"The 1980 Population Census of Papua New Guinea was the first complete enumeration of the population of the country. It was also the first census to be designed and processed within the country. This paper gives the background to the census and outlines the development and implementation of the project, concentrating on the problems encountered and the procedures developed to overcome them in this (in many ways) unique environment. The census project was a thorough exercise and recommendations are also given for the future."
"Planning is under way for the U.S.A. bicentennial census in 1990. The U.S. Census Bureau sponsored a study panel under the U.S. Committee on National Statistics to consider key aspects of methodology for the census and to recommend priority areas for research and testing. The recommendations of the Panel on Decennial Census Methodology, which are summarized in this paper, cover four main topics: adjustment of the census counts for coverage errors, methods of coverage evaluation, uses of sampling in obtaining the count, and uses of administrative records in improving the quality of selected content items."
"The U.S. Bureau of the Census will increase significantly the automation of operations for the 1990 Census of Population and Housing, thus eliminating or reducing many of the labor-intensive clerical operations of past censuses and contributing to the speedier release of data products. An automated address control file will permit the computer to monitor the enumeration status of an address. The automated address file will also make it possible to begin electronic data processing concurrently with data collection, and, thus, 5-7 months earlier than for the 1980 Census. An automated geographic support system will assure consistency between various census geographic products, and computer-generated maps will be possible. Other areas where automation will be introduced or increased are questionnaire editing, coding of written entries on questionnaires, and reporting of progress and cost by field offices."
"The current Canadian census is based on the local enumerator who lists and visits each dwelling in his/her enumeration area to drop off the questionnaires just before census day. Other approaches to compiling an address list exist, however. This paper describes current research into the use of an address register for the next Canadian census, to take place in 1991. Possible applications of an address register are described, earlier research is summarized and potential sources for an address register in the Canadian context are mentioned. The paper concludes with a number of concerns which must be addressed in addition to the more technical issues."
"We present a method for estimating omission rates from censuses. Our method is based on the merger of administrative lists, sampling from these lists, and matching against census rolls. We describe the method, present the results of a test in New York City...in 1980, and evalute the results. We compare our proposed method to other procedures for estimating omission rates."
Demographic trends indicate an aging population, highlighting the importance of collecting valid survey data from older adults. One potential issue when surveying older adults is use of technology to collect data on sensitive topics. Survey technologies like A-CASI and IVR have not been used with older adults to measure elder mistreatment. We surveyed 903 adults age 60 and older in Allegheny County, Pennsylvania (U.S.) with random assignment to one of four survey modes: (1) CAPI, (2) A-CASI, (3) CATI; and (4) IVR. We assessed financial, psychological, and physical mistreatment, and examined feasibility of A-CASI and IVR, and effects on prevalence estimates relative to CAPI and CATI. Approximately 83% of elders randomized to A-CASI/IVR used each technology, although 28% of respondents in the A-CASI condition refused to use headphones and read the questions instead. A-CASI produced higher six month prevalence rates of financial and psychological mistreatment than CAPI. IVR produced higher six month prevalence rates of psychological mistreatment than CATI. We conclude that, while IVR may be useful, A-CASI offers a more promising approach to the measurement of elder mistreatment.
"A census of population and housing, which contributes immensely to the development of social statistics and national statistical services in general, has been carried out at least once in every African country except Chad. The common issues and widespread practices of design, implementation, processing, and evaluation of this census are reviewed. Also reviewed are the problems that arise at each of these stages of conducting the census; problems arising as a result of the socio-economic and physical peculiarities of African conditions."
"Most laws of mortality are partial in the sense that they apply only to a broad age group and not to all ages. This paper focuses on three laws of mortality that apply to all ages. Two of them were developed by the actuaries Thiele and Wittstein in the late 19th century. The third, developed by Heligman and Pollard, is of recent origin. The three laws are discussed with references to Scandinavian mortality data. The results suggest that the most recently proposed law can be used for generation of model life tables, for making population projections, simulations, and other statistical work where there is a need for a realistic model of human mortality."
