To Model or Not To Model? Competing Modes of Inference for Finite Population Sampling

By Roderick J. A. Little. 2004

300 citations recorded by the source corpus.

Subfields: Sociology and Political Science, Statistics and Probability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Sociology and Political Science

In Statistics and Probability

Citations in the corpus

Prediction-Based Decisions and Fairness: A Catalogue of Choices, Assumptions, and Definitions (2018, background): “sampling and measurement) [44, 54] enables modeling that appropriately accounts for the specific conditions under which data has been collected (Chapter 8 of [46], [110, 83]).”
Bayesian Estimation Under Informative Sampling (2015, methodology): “By contrast, Little (2004) offer no modeling approach that parameterizes a proportion-to-size sampling design because they note that each unit is in its own group under the stratum-indexed construction they generally suggest.”
Risk‐Utility Paradigms for Statistical Disclosure Limitation: How to Think, But Not How to Act (2011, background): “Fienberg (2009) argues further for model-based rather than design-based analyses (see also Little (2003)), which would allow weights to be suppressed and obviate risks associated with them.”
Responsive design for household surveys: tools for actively controlling survey errors and costs (2006, background): “Statistical inference from sample surveys has traditionally conditioned on sample design features (Little, 2004).”
Model-Assisted Estimation of Forest Resources With Generalized Additive Models (2007, background): “(For other examples in which calibration yields strange estimates, see Little 1983.)”
Utilization Distribution Estimation Using Weighted Kernel Density Estimators (2007, methodology): “…one might be willing to assume data are independent and then proceed with a model-based analysis, for example, linear regression assuming that residuals are independent and identically distributed (see Thompson 2002:23–26 and Little 2004 for a comparison of design-based and modelbased inference).”
Calibrated Bayes (2006, background): “I have argued that a unified approach based on models that account for the survey design would reduce friction and speed progress (Little 2004).”
Imperfect detection and its consequences for monitoring for conservation (2008, background): “However, it is important to recognize that this creates a danger of losing the ability to generalize to the entire area about which inference is desired, unless selection bias can be modelled (Little 2004).”
Multi-level modelling under informative sampling (2006, methodology): “7), Pfeffermann & Sverchkov (2003) and Little (2004).”
Calibrated Bayes, an alternative inferential paradigm for official statistics (2012, methodology): “…by modeling the outcome as a spline function of size, avoids this problem, and has been shown to yield Bayesian inferences with superior frequentist properties to sample-weighted estimates in simulations (Zheng and Little 2004, 2005; Yuan and Little 2007, 2008; Chen, Elliott and Little 2010).”
Mixed model prediction and small area estimation (2006, background): “In addition to the papers cited in our paper and Professor Pfeffermann’s discussion, we also refer to two nice papers by Ghosh and Sinha (1990) and Little (2004).”
Integrative Data Analysis: The Simultaneous Analysis of Multiple Data Sets (2009, methodology): “…each was also 5For more detailed discussions about the similarities and differences between model-based and design-based procedures as well as current hybrid designs that combine the two approaches see Guo & Hussy (2004), Little (2004), Lenhard (2006), Muthén and Satorra (1995) and Sterba (2009).”
Proxy Pattern-Mixture Analysis for Survey Nonresponse (2011, background): “respondents and nonrespondents (Little 2004) and yields estimates of the mean of Y of the form”