Labour market processes in Tanzania constitute an important but an under-researched topic. This study investigates the recruitment methods of private companies in Mwanza, Tanzania's second largest city. It asks whether employers make use of informal methods more often than formal methods, whether the skills required for a job relate to the choice of methods and whether the vacancy period of a position is linked to a specific approach. A survey consisting of 81 face-to-face interviews with hiring authorities shows that employers prefer informal to formal schemes but tend to rely on formal ones for filling high-ranking positions. Statistically, no influence of the recruitment method on the vacancy period could be found. Additional insights are provided by 10 semi-structured follow-up interviews with respondents from the same group. They suggest an increase in solicited and unsolicited applications that might have caused some hiring authorities to avoid formal methods or modify informal methods. Moreover, it emerges that recruitment choices may be influenced by powerful actors outside or within companies. Future research should explore the benefits and risks of specific recruitment methods as related to the socio-economic context in which they are used, changes in the repertoire of recruitment methods and the role of various actors as potential codeterminants of recruitment methods, especially for lower ranks.

Universitaten und andere Hochschulen haben pragende Wirkung auf die Stadte und Regionen, in denen sie angesiedelt sind. Neben unmittelbaren Effekten als groser Arbeitgeber und Nachfrager nach Gutern und Dienstleistungen ergeben sich mittelbare Effekte durch die Ausgaben der Beschaftigten und Studierenden, die zumindest teilweise ebenfalls regional wirksam werden. Daruber hinaus ergeben sich strukturelle Wirkungen, unter anderem auf die demographische Entwicklung, die regionale Akkumulation von Humankapital, das Grundungsgeschehen und weitere „weiche“ Standortfaktoren wie etwa das kulturelle Angebot. Dieser Beitrag ist auf die Nachfrageeffekte fokussiert. Fur die Stadt Giesen mit der Justus-Liebig-Universitat Giesen und dem Hauptstandort der Technischen Hochschule Mittelhessen werden diese Nachfrageeffekte quantifiziert, wobei auch auf die Unsicherheiten derartiger Berechnungen eingegangen wird. Es zeigt sich, dass – selbst bei einer relativ kleinen Hochschulregion mit entsprechend ausgepragten Verflechtungen nach ausen – die unmittelbaren und mittelbaren Effekte am Standort das Volumen der eingesetzten offentlichen Mittel deutlich ubersteigen.

Survey data are potentially affected by interviewer falsifications with data fabrication being the most blatant form. Even a small number of fabricated interviews might seriously impair the results of further empirical analysis. Besides reinterviews, some statistical approaches have been proposed for identifying this type of fraudulent behaviour. With the help of a small dataset, this paper demonstrates how cluster analysis, which is not commonly employed in this context, might be used to identify interviewers who falsify their work assignments. Several indicators are combined to classify 'at risk' interviewers based solely on the data collected. This multivariate classification seems superior to the application of a single indicator Benford's law.

The literature on enterprises in Sub-Saharan Africa provides evidence that there are significant differences between companies run by members of the majority population and those run by members of minorities. Differences are frequently related to size, age and certain success indicators. However, it remains unclear whether decisions concerning the acquisition of personnel also diverge. This article outlines results of a questionnaire survey on recruitment methods of enterprises in Tanzania. The authors tried to discover differences in the recruitment strategies of Tanzanian-African and Tanzanian-Asian companies, but found none. The interpretation is that companies operate in similar business environments and face comparable exogenously given institutional restrictions. Thus, strategies of personnel recruitment seem to be alike.

This paper empirically investigates how migration and the receipt of remittances affect educational outcomes in Haiti. Based on a theoretical approach it tries to disentangle the effects of both phenomena that have mostly been jointly modeled in previous literature. The results suggest that remittances play an important role for poor households in alleviating budget constraints. Household wealth, captured via an asset index, is found to have a significant impact on education as well, supporting the idea that budget constraints play a crucial role in schooling decisions in Haiti.

Survey data are potentially affected by cheating interviewers. Even a small number of fabricated interviews might seriously impair the results of further empirical analysis. Besides reinterviews some statistical approaches have been proposed for identifying fabrication of interviews. As a novel toolin this context, cluster and discriminant analysis are used. Several indicators are combined to classify 'at risk' interviewers based solely on the collected data. An application to a dataset with known cases of cheating interviewers demonstrates that the methods are able to identify the cheating interviewers with a high probability. The multivariate classiffication is superior to the application of a singleindicator such as Benford's law.

Survey data are potentially affected by cheating interviewers. Even a small number of fabricated interviews might seriously impair the results of further empirical analysis. Besides reinterviews some statistical approaches have been proposed for identifying fabrication of interviews. As a novel tool in this context, cluster and discriminant analysis are used. Several indicators are combined to classify ‘at risk’ interviewers based solely on the collected data. An application to a dataset with known cases of cheating interviewers demonstrates that the methods are able to identify the cheating interviewers with a high probability. The multivariate classification is superior to the application of a single indicator such as Benford’s law.

Sebastian Bredl, Peter Winker and Kerstin Kötschau A statistical approach to detect interviewer falsification of survey data ......................1 Steven Elliott The application of graph theory to the development and testing of survey instruments ........................................................................................................ 11 G. Hussain Choudhry, J.N.K. Rao and Michael A. Hidiroglou On sample allocation for efficient domain estimation.................................................. 23 Ted Chang Calibration alternatives to poststratification for doubly classified data ...................... 31 Paul Knottnerus and Arnout van Delden On variances of changes estimated from rotating panels and dynamic strata ............ 43 Dan Liao and Richard Valliant Variance inflation factors in the analysis of complex survey data .............................. 53 Hung-Mo Lin, Hae-Young Kim, John M. Williamson and Virginia M. Lesser Estimating agreement coefficients from sample survey data ...................................... 63 Jörg Drechsler and Jerome P. Reiter Combining synthetic data with subsampling to create public use microdata files for large scale surveys .......................................................................................... 73 Balgobin Nandram and Myron Katzoff A hierarchical Bayesian nonresponse model for two-way categorical data from small areas with uncertainty about ignorability .................................................... 81