Surveys provide crucial information about the social consequences of armed conflict, but armed conflict can shape surveys in ways that limit their value. We use longitudinal survey data from throughout the recent armed conflict in Nepal to investigate the relationship between armed conflict events and survey response. The Chitwan Valley Family Study (CVFS) provides a rare window into survey data collection through intense armed conflict. The CVFS data reveal that with operational strategies tailored to the specific conflict, duration of the panel study is the main determinant of attrition from the study, just as in most longitudinal studies outside of conflict settings. Though minor relative to duration, different dimensions of armed conflict can affect survey response in opposing directions, with bombings in the local area reducing response rates but nationwide political events increasing response rates. This important finding demonstrates that survey data quality may be affected differently by various dimensions of armed conflict. Overall, CVFS response rates remained exceptionally high throughout the conflict. We use the CVFS experience to identify principles likely to produce higher quality surveys during periods of generalized violence and instability.
Results are reported from a preliminary study testing a new technology for survey data collection: audio computer-assisted self interviewing. This technology has the theoretical potential of providing privacy (or anonymity) of response equivalent to that of paper self-administered questionnaires (SAQs). In addition, it could offer the advantages common to all computer-assisted methods such as the ability to implement complex questionnaire logic, consistency checking, etc.. In contrast to Video-CASI, Audio-CASI proffers these potential advantages without limiting data collection to the literate segment of the population. In this preliminary study, results obtained using RTI's Audio-CASI system were compared to those for paper SAQs and for Video-CASI. Survey questionnaires asking about drug use, sexual behavior, income, and demographic characteristics were administered to a small sample (N = 40) of subjects of average and below-average reading abilities using each method of data collection. While the small sample size renders many results suggestive rather than definitive, the study did demonstrate that both Audio- and Video-CASI systems work well even with subjects who do not have extensive familiarity with computers. Indeed, respondents preferred the Audio- and Video-CASI to paper SAQs. The computerized systems also eliminated errors in execution of "skip" instructions that occurred when subjects completed paper SAQs. In a number of instances, the computerized systems also appeared to encourage more complete reporting of sensitive behaviors such as use of illicit drugs. Among the two CASI systems, respondents rated Audio-CASI more favorably than Video-CASI in terms of interest, ease of use, and overall preference.
"We describe how topics were selected for the Australian Census of Population and Housing. Australia has adopted an open assessment of topics submitted by users and the general public and then performed field tests of difficult or sensitive questions. For the 1986 Census, special attention was given to the development of a census question on ethnic origin."
In sample surveys where sampled units have unequal probabilities of inclusion, associations between the inclusion probabilities and the statistic of interest can induce bias. Weights equal to the inverse of the probability of inclusion are often used to counteract this bias. Highly disproportional sample designs have highly variable weights, which can introduce undesirable variability in statistics such as the population mean or linear regression estimates. Weight trimming reduces large weights to a fixed maximum value, reducing variability but introducing bias. Most standard approaches are ad-hoc in that they do not use the data to optimize bias-variance tradeoffs. This manuscript develops variable selection models, termed "weight pooling" models, that extend weight trimming procedures in a Bayesian model averaging framework to produce "data driven" weight trimming estimators. We develop robust yet efficient models that approximate fully-weighted estimators when bias correction is of greatest importance, and approximate unweighted estimators when variance reduction is critical.
In sample surveys where units have unequal probabilities of inclusion, associations between the inclusion probability and the statistic of interest can induce bias in unweighted estimates. This is true even in regression models, where the estimates of the population slope may be biased if the underlying mean model is misspecified or the sampling is nonignorable. Weights equal to the inverse of the probability of inclusion are often used to counteract this bias. Highly disproportional sample designs have highly variable weights; weight trimming reduces large weights to a maximum value, reducing variability but introducing bias. Most standard approaches are ad hoc in that they do not use the data to optimize bias-variance trade-offs. This article uses Bayesian model averaging to create "data driven" weight trimming estimators. We extend previous results for linear regression models (Elliott 2008) to generalized linear regression models, developing robust models that approximate fully-weighted estimators when bias correction is of greatest importance, and approximate unweighted estimators when variance reduction is critical.
