688 research outputs found
alex-cernat/evs_mm_DE_equivalence: The Impact of Survey Mode Design and Questionnaire Length on Measurement Quality
No full textThis is the code associated with the paper: Cernat, A., Sakshaug, W., J., Christmann, P & Gummer, T. (2022). The Impact of Survey Mode Design and Questionnaire Length on Measurement Quality. Sociological Methods and Research
Do surveys change behaviours?:Evidence from digital trace data
No full textIt is well known that being observed can change our behaviour. Alexandru Cernat and Florian Keusch used digital trace data to see whether taking part in multiple political surveys made participants more interested in news and politic
Supplemental Material - Predicting Web Survey Breakoffs Using Machine Learning Models
No full textSupplemental Material for Predicting Web Survey Breakoffs Using Machine Learning Models by Zeming Chen, Alexandru Cernat, and Natalie Shlomo in Social Science Computer Review</p
Measurement error Skin colour and the role of the interviewer
No full textInterviewer effects are a well-known problem in survey research. Less well known is whether the skin colour of interviewers can affect their assessments of the skin colour of respondents. Alexandru Cernat, Joseph W. Sakshaug and Javier Castillo investigate.</p
Nurse effects in survey biomarkers | Dr Alexandru Cernat & Joe Sakshaug
No full textThis video reports on findings which hope to understand the processes that lead to missing data specifically from the 'nurse effect', in order for this to be corrected for
Measuring crime in place: Distinguishing between area victimisation and area offences
Get PDFCrime data is essential to the running of a safe society. But are we measuring crime in the best way? Ian Brunton-Smith, Alexandru Cernat, David Buil-Gil and Jose Pina-Sánchez explore how the current system could be improve
Estimating systematic response errors using the multitrait-multierror model | Dr Alexandru Cernat
No full textThis video present a new paper which proposes a new method to estimate and control for multiple types of errors concurrently, which we call the “multitrait-multierror” (MTME) approach. MTME combines the theory of experimental design with latent variable modeling to efficiently estimate response errors of different types simultaneously and evaluate which are most impactful on a given question
Supplemental Material - Do You Have Two Minutes to Talk About Your Data? Willingness to Participate and Nonparticipation Bias in Facebook Data Donation
No full textSupplemental Material for Do You Have Two Minutes to Talk About Your Data? Willingness to Participate and Nonparticipation Bias in Facebook Data Donation by Florian Keusch, Paulina K. Pankowska, Alexandru Cernat, and Ruben L. Bach in Field Methods</p
- …
