Skip to content

Latest commit

 

History

History
64 lines (52 loc) · 4.7 KB

dl_similar.md

File metadata and controls

64 lines (52 loc) · 4.7 KB
Raw

Downloads data for all questionnaires with similar names

Description

Exports data from all questionnaires that match a provided pattern. Function uses str_detect to match the pattern so a regular expression pattern can be used. Additionally, there is the option to exclude questionnaires that have certain words in them (ie training, practice). This function uses the dl_one function so it has similar parameters.

Usage

dl_similar(pattern, exclude = NULL, 
      export_type = "tabular", folder, unzip = TRUE, 
      server, user = "APIuser", password = "Password123", tries = 10)

Arguments

  • pattern (string or list of strings): Pattern to match in the name of questionnaires to download. Can use regular expression for this.
  • exclude (string or list of strings): Questionnaire names containing words in this exclude list will not be downloaded. By default, there are no exclusionary terms.
  • export_type (string): The type of data to export. Options are tabular, stata, spss, binary, paradata. Default is tabular.
  • folder (string): The directory to export the data into. Must be a string. Use forward slash (/) instead of backslash (\).
  • unzip (boolean): Option to unzip the downloaded zip file into the same directory. By default, it will unzip. Set this parameter to FALSE if you would like to not unzip the exported data after download.
  • server (string): Prefix for the survey server. It is whatever comes before mysurvey.solutions: [prefix].mysurvey.solutions.
  • user (string): Username for the API user on the server.
  • password (string): Password for the API user on the server.
  • tries (numeric): Number of times to try getting an export before moving on to next one. Default is 10 times. If you are expecting the dataset on your servers to take a long time to export, increase the number of tries.

Examples

I have a survey that uses the cloud server lf2018.mysurvey.solutions for data collection. I want the data in Stata format and I would like the data to be unzipped after download. I want to download the data into the folder: C:/User/LF_Survey/Data/". On my server, I made an API user with the login: APIuser2018 and password: SafePassword123.

The list of questionnaires currently imported on my server looks like this:

Title Version
LF Survey HH Q1 1
LF Survey HH Q1 2
LF Survey HH Q1 3
LF Survey HH Q2 1
LF Survey HH Q2 2
LF Survey COM 1
LF Survey COM 2
LF Survey HH TRAINING 1
LF Survey HH TRAINING 2
LF Survey HH PILOT 1

From this table, I can see that I have the following questionnaires imported:

  • Multiple versions of the household questionnaire from Quarter 1 (LF Survey HH Q1) and Quarter 2 (LF Survey HH Q2) of the survey
  • Two versions of the community questionnaire (LF Survey COM)
  • Two versions of the household questionnaire used for practice during the interviewer training (LF Survey HH TRAINING)
  • One version of a household questionnaire used for the pilot (LF Survey HH PILOT).

Let's say I want to download all the data for all versions of the household questionnaire from both quarters of fieldwork but I do not want the data from the questionnaires used for the interviewer training or the pilot. Since I want to exclude multiple words, I want to put the strings into a vector. To make a vector with strings to use as exclusionary words, I will use the c() function in R. So to exclude the data from the questionnaire used in the training and the pilot, I will set the exclude paramenter to be c("TRAINING", "PILOT"). I can add as many strings as I want into this vector.

The code to download the data I want using the dl_similar function would look like this:

dl_similar(pattern = "HH", exclude = c("TRAINING", "PILOT"), 
        export_type = "stata", folder = "C:/User/LF_Survey/Data/", unzip = TRUE, 
        server = "lf2018", user = "APIuser2018", password = "SafePassword123")

Now let's say I also want the data from the community questionnaire as well as the household questionnaire. I will once again make a list of strings by using the c() function in R. The pattern will not be set to this c("hh", "comm"). So now my code to download the data would look like this:

dl_similar(pattern = c("hh", "com"), exclude = c("TRAINING", "PILOT"), 
        export_type = "stata", folder = "C:/User/LF_Survey/Data/", unzip = TRUE, 
        server = "lf2018", user = "APIuser2018", password = "SafePassword123")