Questions tagged [survey]

Questionnaires created with the purpose of extracting information and deriving conclusions from a set of users. Use this tag for questions related to the creation, storage and handling of surveys. Note: Asking people to take a survey is off-topic.

Surveys generally aren't given too much attention by those who are uninterested in a certain topic, or are given unsolicited requests to answer them. Often times they are prone to intentional negligence by answerers, which can very easily skew results and should be carefully filtered out to prevent this. Nonetheless, surveys can provide powerful insights into a specific area of knowledge if implemented correctly.

Perhaps a website is interested to see if a specific set of users are satisfied with the website, namely those users with IP addresses which have only ordered their first few items. This would be very important to ensure that they can retain new customers and improve any shortcomings. They might also want to ask for suggestions from users who have used their website for an extended period of time, longer than 1 year. Creating surveys with these metrics allows for targeted information extraction, and thus the ability to draw specific conclusions from this data.

Surveys may be implemented in software using a variety of programming languages. Popup windows, modules, and follow-up emails are all viable options.

1210 questions
5
votes
1 answer

Error with svychisq - 'contrast can be applied to factors with 2 or more levels'

Error in contrasts<-(*tmp*, value = contr.funs[1 + isOF[nn]]) : contrasts can be applied only to factors with 2 or more levels I'm getting this error whenever I try using the svychisq function in the survey package. However the function works…
Michael Luu
  • 509
  • 3
  • 10
5
votes
1 answer

Measuring income inequality using the R survey package

I'm working with American Community Survey microdata using the survey package, and am hoping to calculate some basic income inequality statistics. I've set up the following as my design: testsurv <- svrepdesign(data=test, repweights = test[,8:87],…
user115457
  • 51
  • 1
5
votes
3 answers

Cannot find the API to submit a survey response to surveymonkey

I need to submit a survey response from my own mobile app. The user will be shown a survey to fill for which i can pull the survey questions and answer choices using the get_survey_details api call but how do I submit the response to survey monkey?…
Imran Suhail
  • 116
  • 6
5
votes
1 answer

Improve speed of queries using complex survey design in R

I have a large data set (more than 20 million obs) that I analyze with survey package and it is taking me ages to run simple queries. I have tried to find a way to speed up my code but I would like to know if there better ways to make this more…
rafa.pereira
  • 13,251
  • 6
  • 71
  • 109
5
votes
6 answers

What's the most efficient way for a developer to switch between tasks?

I'm looking for a workflow-type description of the series of steps you perform to switch from one software development task to another. If a step involves a tool, please specify which tool and how it's used. The goal of the workflow is to have the…
Jim Lamb
  • 25,355
  • 6
  • 42
  • 48
5
votes
2 answers

How to extract attributes values from svyciprop object?

How can I extract attributes from svyciprop object below to a data.frame: library(survey) api <- read.dta(file = "http://www.ats.ucla.edu/stat/stata/library/apipop.dta") data(api) dclus1 <- svydesign(id=~dnum, fpc=~fpc, data=apiclus1) prop.ci <-…
FFI
  • 392
  • 1
  • 10
5
votes
1 answer

How can I generate marginal effects for a logit model when using survey weights?

I normally generate logit model marginal effects using the mfx package and the logitmfx function. However, the current survey I am using has weights (which have a large effect on the proportion of the DV in the sample because of oversampling in some…
Jon M
  • 1,157
  • 1
  • 10
  • 16
5
votes
2 answers

How to make beautiful borderless geographic thematic/heatmaps with weighted (survey) data in R, probably using spatial smoothing on point observations

Ever since Joshua Katz published these dialect maps that you can find all over the web using harvard's dialect survey, I have been trying to copy and generalize his methods.. but much of this is over my head. josh disclosed some of his methods in…
Anthony Damico
  • 5,779
  • 7
  • 46
  • 77
5
votes
1 answer

how to report goodness of fit for a svyglm object?

a question about the survey package in R. I know this is very basic but I've found nothing. So: library(survey) data(api) dclus2 <- svydesign(id=~dnum, weights=~pw, data=apiclus1, fpc=~fpc) summary(svyglm(api00 ~ ell + meals + mobility, design =…
Florian Oswald
  • 5,054
  • 5
  • 30
  • 38
5
votes
1 answer

Difference between svyglm and weighted glm

I would like to know what the difference is between using svyglm or a weighted glm. For example: M1 = glm(formula = yy ~ age + gender + country , family = binomial(link = "probit"), data = P2013, subset = (P2013$E27>=14…
Natuk
  • 57
  • 1
  • 9
5
votes
1 answer

PHP Solution to Generate Radar Chart

A client of mine currently uses a ASP.net application that allows him to conduct a user survey based on 30 Questions and generate an Excel Report. The process is time consuming and labor intensive as it consists of 5-6 steps.. He wants a PHP…
Pankaj
  • 81
  • 1
  • 3
  • 8
4
votes
4 answers

drawing a stratified sample in R

Designing my stratified sample library(survey) design <- svydesign(id=~1,strata=~Category, data=billa, fpc=~fpc) So far so good, but how can I draw now a sample in the same way I was able for simple sampling? set.seed(67359) samplerows <-…
Roland Kofler
  • 1,332
  • 1
  • 16
  • 33
4
votes
2 answers

Calibration based on the total of a variable other than the population

I've been working on some survey data using the survey package. I read the documentation available on post-stratification and calibration, however I got stuck trying to calibrate the sampling weights on a total known for the population that is not…
christk
  • 87
  • 4
4
votes
1 answer

Using group by to get the value corresponding to the max value of another column

In my dataset respondents are grouped together and there is data available about their age. I want all the people in the same group to have the value of the oldest person in that group. So my example data looks like this. df <-…
4
votes
3 answers

please suggest a hosted survey tool with an API

I'm tasked with implementing a survey addition to an existing website. We have been doing this survey with SurveyMonkey, which seems excellent, but the client doesn't like that it isn't on their site. SurveyMonkey doesn't have an API that I can plug…
jcollum
  • 43,623
  • 55
  • 191
  • 321