QUALITY ISSUES IN TELEPHONE SURVEYS: COVERAGE, NON-RESPONSE and QUOTA SAMPLING

The quality of a telephone survey is affected by several factors: telephone coverage, non-response, the methods used to select households and persons, and the quality of responses obtained from respondents. Data are provided which show that a large proportion of Australian households have telephone connections. However, telephone coverage is not uniform and some subgroups of the population have much lower connection rates. This paper reviews evidence of the effect of non-response and the effectiveness of repeated call backs, and reports the results of a new study. The use of quota sampling to select respondents from randomly selected households is also examined. The results suggest that telephone surveys under-represent older persons and the unemployed, and over-represent middle-aged persons. It is shown that while call backs can increase the response rate, the effect on the composition of the sample and resulting estimates is minimal. The main effects are due to refusals and variation in coverage rates.     

An Evaluation of a Large-scale Cati Household Survey Using Random Digit Dialling

Computer-assisted telephone interviewing and random digit dialling are increasingly being used to conduct household surveys in Australia. However, there is little published information concerning Australian experience with such surveys. In 1995 the Government Statistician's Office in Queensland conducted a household survey to study population migration using these techniques. The survey involved a sample of 110 000 telephone numbers resulting in 38 000 responding households. This article describes a computerized survey management system that was developed and which provided information concerning important operational and quality aspects of the survey.     

Households of migrating and nonmigrating populations: differen

Data from a 1983 special survey concerning causes of internal migration in Poland, which covered 5,400 households are merged with results of quarterly surveys of household income to analyze the characteristics of migrating households and their income levels during the period 1970-1983.     

Estimation bias of different design and analytical strategies in dual-frame telephone surveys: an empirical evaluation

Dual-frame survey designs have become increasingly popular in large-scale telephone surveys. This is due to the lack of coverage of the traditional landline survey design and the escalating use of cell phones in recent years. Several estimation strategies have been proposed and their properties have been discussed under ideal scenarios, including pseudo-maximum-likelihood estimation, single-frame estimation, and simple composite estimation [C.J. Skinner and J.N.K. Rao, Estimation in dual frame surveys with complex designs, J. Am. Statist. Assoc. 91 (1996), pp. 349–356; S.L. Lohr and J.N.K. Rao, Inference from dual frame surveys, J. Am. Statist. Assoc. 95 (2000), pp. 271–280]. In practice, estimation in dual-frame telephone surveys is vulnerable to biases and errors (e.g. inaccessibility, topic/mode salience, and measurement error). The investigation of the performance of popular dual-frame estimation methods is scarce in real and less ideal scenarios. Through an innovatively designed simulation study, we compare the estimation bias under different sampling designs with various estimation strategies. To reduce bias, different raking strategies are compared. Simulated scenarios incorporating sampling costs are examined for practical considerations. Overall, the cell phone-only design yields results with the least bias and variance. When accurate covariate information is available for post-stratification, raking estimates from the cell phone-any design also perform very well. We also provide SAS macros for this simulation evaluation upon request. Survey practitioners can fine-tune the parameters based on their prior knowledge of the target population and run the simulation under different scenarios to gain more insights into how to optimally design and analyse telephone surveys.     

Fitting disposition codes to mobile phone surveys: experiences from studies in Finland, Slovenia and the USA

Summary.  Using mobile phones to conduct survey interviews has gathered momentum recently. However, using mobile telephones in surveys poses many new challenges. One important challenge involves properly classifying final case dispositions to understand response rates and non-response error and to implement responsive survey designs. Both purposes demand accurate assessments of the outcomes of individual call attempts. By looking at actual practices across three countries, we suggest how the disposition codes of the American Association for Public Opinion Research, which have been developed for telephone surveys, can be modified to fit mobile phones. Adding an international dimension to these standard definitions will improve survey methods by making systematic comparisons across different contexts possible.     

