View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 19889 | Bug reports | Import/Export | public | 2024-12-13 12:32 | 2024-12-13 15:32 |
| Reporter | iceclimber81 | Assigned To | |||
| Priority | none | Severity | minor | ||
| Status | new | Resolution | open | ||
| Product Version | 6.6.x | ||||
| Summary | 19889: Exporting response data to R does not honor the choice of "answer codes" or "full answers" for responses | ||||
| Description | When exporting response data in R format, LS will produce identical export files (.R syntax and R data .csv) irrespective of whether "answer codes" or "full answers" is chosen as the response data format. Because the answer codes are always overwritten by the full answer text in the R syntax file, answer codes are never accessible to the analyst. | ||||
| Steps To Reproduce | Steps to reproduce
Expected resultTo be able to work with 'answer codes' as the response format, the R syntax file would need to repeat what's in the factor LEVELS for the factor LABELS, like so: data[, 8] <- factor(data[, 8], levels=c("a","b","c"),labels=c("a", "b", "c")) Actual resultLS always generates identical R syntax (.R) and R data files (.csv), irrespective of the response format chosen (answer codes or full answers). In the R syntax file, this is an example of the problem: data[, 8] <- factor(data[, 8], levels=c("a","b","c"),labels=c("Option A", "Option B", "Option C")) Explained: The way R works, the factor labels (= full answers, e.g. "Option A", "Option B", "Option C") will effectively overwrite the factor levels (= answer codes, e.g. "a","b","c"), as explained here (https://stackoverflow.com/a/5869675/5389846), making them inaccessible. This is different from how levels and labels work in SPSS. | ||||
| Tags | No tags attached. | ||||
| Attached Files | TEST_syntax_file_AnswerCodes.R (1,923 bytes)
data <- read.csv("survey_599469_R_data_file.csv", quote = "'\"", na.strings=c("", "\"\""), stringsAsFactors=FALSE, fileEncoding="UTF-8-BOM")
# LimeSurvey Field type: F
data[, 1] <- as.numeric(data[, 1])
attributes(data)$variable.labels[1] <- "id"
names(data)[1] <- "id"
# LimeSurvey Field type: DATETIME23.2
data[, 2] <- as.character(data[, 2])
attributes(data)$variable.labels[2] <- "submitdate"
names(data)[2] <- "submitdate"
# LimeSurvey Field type: F
data[, 3] <- as.numeric(data[, 3])
attributes(data)$variable.labels[3] <- "lastpage"
names(data)[3] <- "lastpage"
# LimeSurvey Field type: A
data[, 4] <- as.character(data[, 4])
attributes(data)$variable.labels[4] <- "startlanguage"
names(data)[4] <- "startlanguage"
# LimeSurvey Field type: A
data[, 5] <- as.character(data[, 5])
attributes(data)$variable.labels[5] <- "seed"
names(data)[5] <- "seed"
# LimeSurvey Field type: DATETIME23.2
data[, 6] <- as.character(data[, 6])
attributes(data)$variable.labels[6] <- "startdate"
names(data)[6] <- "startdate"
# LimeSurvey Field type: DATETIME23.2
data[, 7] <- as.character(data[, 7])
attributes(data)$variable.labels[7] <- "datestamp"
names(data)[7] <- "datestamp"
# LimeSurvey Field type: A
data[, 8] <- as.character(data[, 8])
attributes(data)$variable.labels[8] <- "Single choice question with answer codes and labels (code a/b/c):"
data[, 8] <- factor(data[, 8], levels=c("a","b","c"),labels=c("Option A", "Option B", "Option C"))
names(data)[8] <- "Q00"
# LimeSurvey Field type: F
data[, 9] <- as.numeric(data[, 9])
attributes(data)$variable.labels[9] <- "Yes/No question (code 1/0):"
data[, 9] <- factor(data[, 9], levels=c(1,0),labels=c("Yes", "No"))
names(data)[9] <- "Q01"
# LimeSurvey Field type: A
data[, 10] <- as.character(data[, 10])
attributes(data)$variable.labels[10] <- "Yes/No question (code Y/N):"
data[, 10] <- factor(data[, 10], levels=c("Y","N"),labels=c("Yes", "No"))
names(data)[10] <- "Q02"
TEST_syntax_file_FullAnswers.R (1,923 bytes)
data <- read.csv("survey_599469_R_data_file.csv", quote = "'\"", na.strings=c("", "\"\""), stringsAsFactors=FALSE, fileEncoding="UTF-8-BOM")
# LimeSurvey Field type: F
data[, 1] <- as.numeric(data[, 1])
attributes(data)$variable.labels[1] <- "id"
names(data)[1] <- "id"
# LimeSurvey Field type: DATETIME23.2
data[, 2] <- as.character(data[, 2])
attributes(data)$variable.labels[2] <- "submitdate"
names(data)[2] <- "submitdate"
# LimeSurvey Field type: F
data[, 3] <- as.numeric(data[, 3])
attributes(data)$variable.labels[3] <- "lastpage"
names(data)[3] <- "lastpage"
# LimeSurvey Field type: A
data[, 4] <- as.character(data[, 4])
attributes(data)$variable.labels[4] <- "startlanguage"
names(data)[4] <- "startlanguage"
