Anders Holm and colleagues examine the challenges of explaining variance in multilevel models. Their method helps researchers better decompose variance components and account for level 1 and level 2 predictors.
https://t.co/CjmR0g2iWi
R. Gordon Rinderknecht and colleagues assess the daily lives of on MTurk and Prolific respondents. Their findings urge researchers to consider how differing daily routines might affect study outcomes when using crowdsourcing samples.
https://t.co/EjH3Ijsokw
Jaclyn S. Wong and colleagues explore the challenges and opportunities of using "other, describe" open-text boxes in survey questionnaires, demonstrating that there's no one-size-fits-all approach to these complex decisions.
https://t.co/5cdf6dpfBi
New study by Anna-Carolina Haensch and Reinhard Schunck explores using external data to improve accuracy in surveys where secondary respondent info is often missing.
https://t.co/EWq9hYeAmN
Moeen Mostafavi and colleagues introduce BERTNN: a breakthrough in expanding affective lexicons using Bidirectional Encoder Representations from Transformers. It enhances sociological research by efficiently estimating the emotional context of new words.
https://t.co/LhZygml271
New study by Siwei Cheng and colleagues introduces a method to measure economic polarization, focusing on changes within job sectors and geographic areas, offering insights into wage and income disparities.
https://t.co/9FTmTwHL5t
Exploring topic modeling validation, Bolun Zhang and colleagues challenge current methods that rely on human interpretation and propose a new procedure to ensure consistent results across semantically similar analyses.
https://t.co/JVDhxsKmV5
Evaluating nonprobability vs. probability surveys, Björn Rohr and colleagues find that nonprobability surveys are cost-effective, show less accuracy in univariate measures compared to probability surveys but perform similarly in multivariate analyses.
https://t.co/vUGtOIzMyV
Brian C. Kelly and colleagues discuss event-centered interviewing, an innovative approach that uses smartphone apps to enhance traditional qualitative interviews.
https://t.co/6vrw9InILK
OnlineFirst: Sarah Brothers and colleagues explore new frontiers in community-engaged research, this article delves into a virtual community-driven research project with the National Survivors Union during COVID-19.
https://t.co/wOxnPAp9n0
Volume 54 Issue 2: Zsófia Papp and colleagues examine question-order effects in measuring satisfaction with democracy (SWD). They recommend placing SWD questions before state of the economy (SWE) questions or randomize question order.
https://t.co/TmFeHAdUmX
Volume 54 Issue 2: Jessica P. Kunke and colleagues take the network scale-up method (NSUM) to estimate the size or prevalence of hard to reach populations.
https://t.co/SfEwV5Cz7Q
Volume 54 Issue 2: Kazuo Yamagushi and Jesse Zhou introduce a new group of multinomial logit models to identify covariate effects on dependent variables that are associated with each other. An example of modeling occupational attainments is offered.
https://t.co/onYpqpYwNI
Volume 54 Issue 2: @calliehburt offers a primer on Polygenic Indices (PGI), the new standard for genetic summary scales in social science research.
https://t.co/C7aMx1htxx
Volume 54 Issue 2: Scott W. Duxbury uses the average mediated micro effect (AMME) to re-examine mediation analysis in social networks, taking into account both micro and macro outcomes.
https://t.co/xblwTjVW4h
Volume 54 Issue 2: @MarkDVerhagen proposes a machine learning driven framework to revolutionize how quantitative sociologists specify their models.
https://t.co/1Cp3EhdWT3
Volume 54 Issue 2: Inge Kryger Pedersen and Anders Blok integrate cross-case comparison with abductive analysis through an illustration of transnational professional boundary work.
https://t.co/46YdLW3CXp
Vol 51 issue 2: are you using fixed and random effects models and wish someone would provide a general model and summarize what to do based on your research q, then look no further than Scott W Duxbury’s article https://t.co/380F3wawjc
And the #RCode for the @SocMethod article (validation of #sequenceanalysis typologies) is available in WeightedCluster development version
To install see https://t.co/vvLxCjIqR9
Then in R:
library(WeightedCluster)
help(seqnullcqi)
example(seqnullcqi)