"This paper surveys the literature on the major censuses recorded in the Old Testament, those taken by Moses and that taken by David. It also reviews the Biblical origins and ramifications of the superstition against being counted in a census."
"This paper examines attempts to collect data on a politically controversial topic, race and ethnicity, in the British Census of Population in the post-war period. It discusses an indirect, proxy method of inferring race or ethnicity by asking for the country of birth of the respondent and of his parents, and a direct question where the respondent is asked to identify his racial or ethnic group. Different versions of the direct question are examined, as is the 1979 Census test, which resulted in considerable public resistance to the question. Following the exclusion of the direct question from the 1981 Census, the subject was reviewed by the Parliamentary Home Affairs Committee, the results of whose report--including practical suggestions as to question wording--are discussed."
"During the past twenty years Scandinavian countries have made changes in the methods of taking population and housing censuses that are more fundamental than any seen since modern census methods were first introduced two hundred years ago. These countries extract their census data in part or in whole from administrative registers. If other countries in Western Europe were to adopt this approach, most of them would have to make major improvements to their administrative records. But the primary reasons for making such improvements are concerned with administration and policy rather than statistics, namely, the need to secure a more effective and fairer system of public administration and to enable governments to exercise a wider range of policy options."
"Net undercount rates in the U.S. decennial census have been steadily declining over the last several censuses. Differential undercounts among race groups and geographic areas, however, appear to persist. In the following, we examine and compare several methodologies for providing small area estimates of census coverage by constructing artificial populations. Measures of performance are also introduced to assess the various small area estimates. Synthetic estimation in combination with regression modelling provide the best results over the methods considered. Sampling error effects are also simulated. The results form the basis for determining coverage evaluation survey small area estimates of the 1900 decennial census."
"A parametric method for estimating child mortality from reports concerning children ever born and surviving children is presented. In contrast with previously proposed methods, it facilitates use of single-year age reports by mothers on the survival of their children. In addition, the new method makes it possible to incorporate a priori knowledge of child mortality and fertility in the estimation process. The new method is illustrated by means of an application to data from the 1976 Western Samoa census."
"This paper presents the perspective of a major user of both decennial and economic [U.S.] census data. It illustrates how these data are used as a framework for commercial marketing research surveys that measure television audiences and sales of consumer goods through retail stores, drawing on Nielsen's own experience in data collection and evaluation. It reviews Nielsen's analyses of census data quality based, in part, on actual field evaluation of census results. Finally, it suggests ways that data quality might be evaluated and improved to enhance the usefulness of these census programs."
The use of microcomputers to process census data in developing countries is discussed. Specifically, the authors describe the Integrated Microcomputer Processing System (IMPS) developed by the U.S. Bureau of the Census and focus "on the recent experience of Burkina Faso and Senegal in using microcomputers. A summary is given of the use of IMPS by an increasing number of countries. Additionally, this paper describes the latest enhancements and future plans for IMPS. It concludes with some thoughts on the use of microcomputers in the areas of data collection and processing, as well as in the use and dissemination of census data."
The development of computer systems that support the geography of the census in Canada is described. The authors "present proposals for the 1991 Census that increase the level of automation, whether the geography changes or not. In addition, opportunities for change and improvement, should the geography change, are outlined. To minimize the risks involved, it is important to give these different proposals serious consideration early in the planning of a census."
"Complete decennial censuses are needed for small areas and other domains. Sample surveys yield diverse and timely data. Censuses can also be combined with samples, and sometimes with data from registers, for diverse estimates that are detailed over both space and time, and hence are timely for small domains. Methods of 'postcensal estimates' for small domains are described. We note uses of censuses for improving samples and of samples for improving censuses, and propose a method for cumulating data from 'rolling' (or rotating) periodic (weekly, monthly or quarterly) samples specifically designed to cover the population in detail over designed spans (annual and quinquennial)."
