Search

Prospect theory states that the influential power of avoiding negative outcomes is stronger than that of achieving positive outcomes. In a survey context, this theory has been tested with respect to not only participation in a CATI survey, but also giving consent to data linkage in CATI surveys. No study, however, has tested the theory with respect to participation in a CAPI or web survey. This study does so in a mixed-mode panel context; it also tests the moderating effects of time-in-panel, response history, and mode protocol. Results show that the framing of the survey participation request influences participation propensity in a way consistent with prospect theory, but only for relatively recent panel entrants. The opposite effect is found for long-term panel participants. No difference is found between mode protocols.

Previous research into survey invitation letters and prenotification letters has sought to identify optimal features of a standard letter that can be sent to all sample members. In the spirit of adaptive design, this study seeks to establish whether it can be advantageous to target different versions of a letter to different sample subgroups. Specifically, a paragraph intended to heighten perceptions of relevance of the survey is varied among six subgroups in a panel survey. Random allocation to control and treatment groups is crossed with variants of two other design features: time in sample and data collection mode. This enables analysis of the effects of the targeted letter in different survey contexts. The targeted version of the letter improved response rates for two operationally important low-response propensity groups, but only in certain survey design contexts.

A 78-year-old female underwent a right pneumonectomy for a neuroendocrine tumour. In the early postoperative phase, she developed a paralytic ileus and went on to develop breathlessness and orthostatic desaturation, which characterizes platypnoea–orthodeoxia. She was found to have an intra-atrial shunt on bubble echo, with equal atrial pressures. This was managed by device closure of the atrial connection, following which her condition improved.

Web questionnaires, including those used in mixed-mode surveys, generally produce higher levels of item nonresponse than interviewer-administered questionnaires. Item nonresponse is generally seen as having a detrimental impact on data quality. The current study examines using motivational statements to reduce item nonresponse in a web survey component of a mixed-mode design. The effects of alternative implementations are compared, both for web surveys and for mixed-mode surveys. In addition, the mixed-mode results are compared to a face-to-face survey. The current study adds to the literature on the use of motivational statements by using a unique large-scale randomized experiment to examine the impact of the timing of the motivational statement, and to compare with the same survey in an interviewer-administered context. Findings show that a motivational statement following immediately after an item is left unanswered greatly outperforms either the control or a motivational statement at a later point in the survey. Using this immediate reactive prompt reduces item nonresponse to levels equivalent to a face-to-face version. Conversely, the control (no statement) and later placed motivational statement lead to significantly greater item nonresponse. Point estimates for the tested variables are not affected by the additional responses obtained. The results suggest practical design implications to reduce item nonresponse when using a web design, specifically the use of a reactive motivational prompt in a planned way.

This article considers the effect that a wave of mixed-mode data collection (telephone and face-to-face), in an otherwise face-to-face survey, can have on panel attrition and the extent to which this effect is dependent on the nature of the mode-switch protocol. Findings are reported from an experiment carried out at wave 2 of the UK Household Longitudinal Study Innovation Panel, a survey in which the objective is to interview each adult member of the household. One protocol involves making intensive efforts to interview each household member by telephone before switching to face-to-face, and the other involves switching a household to face-to-face as soon as it is apparent that an interviewer visit will be needed for at least one household member. I assess effects on response at waves 2, 3, and 4. With both protocols response rate is lower at wave 2 than with face-to-face single-mode data collection, but with the protocol involving intensive efforts, this response differential is eroded by wave 4, whereas with the other protocol the difference remains. This difference in effect may be caused by intra-household communications.

SummarySurvey organizations often attempt to ‘convert’ sample members who refuse to take part in a survey. Persuasive techniques are used in an effort to change the refusers’ minds and to agree to an interview. This is done to improve the response rate and, possibly, to reduce non-response bias. However, refusal conversion attempts are expensive and must be justified. Previous studies of the effects of refusal conversion attempts are few and have been restricted to cross-sectional surveys. The criteria for ‘success’ of a refusal conversion attempt are different in a longitudinal survey, where for many purposes the researcher requires complete data over multiple waves. The paper uses data from the British Household Panel Survey from 1994 to 2003 to assess the long-term effectiveness of refusal conversion procedures in terms of sample sizes, sample composition and data quality.

We have demonstrated that the use of PDI results in lower levels of observed change in each of a number of characteristics of the respondent's employment. The reduction in observed change is particularly pronounced for SOC, SIC, and the number of employees at the respondent's workplace. This reduction in observed change appears to represent a reduction in measurement error as the effect of PDI is particularly pronounced amongst respondents who have not reported a change in job between survey waves. Levels of change in employment characteristics amongst INDI respondents who have not reported a change in job remain implausibly high. Furthermore, it seems that the reduction in measurement error brought about by PDI is particularly associated with certain employment characteristics. Finally, we found that measurement error was particularly reduced amongst respondents aged 36 or over and amongst the most highly qualified respondents. In attempting to summarize the nature of the characteristics, both employment and demographic, associated with a propensity for PDI to reduce measurement error in estimates of change, we would suggest that many of these characteristics may be associated with an increased propensity to have complex jobs, the characteristics of which are difficult to describe, or are associated with increased uncertainty on the part of the respondent regarding the information being sought. Complex jobs are likely to be more common amongst more highly qualified people, those in managerial, administrative or professional occupations and those working at large workplaces and in the private sector. These are the groups amongst whom PDI reduced the observed levels of change in SOC and SIC. The more difficult it is to describe a particular characteristic of a job, the more likely it is that it will be described inconsistently in two survey interviews some time apart. This leads to measurement error in estimates of change. Similarly, number of employees at the workplace may be something that respondents are less likely to know accurately if the workplace is large and if they are in the public sector (where workplaces may be larger on average, and also more impersonal). Thus, perhaps what we are observing here is simply that survey respondents who are more prone to measurement error with traditional independent interviewing will be more likely to have reduced measurement error if interviewed using PDI. An alternative explanation might be that PDI reduces the error disproportionately across sample subgroups due to differences in the way the change in interviewing method affects the cognitive processes during the interview. We find no obvious support for this. For example, if this explanation held we might expect to find a greater reduction in measurement error amongst respondents with lower levels of qualifications. We find the opposite. However, we would also note that such an effect might be difficult to identify due to the association of qualifications with employment characteristics. For example, people with lower qualifications may indeed be prone to greater levels of measurement error other things being equal, only for this effect to be outweighed by the greater tendency of people with lower qualifications to be in occupations that are easy to describe and easy for coders to recognize - that is occupations that are prone to lower levels of measurement error. We conclude that PDI is an effective questioning method for the reduction of measurement error associated with measures of change in employment characteristics. Additionally, measurement error in measures of change from traditional independently-collected interviews is likely to be greater amongst occupations that are difficult to describe consistently. This is likely to include jobs where employees have a range of varied tasks, as opposed to routine jobs involving a single task, or a limited number of closely-related tasks. Data analysts might be advised to allow for this possibility. © The Author 2006. Published by Oxford University Press on behalf of The World Association for Public Opinion Research. All rights reserved.

Summary In 1992, the British Crime Survey used the postcode address file as the sampling frame for the first time, having previously used the electoral registers (ERs). This paper addresses concerns that the change in frame may have affected the survey data, particularly measures of change over time in victimization rates. The general conclusion is that the change may have caused increases in crime rates between 1988 and 1992 to be slightly overestimated, but only by an amount which is very small compared with other sources of survey error. Information is also provided on the nature of the biased coverage of the ERs, and the population subgroups that would be under-represented by surveys involving ER samples.