_CORRECT <- function(PLO_data_clean){
##~~~~~~~~~~~~~~~~~~~

#....total respondents that continued onto answer question 18b...
total_post_resp <- meds2024_after_clean |>
select(learning_from_model) |>
group_by(learning_from_model) |>
# filter(learning_from_model != "1 (never heard of it)") |> # include this if you want % calculation to be out of only students who advanced to this question
count() |>
ungroup() |>
summarize(n = sum(n)) |>
pull()

post_meds <- df |>
filter(timepoint == "Post-MEDS") |>
group_by(correctness) |>
count() |>
# summarize(total = sum(n)) |>
# ungroup() |>
mutate(timepoint = rep("Post-MEDS")) |>
mutate(total_respondents = total_post_resp,
percentage = round((n/total_respondents)*100, 1),
perc_label = paste0(percentage, "%")) |>
mutate(perc_label_long = paste0(perc_label, "\n(", n, "/", total_respondents, " respondents)"))
# total_post_resp <- meds2025_after_clean |>
# select(learning_from_model) |>
# group_by(learning_from_model) |>
# # filter(learning_from_model != "1 (never heard of it)") |> # include this if you want % calculation to be out of only students who advanced to this question
# count() |>
# ungroup() |>
# summarize(n = sum(n)) |>
# pull()
#
# post_meds <- df |>
# filter(timepoint == "Post-MEDS") |>
# group_by(correctness) |>
# count() |>
# # summarize(total = sum(n)) |>
# # ungroup() |>
# mutate(timepoint = rep("Post-MEDS")) |>
# mutate(total_respondents = total_post_resp,
# percentage = round((n/total_respondents)*100, 1),
# perc_label = paste0(percentage, "%")) |>
# mutate(perc_label_long = paste0(perc_label, "\n(", n, "/", total_respondents, " respondents)"))

##~~~~~~~~~~~~~~~~~~~~~~~
## ~ recombine dfs ----
##~~~~~~~~~~~~~~~~~~~~~~~

all_q26c_data <- rbind(pre_meds, post_meds) |>
all_q26c_data <- pre_meds |> #rbind(pre_meds, post_meds) |>
filter(correctness == "yes") |>
mutate(timepoint = fct_relevel(timepoint, c("Pre-MEDS", "Post-MEDS")))

Expand Down
11 changes: 11 additions & 0 deletions R/q31_lang.R
View file
Original file line number Diff line number Diff line change
Expand Up @@ -70,6 +70,8 @@ clean_q31_lang <- function(PLO_data_clean){
post_meds <- df |>
filter(timepoint == "Post-MEDS")

if("Post-MEDS" %in% df$timepoint){

#................add 0s where missing (post-MEDS)................
for (i in 1:length(options)){

Expand Down Expand Up @@ -102,6 +104,15 @@ clean_q31_lang <- function(PLO_data_clean){
perc_label_long = paste0(perc_label, "\n(", n, "/", total_respondents, " respondents)")) |>
mutate(xvar = what_lang_is_this)

} else{
post_meds <- post_meds %>%
mutate(total_respondents = 0,
percentage = 0,
perc_label = NA,
per_label_long = NA,
what_lang_is_this = NA)
}

##~~~~~~~~~~~~~~~~~~~~~~~
## ~ recombine dfs ----
##~~~~~~~~~~~~~~~~~~~~~~~
Expand Down
19 changes: 16 additions & 3 deletions R/q8_workflow_satisfaction.R
View file
Original file line number Diff line number Diff line change
Expand Up @@ -58,21 +58,26 @@ clean_q8_workflow_sat <- function(PLO_data_clean){
mutate(total_respondents = sum(n),
percentage = round((n/total_respondents)*100, 1),
perc_label = paste0(percentage, "%")) |>
mutate(xvar = rate_satisfaction)
mutate(xvar = rate_satisfaction) %>%
drop_na()

##~~~~~~~~~~~~~~~~~~~
## ~ post-MEDS ----
##~~~~~~~~~~~~~~~~~~~




#........separate post-MEDS (to add 0s for missing cats).........
post_meds <- df |>
filter(timepoint == "Post-MEDS")

if("Post-MEDS" %in% df$timepoint){

#................add 0s where missing (post-MEDS)................
for (i in 1:length(options)){

cat_name <- options[i]

print(cat_name)
# if category already exists in df, skip to next one ----
if (cat_name %in% pull(post_meds[,2])) {

Expand All @@ -99,6 +104,14 @@ clean_q8_workflow_sat <- function(PLO_data_clean){
perc_label = paste0(percentage, "%")) |>
mutate(xvar = rate_satisfaction)

} else{
post_meds <- post_meds %>%
mutate(total_respondents = 0,
percentage = 0,
rate_satisfaction = NA)
}


##~~~~~~~~~~~~~~~~~~~~~~~
## ~ recombine dfs ----
##~~~~~~~~~~~~~~~~~~~~~~~
Expand Down
Loading

0 comments on commit 45d8902

Please sign in to comment.