Designing stratified sampling in economic and business surveys

In most economic and business surveys, the target variables (e.g. turnover of enterprises, income of households, etc.) commonly resemble skewed distributions with many small and few large units. In such surveys, if a stratified sampling technique is used as a method of sampling and estimation, the convenient way of stratification such as the use of demographical variables (e.g. gender, socioeconomic class, geographical region, religion, ethnicity, etc.) or other natural criteria, which is widely practiced in economic surveys, may fail to form homogeneous strata and is not much useful in order to increase the precision of the estimates of variables of interest. In this paper, a stratified sampling design for economic surveys based on auxiliary information has been developed, which can be used for constructing optimum stratification and determining optimum sample allocation to maximize the precision in estimate.     

Surveying migrant households: a comparison of census-based, snowball and intercept point surveys

Summary.  Few representative surveys of households of migrants exist, limiting our ability to study the effects of international migration on sending families. We report the results of an experiment that was designed to compare the performance of three alternative survey methods in collecting data from Japanese–Brazilian families, many of whom send migrants to Japan. The three surveys that were conducted were households selected randomly from a door-to-door listing using the Brazilian census to select census blocks, a snowball survey using Nikkei community groups to select the seeds and an intercept point survey that was collected at Nikkei community gatherings, ethnic grocery stores, sports clubs and other locations where family members of migrants are likely to congregate. We analyse how closely well-designed snowball and intercept point surveys can approach the much more expensive census-based method in terms of giving information on the characteristics of migrants, the level of remittances received and the incidence and determinants of return migration.     

中等收入群体的持续期与退出风险估计———基于EM算法的收入群体划分

本文在利用EM算法求解有限混合分布模型以识别CFPS数据库中各年度中等收入群体的基础上,使用生存分析方法估计了中等收入群体的持续期与退出风险。两种情境下的估计结果显示,20%以上的中等收入家庭在样本期内表现出了脆弱性,这意味着传统的仅考虑一期家庭收入状态的中等收入群体比例测算可能高估了真实的中等收入群体比例。此外,家庭面临倒U型退出风险,持续期与退出风险存在显著的区域与城乡间差异:东部地区与城市地区的中等收入家庭表现出更强的收入稳固性。本文还进一步分析了家庭退出中等收入群体的决定因素,估计结果表明,家庭收入结构、家庭人口特征、家庭职业特征等因素对家庭退出风险影响显著。     

Cellular Telephone,New Products,and the CPI

Since their introduction in 1983, cellular telephones' adoption has grown at 25%–35% per year. At year end 1997, about 55 million cellular telephones were in use in the United States. The Bureau of Labor Statistics (BLS) did not know that cellular telephones existed, at least in terms of calculating the Consumer Price Index (CPI), until 1998 when they were finally included in the CPI. Omitting cellular telephones from the CPI created a significant bias. I estimate a bias in the BLS estimate of the telecommunications-services index of between .8%–1.9% per year because of the omission of the cellular telephone. Rather than telecommunications-service prices increasing at about 1.1% per year, the correct calculation has them decreasing at about .8% per year. Omission of new goods from the CPI, for significant periods of time, leads to important bias in the calculation of the CPI.     

Obtaining cancer risk factor prevalence estimates in small areas: combining data from two surveys

Summary.  Cancer surveillance research requires accurate estimates of risk factors at the small area level. These risk factors are often obtained from surveys such as the National Health Interview Survey (NHIS) or the Behavioral Risk Factors Surveillance System (BRFSS). The NHIS is a nationally representative, face-to-face survey with a high response rate; however, it cannot produce state or substate estimates of risk factor prevalence because the sample sizes are too small and small area identifiers are unavailable to the public. The BRFSS is a state level telephone survey that excludes non-telephone households and has a lower response rate, but it does provide reasonable sample sizes in all states and many counties and has publicly available small area identifiers (counties). We propose a novel extension of dual-frame estimation using propensity scores that allows the complementary strengths of each survey to compensate for the weakness of the other. We apply this method to obtain 1999–2000 county level estimates of adult male smoking prevalence and mammogram usage rates among females who were 40 years old and older. We consider evidence that these NHIS-adjusted estimates reduce the effects of selection bias and non-telephone coverage in the BRFSS. Data from the Current Population Survey Tobacco Use Supplement are also used to evaluate the performance of this approach. A hybrid estimator that selects one of the two estimators on the basis of the mean-square error is also considered.     

