MOWDOC: A Dataset of Documents From Taking the Measure of Work for Building a Latent Semantic Analysis Space

Measuring the menu, not the food: “psychometric” data may instead measure “lingometrics” (and miss its greatest potential)

“…If we compare this with the performance in the previously cited articles, we can see that the language algorithms are able to predict data patterns that range from construct definition levels via item correspondent levels ( Larsen et al, 2013 ; Rosenbusch et al, 2020 ; Nimon, 2021 ), down to patterns in observed statistics bearing on construct relationships and correlation patterns from human respondents ( Arnulf et al, 2014 ; Nimon et al, 2016 ; Gefen and Larsen, 2017 ; Arnulf et al, 2018a , b , c , d ).…”

“…At the moment of writing, studies demonstrating semantically predictable research findings and picking up at an increasing pace covering state-of-the-art research instruments in leadership and motivation ( Arnulf et al, 2014 ), engagement, job-satisfaction and well-being ( Nimon et al, 2016 ), the technology acceptance model ( Gefen and Larsen, 2017 ), job analysis ( Kobayashi et al, 2018 ), personality scale construction ( Abdurahman et al, 2023 ; Fyffe et al, 2023 ), entrepreneurship ( Freiberg and Matz, 2023 ) personality and mental health ( Kjell K. et al, 2021 ; Kjell O. et al, 2021 ) or even near-death-experiences ( Lange et al, 2015 ). Overlapping meanings between a vast group of constructs have been demonstrated ( Larsen and Bong, 2016 ) and new scales can be checked for overlaps ( Rosenbusch et al, 2020 ; Nimon, 2021 ).…”

“…When this happens, constructs are in no way independent of each other. They are simply various ways to phrase statements about working conditions that overlap in meaning – a sort of second-order jingle/jangle relationships ( Larsen and Bong, 2016 ; Nimon, 2021 ). In our example, it means that definitions and operationalizations of work imply a little bit of motivation.…”

“…In this way the semantic matrix could pose as a Bayesian prior to research in the social sciences ( Gelman and Shalizi, 2013 ). One can now compute the likely relationships among all variables prior to making empirical data collections ( Gefen and Larsen, 2017 ; Arnulf et al, 2018a ; Kjell et al, 2019 ; Gefen et al, 2020 ; Rosenbusch et al, 2020 ; Kjell K. et al, 2021 ; Nimon, 2021 ). From a statistical point of view, one should ask questions like who, how, why, and how much people will comply with what is semantically expected.…”

Preparing Students for Certifications: A Model for Assessing Curriculum Alignment to Professional Standards

“…For further discussion of using LSA in social sciences see the work of Valdez, Pickett, and Goodson (2018). Nimon (2021) also presents an excellent use case of LSA that may be helpful for readers.…”