Questions tagged [survey-weights]
survey weights are used when data are collected according to a probability sampling design with unequal probabilities of selection and/or response
227 questions
0 votes
0 answers
19 views
How can I test the proportional odds assumption for cumulative logit models estimated with svyVGAM?
I'm trying to estimate a cumulative logit model with partial proportional odds while taking into account sampling weights (to make the sample representative of the population). My best option seems to ...
1 vote
0 answers
94 views
Raking survey weights of specific subgroups
The Question: what are the best practices for raking the survey weights of a specific subsample while leaving the remaining sample (relatively) unchanged for later comparative analyses? The Situation: ...
- 11
0 votes
1 answer
51 views
How to predict household income in census sample using income & expenditure survey data in R? [closed]
I have two datasets at the individual level, which I aggregated to the household level: Census Sample: A 2% sample of the total population from a national census. This includes household-level ...
- 101
2 votes
1 answer
57 views
Comparing distributions of three likert variables within the levels of a fourth (categorical) variable
I'm having a bit of a brain fart on a very basic question. Background: I am analysing data from a survey with 2,000 respondents. The survey had 10 questions on the respondents' familiarity and comfort ...
- 1,837
2 votes
1 answer
97 views
Effect sizes for regression associations in complex survey data
I am conducting regression analyses using complex survey data that incorporates weights, clustering, and stratification (implemented in R using the survey package with svydesign() and svyglm()). A ...
3 votes
1 answer
93 views
How to use R survey with survey data that had a complex sampling process but offers only one sampling weight with no info on strata or clusters."
I'm working with survey data (not my own) that had a relatively complex sampling process (selection of regions, then households, then one or more people within the household). However, the scientific ...
- 31
1 vote
1 answer
124 views
Post-stratification weights for subsample
I have a survey of around 3000 respondents quota sampled to match population demographics (age-sex-education) with post-hoc weights (age-sex-education) computed from population statistics to correct ...
- 45
2 votes
2 answers
122 views
Two-sample randomisation test using survey weights - biased?
I like randomisation methods for two-sample tests because they relax assumptions without losing power and that's really important to me. Now I have data with survey weights that I'd like to ...
- 1,241
2 votes
1 answer
131 views
Are there good references discussing the use of survey weights with PCA and MCA?
I've found that the survey package in R allows using survey weights with principal component analysis, which is great. However, it doesn't seem to provide the same for correspondence analysis or ...
- 23
1 vote
1 answer
119 views
How to calculate BIC values for logit models using the survey package in R?
I am using lumley's survey package in R to estimate logit models and I want to compare BIC statistics of several nested models. Given that they are not estimated with maximum likelihood, I am having ...
- 41
2 votes
1 answer
199 views
Alternative to R-squared when calculating glm model using survey package
I'm writing my master thesis using the European Social Survey. The data requires weighting, so therefore I have to use the survey package and its ...
- 41
1 vote
0 answers
47 views
Incidence influenced by test frequency
I’m puzzled with a study design. Say I want to study a disease incidence, but this disease is asymptomatic mostly and its detection merely relies on regular testing. Some people test more frequently ...
- 11
1 vote
0 answers
140 views
Test (quantify) the association between categorical and numerical variable in survey data
I'm working on the analysis of survey data where the weights are the results of a marginal calibration. The dimension are 968 rows and 168 columns. My first goal is to compare two subgroups A and B ...
- 11
1 vote
0 answers
86 views
Quantile regression to compare two populations at different time points
I am trying to compare two different populations (e.g., 2001 and 2011) on the same variable. These are different, independent populations (this is NOT a longitudinal study; rather, two cross-sectional ...
- 11
6 votes
0 answers
64 views
How are effect sizes typically defined under complex sampling?
In some fields it is common to quote comparisons in terms of standardised effect sizes. For example, a difference in means gets scaled by the sample standard deviation to give an effect size. This ...
- 57k