This article reports on a web-based vignette experiment investigating how likely subjects would be to participate in surveys varying in topic sensitivity and risk of disclosure. A total of 3,672 participants each responded to a series of eight vignettes, along with a variety of background questions, concerns about confidentiality, trust in various institutions, and the like.Vignettes were randomly assigned to respondents, such that each respondent was exposed to four levels of disclosure risk for each level of topic sensitivity (high versus low). Half the sample was assigned to receive a confidentiality statement for all eight vignettes, while the other half received no mention of confidentiality in the vignettes. The order of presentation of vignettes was randomized for each respondent.Respondents were also asked for their subjective perceptions of risk, harm, and social as well as personal benefits for one of the eight vignettes. Adding these questions permits us to examine how objective risk information presented by the researcher relates to the subjective perception of risk by the participant, and to assess the importance of both for their willingness to participate in the surveys described.Under conditions resembling those of real surveys, objective risk information does not affect willingness to participate. On the other hand, topic sensitivity does have such effects, as do general attitudes toward privacy and survey organizations as well as subjective perceptions of risk, harm, and benefits. We discuss the limitations and implications of these findings.
The Panel Study of Income Dynamics (PSID) is a nationally representative longitudinal survey of approximately 9,000 families and their descendants that has been ongoing since 1968. Since 1969, families have been sent a mailing asking them to update or verify their contact information to keep track of their whereabouts between waves. Having updated contact information prior to data collection is associated with fewer call attempts and refusal conversion efforts, less tracking, and lower attrition. Given these apparent advantages, a study was designed in advance of the 2009 PSID field effort to improve the response rate of the contact update mailing. Families were randomly assigned to the following conditions: mailing design (traditional versus new), $10 as a prepaid versus postpaid incentive, timing and frequency of the mailing (July 2008 versus October 2008 versus both times) and whether or not they were sent a study newsletter. This paper reports on findings with regards to response rates to the mailing and the effect on production outcomes including tracking rates and number of calls during 2009 by these different conditions, examines whether the treatment effects differ by key characteristics of panel members including likelihood of moving and anticipated difficulty in completing an interview, and provides some recommendations for the use of contact update strategies in panel studies.
Since 1969, families participating in the U.S. Panel Study of Income Dynamics (PSID) have been sent a mailing asking them to update or verify their contact information in order to keep track of their whereabouts between waves. Having updated contact information prior to data collection is associated with fewer call attempts, less tracking, and lower attrition. Based on these advantages, two experiments were designed to increase response rates to the between-wave contact mailing. The first experiment implemented a new protocol that increased the overall response rate by 7 - 10 percentage points compared to the protocol in place for decades on the PSID. This article provides results from the second experiment which examines the basic utility of the between-wave mailing, investigates how incentives affect article cooperation to the update request and field effort, and attempts to identify an optimal incentive amount. Recommendations for the use of contact update strategies in panel studies are made.
A common strategy for handling item nonresponse in survey sampling is hot deck imputation, where each missing value is replaced with an observed response from a "similar" unit. We discuss here the use of sampling weights in the hot deck. The naive approach is to ignore sample weights in creation of adjustment cells, which effectively imputes the unweighted sample distribution of respondents in an adjustment cell, potentially causing bias. Alternative approaches have been proposed that use weights in the imputation by incorporating them into the probabilities of selection for each donor. We show by simulation that these weighted hot decks do not correct for bias when the outcome is related to the sampling weight and the response propensity. The correct approach is to use the sampling weight as a stratifying variable alongside additional adjustment variables when forming adjustment cells.
Survey respondents may misinterpret the questions they are asked, potentially undermining the accuracy of their answers. One way to reduce this risk is to make definitions of key question concepts available to the respondents. In the current study we compared two methods of making definitions available to web survey respondents - displaying the definition with the question text and displaying the definition when respondents roll the mouse over the relevant question terms. When definitions were always displayed they were consulted more than when they required a rollover request. The length of the definitions did not affect how frequently they were used under either method of display. Respondents who completed training items designed to encourage definition use actually requested definitions less often, suggesting that they may value minimal effort over improved understanding. We conclude that at least for small numbers of questions, providing definitions with the question is likely to be the more effective approach than rollovers or hyperlinks.
