Survey Sampling

Sampling - bridging probability and non probability designs - final Word version

by Emma Uprichard

Full reference of final published paper:
Uprichard, E. (2011) 'Sampling - bridging probability and non probability designs', International Journal of Social Research Methodology, iFirst/Online Early, 1-11.

This article reconceptualizes sampling in social research. It is argued that three inter-related a priori assumptions... more This article reconceptualizes sampling in social research. It is argued that three inter-related a priori assumptions limit on the possibility of sample design, namely: (a) the ontology of the case, (b) the epistemological assumptions underpinning what properties are necessary to know the case and (c) the logistics
involved in the process of ‘casing’ the case. In considering sampling in this way, not only are key criteria commonly used to gauge the validity of sample problematized, but a genuine epistemological bridge between probability and non-probability sample designs is also forged.

Herrera, V. M., A. Uriarte, et al. (2011). ‘Background noise’ and landscape exploitation in the Late Iron Age Andalusian countryside. Hidden Landscapes of Mediterranean Europe. Cultural and methodological biases in pre- and protohistoric landscape studies. M. v. Leusen, G. Pizziolo and L. Sarti. Collegio S. Chiara, University of Siena, BAR International Series, Archeopress. 2320: 265-270.

by Victorino Mayoral Herrera

In this paper we present the results from a survey carried out from 2000 to 2003 in the Guadiana Menor valley (Jaen... more In this paper we present the results from a survey carried out from 2000 to 2003 in the Guadiana Menor valley (Jaen province,
Spain). The sampling strategy was designed to explore the presence of archaeological fi nds in very different landscape units, from the fl oodplains to the uncultivated slopes of the Cazorla massif. Field survey revealed very high contrasts in fi nd densities, suggesting different patterns of land use. On the valley fl oor we recorded a weak carpet-like pattern of sherds dating from the Iron Age to Roman times. Its distribution suggests intensive agricultural activities such as manuring in a context of increasing pressure on resources during the Romanization process. In the highland area, the much more scarce evidence seems to be connected with pastoral activity,
intensive farming of small valleys, and control over strategic mountain passes. We will here present methodological issues relating to the fi eld survey and reliability problems raised by erosion and land use history of the study area.

Rotation Sampling for Functional Data

by David Degras

Survey sampling methods provide cost-effective solutions for monitoring global parameters in large populations.... more Survey sampling methods provide cost-effective solutions for monitoring global parameters in large populations. Although time-varying samples are known to outperform fixed panels in various instances of discrete-time repeated surveys, they have not yet been examined in the continuous-time setup of sensor networks. In this paper we devise sampling designs for functional data (that is, continuous signals) based on rotation sampling and stratification. We propose to periodically replace the sample according to a Markov chain, which allows for spatial and temporal adaptation to the network. Considering the Horvitz-Thompson estimator of the mean temporal signal, we show that the variance of the Integrated Squared Error (ISE) can be dramatically reduced by increasing the frequency or intensity of sample replacements. Further, the average ISE can be decreased by suitably allocating the sample across strata at replacement times. An application to simulated electricity consumption data illustrates the good performances of our sampling designs relative to fixed panels.

Internet Marketing Research: Opportunities and Problems

by Olivier Furrer

Olivier Furrer and D. Sudharshan
Qualitative Market Research: An International Journal
Vol. 4, Iss: 3, 2001, pp.123 - 129

The Internet is promised a brilliant future among the favorite tools of marketing researchers. Develops a typology of... more The Internet is promised a brilliant future among the favorite tools of marketing researchers. Develops a typology of Internet marketing surveys showing the existence of eight different designs that can be used by marketers. However, researchers who plan to develop research using the Internet need to be aware of several problems related to this new tool. In particular we show that the nature of the Internet creates different sampling problems. To identify these problems, a seven-step procedure following the steps of the sampling process is proposed. Several practical problems are then discussed.

CUESTIONARIOS PSICOLÓGICOS E INVESTIGACIÓN EN INTERNET: UNA REVISIÓN DE LA LITERATURA Anales de Psicología, junio, año/vol. 24, …

by Francisco José Eiroa Orosa

Combined Survey Sampling Inference

by Steven LaLonde

AN APPLICATION ABOUT THE UNIVERSITY STUDENTS IN IZMIR WITH STRATIFIED CLUSTER SAMPLING

by Aslı Suner

Suleyman Demirel University The Journal of Faculty of Economics and Administrative Sciences Y.2008, Vol.13, No.3 pp.407-424.

The ratio of young population is quite high in Turkey, so identifying opinions of the young people are of great... more The ratio of young population is quite high in Turkey, so identifying opinions of the young people are of great importance in terms of solving the various kinds of the problems of Turkey. The subject of this study was to determine opinions of the undergraduate students studying in İzmir concerning Turkey – European Union relations, primary problems of Turkey and habits of the undergraduate students. In this study, 1036 undergraduate student were selected from five different universities located in İzmir and prepared questionnaire forms were applied on this study group during 2004-2005 spring term and 2006-2007 fall term. With the purpose of assigning differences between the results that were obtained two different terms, Stratified Cluster Sampling and Neyman Allocation Approach were used. After analysis of the data, various statistical results were observed between the terms.
Key Words: Stratified Cluster Sampling, Neyman Allocation,
zmir, University Students

Confidence Bands for Horvitz-Thompson Estimators Using Sampled Noisy Functional Data

by David Degras

Co-authored with Hervé Cardot and Etienne Josserand (Université de Bourgogne). Submitted to Bernoulli.

When collections of functional data are too large to be exhaustively observed, survey sampling techniques provide an... more When collections of functional data are too large to be exhaustively observed, survey sampling techniques provide an effective way to estimate global quantities such as the population mean function. Assuming functional data are collected from a finite population according to a probabilistic sampling scheme, with the measurements being discrete in time and noisy, we propose to first smooth the sampled trajectories with local polynomials and then estimate the mean function with a Horvitz-Thompson estimator. Under mild conditions on the population size, observation times, regularity of the trajectories, sampling scheme, and smoothing bandwidth, we prove a Central Limit Theorem in the space of continuous functions. We also establish the uniform consistency of a covariance function estimator and apply the former results to build global confidence bands for the mean function. The bands attain nominal coverage and are obtained through Gaussian process simulations conditional on the estimated covariance function. To select the bandwidth, we propose a cross-validation method that accounts for the sampling weights. A simulation study assesses the performance of our approach and highlights the influence of the sampling scheme and bandwidth choice.

Operator bias with optical wedges in point sampling

by Jerome Vanclay

Commonwealth Forestry Review 74:158 (1995)

Point sampling, or sampling with probability proportional to size (PPS), is an efficient technique often used in... more Point sampling, or sampling with probability proportional to size (PPS), is an efficient technique often used in timber cruising, but care must be taken to avoid operator bias.

Sampling Methods Used In Developed Countries for Behavioural Surveillance Among Men Who Have Sex With Men

by Dana Paquette