Explore the vast range of titles available.

There is evidence that survey interviewers may be tempted to manipulate answers to filter questions in a way that minimizes the number of follow-up questions. This becomes relevant when ego-centered network data are collected. The reported network size has a huge impact on interview duration if multiple questions on each alter are triggered. We analyze interviewer effects on a network-size question in the mixed-mode survey “Panel Study ‘Labour Market and Social Security’” (PASS), where interviewers could skip up to 15 follow-up questions by generating small networks. Applying multilevel models, we find almost no interviewer effects in CATI mode, where interviewers are paid by the hour and frequently supervised. In CAPI, however, where interviewers are paid by case and no close supervision is possible, we find strong interviewer effects on network size. As the area-specific network size is known from telephone mode, where allocation to interviewers is random, interviewer and area effects can be separated. Furthermore, a difference-in-difference analysis reveals the negative effect of introducing the follow-up questions in Wave 3 on CAPI network size. Attempting to explain interviewer effects we neither find significant main effects of experience within a wave, nor significantly different slopes between interviewers.

Falsified interviews represent a serious threat to empirical research based on survey data. The identification of such cases is important to ensure data quality. Applying cluster analysis to a set of indicators helps to identify suspicious interviewers when a substantial share of all of their interviews are complete falsifications, as shown by previous research. This analysis is extended to the case when only a share of questions within all interviews provided by an interviewer is fabricated. The assessment is based on synthetic datasets with a priori set properties. These are constructed from a unique experimental dataset containing both real and fabricated data for each respondent. Such a bootstrap approach makes it possible to evaluate the robustness of the method when the share of fabricated answers per interview decreases. The results indicate a substantial loss of discriminatory power in the standard cluster analysis if the share of fabricated answers within an interview becomes small. Using a novel cluster method which allows imposing constraints on cluster sizes, performance can be improved, in particular when only few falsifiers are present. This new approach will help to increase the robustness of survey data by detecting potential falsifiers more reliably.

What should researchers do when their baseline model is falsified? We recommend reporting the set of parameters that are consistent with minimally nonfalsified models. We call this the falsification adaptive set (FAS). This set generalizes the standard baseline estimand to account for possible falsification. Importantly, it does not require the researcher to select or calibrate sensitivity parameters. In the classical linear IV model with multiple instruments, we show that the FAS has a simple closed-form expression that only depends on a few 2SLS coefficients. We apply our results to an empirical study of roads and trade. We show how the FAS complements traditional overidentification tests by summarizing the variation in estimates obtained from alternative nonfalsified models.