Web surveys often collect information such as frequencies, currency amounts, dates, or other items requiring short structured answers in an open-ended format, typically using text boxes for input. We report on several experiments exploring design features of such input fields. We find little effect of the size of the input field on whether frequency or dollar amount answers are well-formed or not. By contrast, the use of templates to guide formatting significantly improves the well-formedness of responses to questions eliciting currency amounts. For date questions (whether month/year or month/day/year), we find that separate input fields improve the quality of responses over single input fields, while drop boxes further reduce the proportion of ill-formed answers. Drop boxes also reduce completion time when the list of responses is short (e.g., months), but marginally increases completion time when the list is long (e.g., birth dates). These results suggest that non-narrative open questions can be designed to help guide respondents to provide answers in the desired format.
This article extends earlier work (Couper et al. 2008) that explores how survey topic and risk of identity and attribute disclosure, along with mention of possible harms resulting from such disclosure, affect survey participation. The first study uses web-based vignettes to examine respondents' expressed willingness to participate in the hypothetical surveys described, whereas the second study uses a mail survey to examine actual participation. Results are consistent with the earlier experiments. In general, we find that under normal survey conditions, specific information about the risk of identity or attribute disclosure influences neither respondents' expressed willingness to participate in a hypothetical survey nor their actual participation in a real survey. However, when the possible harm resulting from disclosure is made explicit, the effect on response becomes significant. In addition, sensitivity of the survey topic is a consistent and strong predictor of both expressed willingness to participate and actual participation.
"Population counts for small areas, as in a census, are directly affected by the rules that determine where each person is to be counted. Clear rules are needed for groups such as migratory workers, persons whose work requires long absences from the family home, homeless persons, military personnel, persons maintaining two or more homes, persons in hospitals, correctional, or other institutions, crews of sea-going vessels, and others. The rules may have a bearing on family and household statistics and statistics on socioeconomic status as well as on counts of numbers of people."
In this paper we follow up two notes of Thomsen (1973, 1978) and present some results on the estimation effect of post-stratification when analyzing binary survey data subject to non-response. Using an alternative parameterisation and assuming that the non-response depends on the variable of interest which can not be fully observed, we show that the relative reduction in the bias can be estimated from the response group alone. In addition, the relative bias and variance reduction are both shown to be approximately equal, under certain conditions, to one minus the square of the correlation coefficient between the auxiliary and object variable among the respondents.
Empirical evidence suggests that the respondents� approval motive, their desirability beliefs and the privacy of the response situation affects respondents� susceptibility to social desirability bias. Previous research has analyzed the explanatory power of these factors separately and has not taken their possible interdependence as determinants for social desirability bias into account. This article tests the prediction from rational-choice theory that a strong approval motive, clear differences in the perceived desirability of response options and a lack of privacy are all necessary but not sufficient conditions for social desirability bias. According to the empirical results from our first study a three-way interaction between the analyzed factors predicts respondents� racial attitude reports. However, since attitude answers and desirability beliefs were collected in the same interview, the observed associations may be an artifact due to subjects� sensitization for social desirability concerns. This possibility is tested in a second study, where only racial attitude answers were collected under conditions of varying response privacy. Aggregated response differences between the utilized attitude items and respondents� social group affiliation were matched with equivalent differences in the desirability beliefs found in the first study. The results from the main study were replicated with this independent sample of respondents.