Imputation of Household Survey Data Using Linear Mixed Models

Mixed models are regularly used in the analysis of clustered data, but are only recently being used for imputation of missing data. In household surveys where multiple people are selected from each household, imputation of missing values should preserve the structure pertaining to people within households and should not artificially change the apparent intracluster correlation (ICC). This paper focuses on the use of multilevel models for imputation of missing data in household surveys. In particular, the performance of a best linear unbiased predictor for both stochastic and deterministic imputation using a linear mixed model is compared to imputation based on a single level linear model, both with and without information about household respondents. In this paper an evaluation is carried out in the context of imputing hourly wage rate in the Household, Income and Labour Dynamics of Australia Survey. Nonresponse is generated under various assumptions about the missingness mechanism for persons and households, and with low, moderate and high intra‐household correlation to assess the benefits of the multilevel imputation model under different conditions. The mixed model and single level model with information about the household respondent lead to clear improvements when the ICC is moderate or high, and when there is informative missingness.     

Estimating Engel curves: a new way to improve the SILC-HBS matching process using GLM methods

Microdata are required to evaluate the distributive impact of the taxation system as a whole (direct and indirect taxes) on individuals or households. However, in European Union countries this information is usually distributed into two separate surveys: the Household Budget Surveys (HBS), including total household expenditure and its composition, and EU Statistics on Income and Living Conditions (EU-SILC), including detailed information about households'' income and direct (but not indirect) taxes paid. We present a parametric statistical matching procedure to merge both surveys. For the first stage of matching, we propose estimating total household expenditure in HBS (Engel curves) using a GLM estimator, instead of the traditionally used OLS method. It is a better alternative, insofar as it can deal with the heteroskedasticity problem of the OLS estimates, while making it unnecessary to retransform the regressors estimated in logarithms. To evaluate these advantages of the GLM estimator, we conducted a computational Monte Carlo simulation. In addition, when an error term is added to the deterministic imputation of expenditure in the EU-SILC, we propose replacing the usual Normal distribution of the error with a Chi-square type, which allows a better approximation to the original expenditures variance in the HBS. An empirical analysis is provided using Spanish surveys for years 2012–2016. In addition, we extend the empirical analysis to the rest of the European Union countries, using the surveys provided by Eurostat (EU-SILC, 2011; HBS, 2010).     

Sampling within households in household surveys

Summary.  The number of people to select within selected households has significant consequences for the conduct and output of household surveys. The operational and data quality implications of this choice are carefully considered in many surveys, but the effect on statistical efficiency is not well understood. The usual approach is to select all people in each selected household, where operational and data quality concerns make this feasible. If not, one person is usually selected from each selected household. We find that this strategy is not always justified, and we develop intermediate designs between these two extremes. Current practices were developed when household survey field procedures needed to be simple and robust; however, more complex designs are now feasible owing to the increasing use of computer-assisted interviewing. We develop more flexible designs by optimizing survey cost, based on a simple cost model, subject to a required variance for an estimator of population total. The innovation lies in the fact that household sample sizes are small integers, which creates challenges in both design and estimation. The new methods are evaluated empirically by using census and health survey data, showing considerable improvement over existing methods in some cases.     

The Use of Multiplicity in a Household Survey of Disabilities

The multiplicity technique has been proposed (Sirken, 1970) as a means of improving estimates of the number of rare “events” in a population. This paper describes one use of this technique, namely estimation of the number of persons with certain types of disability living in private dwellings in Canberra. The results are taken from three household surveys conducted in Canberra during 1978–79 in which a multiplicity rule linking parents, siblings and children of residents of sample households who were also living in private dwellings in Canberra was used. While no direct evidence on the level of non-sampling error is available here, it appears that net response errors at least are not substantial using this multiplicity rule and that a reasonable gain in sampling efficiency results from the use of multiplicity.     