# LimeSurvey Field type: A
data[, 5] <- as.character(data[, 5])
attributes(data)$variable.labels[5] <- "seed"
names(data)[5] <- "seed"
# LimeSurvey Field type: DATETIME23.2
data[, 6] <- as.character(data[, 6])
attributes(data)$variable.labels[6] <- "startdate"
names(data)[6] <- "startdate"
# LimeSurvey Field type: DATETIME23.2
data[, 7] <- as.character(data[, 7])
attributes(data)$variable.labels[7] <- "datestamp"
names(data)[7] <- "datestamp"
# LimeSurvey Field type: A
data[, 8] <- as.character(data[, 8])
attributes(data)$variable.labels[8] <- "Single choice question with answer codes and labels (code a/b/c):"
data[, 8] <- factor(data[, 8], levels=c("a","b","c"),labels=c("Option A", "Option B", "Option C"))
names(data)[8] <- "Q00"
# LimeSurvey Field type: F
data[, 9] <- as.numeric(data[, 9])
attributes(data)$variable.labels[9] <- "Yes/No question (code 1/0):"
data[, 9] <- factor(data[, 9], levels=c(1,0),labels=c("Yes", "No"))
names(data)[9] <- "Q01"
# LimeSurvey Field type: A
data[, 10] <- as.character(data[, 10])
attributes(data)$variable.labels[10] <- "Yes/No question (code Y/N):"
data[, 10] <- factor(data[, 10], levels=c("Y","N"),labels=c("Yes", "No"))
names(data)[10] <- "Q02"
TEST_data_file_AnswerCodes.csv (974 bytes)
"id. Response ID","submitdate. Date submitted","lastpage. Last page","startlanguage. Start language","seed. Seed","startdate. Date started","datestamp. Date last action","Q00. Single choice question with answer codes and labels (code a/b/c):","Q01. Yes/No question (code 1/0):","Q02. Yes/No question (code Y/N):","interviewtime. Total time","groupTime577. Group time: My first question group","Q00Time. Question time: Q00","Q01Time. Question time: Q01","Q02Time. Question time: Q02" "1","2024-12-13 11:59:08","1","en","1331703622","2024-12-13 11:58:59","2024-12-13 11:59:08","a","0","Y","9.01","9.01",,, "2","2024-12-13 11:59:17","1","en","971659123","2024-12-13 11:59:11","2024-12-13 11:59:17","b","1","N","6.91","6.91",,, "3","2024-12-13 11:59:26","1","en","1262037335","2024-12-13 11:59:21","2024-12-13 11:59:26","c","0","Y","5.26","5.26",,, "4","2024-12-13 11:59:57","1","en","2012052566","2024-12-13 11:59:52","2024-12-13 11:59:57","b","0","N","5.97","5.97",,, TEST_data_file_FullAnswers.csv (834 bytes)
"Response ID","Date submitted","Last page","Start language","Seed","Date started","Date last action","Single choice question with answer codes and labels (code a/b/c):","Yes/No question (code 1/0):","Yes/No question (code Y/N):","Total time","Group time: My first question group","Question time: Q00","Question time: Q01","Question time: Q02" "1","2024-12-13 11:59:08","1","en","1331703622","2024-12-13 11:58:59","2024-12-13 11:59:08","a","0","Y","9.01","9.01",,, "2","2024-12-13 11:59:17","1","en","971659123","2024-12-13 11:59:11","2024-12-13 11:59:17","b","1","N","6.91","6.91",,, "3","2024-12-13 11:59:26","1","en","1262037335","2024-12-13 11:59:21","2024-12-13 11:59:26","c","0","Y","5.26","5.26",,, "4","2024-12-13 11:59:57","1","en","2012052566","2024-12-13 11:59:52","2024-12-13 11:59:57","b","0","N","5.97","5.97",,, | ||||
| Bug heat | 6 | ||||
| Complete LimeSurvey version number (& build) | 6.8.2 | ||||
| I will donate to the project if issue is resolved | No | ||||
| Browser | |||||
| Database type & version | LS Cloud Hosting | ||||
| Server OS (if known) | |||||
| Webserver software & version (if known) | |||||
| PHP Version | Info not visible | ||||
 |
Oh ! Complex situation R data cabn be same , but allow R syntax to use |
 |
Yes, exactly, the R data file can be the same (it's a .CSV with only 'answer codes' for both response format types, answer code and full answers), but the R syntax file must be different so it doesn't overwrite the answer codes. |
| Date Modified | Username | Field | Change |
|---|---|---|---|
| 2024-12-13 12:32 | iceclimber81 | New Issue | |
| 2024-12-13 12:32 | iceclimber81 | File Added: TEST_survey_archive.lsa | |
| 2024-12-13 12:32 | iceclimber81 | File Added: TEST_syntax_file_AnswerCodes.R | |
| 2024-12-13 12:32 | iceclimber81 | File Added: TEST_syntax_file_FullAnswers.R | |
| 2024-12-13 12:32 | iceclimber81 | File Added: TEST_data_file_AnswerCodes.csv | |
| 2024-12-13 12:32 | iceclimber81 | File Added: TEST_data_file_FullAnswers.csv | |
| 2024-12-13 14:59 | DenisChenu | Issue Monitored: DenisChenu | |
| 2024-12-13 14:59 | DenisChenu | Bug heat | 0 => 2 |
| 2024-12-13 15:26 | DenisChenu | Note Added: 81625 | |
| 2024-12-13 15:26 | DenisChenu | Bug heat | 2 => 4 |
| 2024-12-13 15:32 | iceclimber81 | Note Added: 81626 | |
| 2024-12-13 15:32 | iceclimber81 | Bug heat | 4 => 6 |