Carling et al (1996) analyze a large data set of unemployed workers in order to examine, inter alia, the effect of unemployment benefits on the escape rate to employment. In this paper we take a closer look at the 20 per cent of workers who were drop-outs and check the empirical justification for modeling attrition as independent right censoring in the analysis of unemployment duration. It may very well be that dropping out, i.e. attrition, often occurs due to employment. In the analysis, we refer to these individuals as misclassified in that they are typically treated as if their unemployment spell went beyond the time of attrition. We propose to follow up the drop-outs by a supplementary sample and apply a Multiple Imputation approach to incorporate the supplementary information. Our follow-up study revealed that 45% dropped out due to employment. The escape rate to employment was as a consequence under-estimated by 20 per cent, implying that the effect of unemployment benefits on the escape rate is likely to be much greater than reported in Carling et al (1996)
In sampling theory the large concentration of the population with respect to most surveyed variables constitutes a problem which is difficult to tackle by means of classical tools. One possible solution is given by cut-off sampling, which explicitly prescribes to discard part of the population; in particular, if the population is composed by firms or establishments, the method results in the exclusion of the “smallest” firms. Whereas this sampling scheme is common among practitioners, its theoretical foundations tend to be considered weak, because the inclusion probability of some units is equal to zero. In this paper we propose a framework to justify cut-off sampling and to determine the census and cut-off thresholds. We use an estimation model which assumes as known the weight of the discarded units with respect to each variable; we compute the variance of the estimator and its bias, which is caused by violations of the aforementioned hypothesis. We develop an algorithm which minimizes the MSE as a function of multivariate auxiliary information at the population level. Considering the combinatorial optimization nature of the model, we resort to the theory of stochastic relaxation: in particular, we use the simulated annealing algorithm.
Using rich register data to analyze response behavior in a survey on health and economic standard, a model to explain contact and participation probabilities is estimated. A main result is that both probabilities are lower among respondents out of the labor market, who are immigrants and on benefits.
The next round of census will be completely register-based in all the Nordic countries. Household is a key statistical unit in this context, which however does not exist as such in the administrative registers available, and needs to be created by the statistical agency based on the various information available in the statistical system. Errors in such register households are thus unavoidable, and will propagate to various induced household statistics. In this paper we outline a unit-error theory which provides a framework for evaluating the statistical accuracy of these register-based household statistics, and illustrate its use through an application to the Norwegian register household data.
This article shows that respondents gain meaning from verbal cues (words) as well as nonverbal cues (layout; numbers) in a web survey. We manipulated the layout of a five-point rating scale in two experiments. In the first experiment, we compared answers for different presentations of the responses: in one column with separate rows for each answer ("linear"), in three columns and two rows ("nonlinear") in various orders, and after adding numerical labels to each response option. Our results show significant differences between a linear and nonlinear layout of response options. In the second experiment we looked at effects of verbal, graphical, and numerical language. We compared two linear vertical layouts with reverse orderings (from positive to negative and from negative to positive), a horizontal layout, and layouts with various numerical labels (1 to 5, 5 to 1, and 2 to 22). We found effects of verbal and graphical language. The effect of numerical language was only apparent when the numbers 2 to 22 were added to the verbal labels. We also examined whether the effects of design vary with personal characteristics. Elderly respondents appeared to be more sensitive to verbal, graphical, and numerical language.
We analyze five vintages of eighteen quarterly macroeconomic variables for the Netherlands and we focus on the degree of deterministic seasonality in these series. We document that the data show most such deterministic seasonality for their first release vintage and for the last available vintage. In between vintages show a variety of seasonal patterns. We show that seasonal patterns in later vintages can hardly be predicted by those in earlier vintages. The consequences of these findings for the interpretation and modeling of macroeconomic data are discussed.
Systematic sampling is a widely used technique in survey sampling. It is easy to execute, whether the units are to be selected with equal probability or with probabilities proportional to auxiliary sizes. It can be very efficient if one manages to achieve favourable stratification effects through the listing of units. The main disadvantages are that there is no unbiased method for estimating the sampling variance, and that systematic sampling may be poor when the ordering of the population is based on inaccurate knowledge. In this paper we examine an aspect of the systematic sampling that previously has not received much attention. It is shown that in a number of common situations, where the systematic sampling has on average the same efficiency as the corresponding random sampling alternatives under an assumed model for the population, the sampling variance fluctuates much more with the systematic sampling. The use of systematic sampling is associated with a risk that in general increases with the sampling fraction. This can be highly damaging for large samples from small populations in the case of single-stage sampling, or large sub-samples from small sub-populations as in the case of multi-stage sampling.