我国农村住户连续性抽样调查方案设计

 在改革开放的新形势下,我国政府统计部门开展了农村住户等一系列农村统计调查,为解决“三农”问题提供了多方面的数据信息。本文通过分析总结现行农村住户抽样调查方案中存在的各种矛盾和问题,利用国际上前沿的连续性抽样调查方法作为理论基础,分别从农村住户抽样框的构建、连续各期调查样本的抽取、二维平衡轮换模式的设计、连续性抽样估计及其方差估计和连续时间序列数据的调整分析等角度提出一系列改革措施,从而设计出更加科学的调查方案,为及时、准确地搜集和提供关于“三农”问题的数据信息服务。关于其它类型的抽样调查方案亦可按照本文研究的思路类似地加以设计和解决。     

New Zealand Consumption Equivalence Scales From Quasi-unit Record Data

This paper uses quasi-unit record data on expenditures and incomes of New Zealand households of different demographic profiles to estimate consumption equivalence scales which measure the relative levels of spending required by households of different composition to attain given levels of utility. The equivalence scales with respect to specific items of consumption, as well as total consumption, have been obtained from the parameters of the preference-consistent Extended Linear Expenditure System and a few alternative versions of a rank 3 complete demand system with demand shifters. Among the demographic variables included in the models, the type of housing tenure and the actual age of children have special relevance. For the sake of comparison, a few heuristic scales are also computed. The results are largely in line with those found in similar studies of Australia and New Zealand. The analysis is developed within the framework of hypothesized utility-maximizing behaviour of households.     

基于社会核算矩阵乘数的中国产业、居民相对收入分析

 本文使用社会核算矩阵(SAM)收入分配乘数及其分解方法,对中国2000年细分SAM进行了分析,研究了经济系统中产业部门、居民部门相对收入的变化,以及政府部门在收入分配过程中的作用,并分析了引起相对收入变化的构成。通过分析我们得出,从相对收入角度来说,大部分产业之间是非相互促进的;产业对居民相对收入的提高与居民的初始收入水平有较高的相关性;政府部门的投入会提高城镇居民的相对收入,但却会降低农村居民的相对收入。     

A random effects model for diseases with heterogeneous rates of infection

One form of data collected in the study of infectious diseases is on the transmission of a disease within households. We consider a model which allows the rate of disease transmission to vary between households. A Bayesian hierarchical approach to fitting the model is proposed and is implemented by the Metropolis–Hastings algorithm, a standard Markov chain Monte Carlo (MCMC) method. Results are presented for both simulated epidemic chain data and the Providence measles data, illustrating the potential that MCMC methods have to dealing with heterogeneity in infectious disease transmission.     

西安市天然气阶梯价格城市家庭门槛消费量的估计

以西安市为例,通过分层随机抽样获取样本,对家庭天然气阶梯定价初始档的消费量及决定因素进行分析。通过调研发现,在燃气灶、燃气热水器和壁挂锅炉三个主要用能终端中,壁挂锅炉家庭和非壁挂锅炉家庭的消费特征显著不同。运用He和Reiner的能源消费门槛值(Threshold)模型,对这两类家庭样本进行估计,得到壁挂锅炉和非壁挂锅炉家庭天然气阶梯价格门槛值消费量分别为898.07立方米/年和182.74立方米/年,证实西安市实施中的天然气阶梯价格划定的门槛值(一档线)过高,不能实现高收入家庭交叉补贴低收入家庭的收入再分配效果。     

中国城镇居民收入差距对消费结构的影响分析

一、引言改革开放以来 ,我国城镇居民收入的决定机制发生了根本性变化 ,收入来源由单一性向多样化过渡 ,由改革开放前的基本是工薪收入 ,变为以工薪收入为主 ,同时利息、股息、租金等财产收入及经营利润收入比重逐渐增加 ;工薪收入中 ,由过去基本是单一职业收入 ,变为以第一职业为主 ,同时第二职业收入比重逐渐增加 ,居民间的收入差距不断扩大 ,呈现出如下特点[1 ] :其一 ,新兴产业和风险较大产业的职工收入较高 ,如金融保险、房地产业 ,平均月收入 80 0元 ;而开采业、制造业职工收入较低 ,月平均收入 64 0元。其二 ,私营单位就业人员收入较…