The cognitive processes that underlie self- and proxy-reports of attitudes and behaviors have recently received increased attention in the survey methodology literature. However, this research has not yet fully exploited the wealth of psychological theorizing that may potentially be relevant to self and proxy reporting. Most importantly, psychological research on person perception and attribution bears on how individuals form mental representations of self and other, and on how they use these representations in recall and estimation processes. The present paper provides a selective review of this psychological literature; relates its key findings to methodological issues of self and proxy reporting; and reports the findings of several laboratory experiments and an experimental survey designed to provide first tests of key hypotheses derived from cognitive theorizing. Implications for future research, respondent selection rules, and questionnaire construction are discussed. Cognitive D...
A nondisclosure policy for tabular data or microdata restricts release of information that could be related to a specific individual. Pannekoek and de Waal (1998) describe a rule that suppresses data release when the number of people in a cell defined by a rare characteristic falls below a fixed floor, and show how empirical Bay es methods can be used to improve the estimation of that number. We argue that the nondisclosure problem can be formulated as a decision problem in which one loss is associated with the possibility of disclosure and another with nonpublication of data. This analy sis supports a decision on whether to disclose information in each cell, minimizing the expected sum of the two losses. We present arguments for several loss functions, considering both tabular and microdata releases, and illustrate their application to simple simulated data.
It is common practice to use weighting, poststratification, and raking to correct for sampling and nonsampling biases and to improve efficiency of estimation in sample surveys. However, there is no standard method for computing sampling variances of estimates that use these adjustments in combination. In this paper we develop such a method, using three ideas: (1) a general notation that unifies the different forms of weighting adjustment, (2) a variance decomposition to estimate sampling variances conditional and unconditional on sample sizes within poststratification categories, and (3) numerical computation using a delta method. We apply our approach to the problem that motivated this research, the New York City Social Indicators Survey, a telephone survey that uses inverse-probability weighting, poststratification, and raking to correct for sampling design and nonresponse. Our variance estimates systematically differ from those obtained using methods that do not account for the desi...
Survey questionnaires commonly include retrospective questions on attitudes as a substitute for concurrently obtained assessments. Empirical studies have shown that these retrospectively obtained attitudes often deviate seriously from the attitudes for which they are substitutes: the attitude that respondents held in the past. In spite of this, hardly any research has been done to examine whether aided recall methods might be able to minimize such a discrepancy. The present study provides a first empirical test of a cue-list which was used as an aided recall method for retrospective attitude questions about unemployment. The cue-list consisted of a standardized set of cues – advantages and disadvantages of being unemployed – that was added to a retrospective attitude question on unemployment. It was tested in a randomized field experiment during the second wave (1991) of a longitudinal social survey in the Netherlands. The cue-list was expected to enhance the reconstruction of the past attitudes of those respondents who had actually been unemployed at the time of the first interview. Respondents who had not been unemployed were treated as a control group. The agreement between the recalled attitude and the attitude reported in the first wave of the survey, four years earlier, was used to evaluate the effects of the cue-list. Contrary to expectation, the cue-list led to a lower instead of a higher level of agreement. However, in line with Tulving’s encoding specificity principle, the “unemployment” cue-list did have this unexpected effect only for the respondents who had been unemployed at the time of the first interview and not for the control group. In addition, the cue-list led to a more positive attitude towards unemployment, in retrospect, for the former group of respondents. Exploratory analyses suggest that the cues might have triggered positive features of unemployment that could have been easily overlooked otherwise. Based on the outcomes, the risks of applying cues are discussed
The measures of household income in UK household surveys refer to income received round about the time of the interview (“current income”). By contrast the measure of income in surveys for most other countries refers to annual income. The British Household Panel Survey provides a unique opportunity to compare estimates of Britain’s income distribution based upon both current and annual measures. This article compares the measures, shows that they provide remarkably similar estimates, and explains why. The results suggest that differences in income distributions between Britain and other nations do not arise because of the different survey measures of income.
Policymakers, faced with increasing demands to make decisions at a local level, are turning to statistical agencies to provide local data. Advances in matching technology, combined with the reduced cost of archiving, indexing, storing, and curating large-scale datasets, now mean that it is technically feasible to provide information at fine levels of geographic detail by means of combining administrative and survey datasets at lower cost and with potentially greater coverage. This article describes an approach that uses administrative data from U.S. unemployment insurance records to enhance the coverage and accuracy of work location information in the American Community Survey.