7739 research outputs found
Sort by
Using AI for Qualitative Labeling: Consistency and Comparisons
Get PDFThis paper continues research that evaluates the capacity of artificial intelligence (AI) to perform qualitative coding tasks. The previous study found that AI models lacked consistency with themselves and did not agree with human coded data. Since that study, AI’s general level of intelligence has increased. Hence, this study re-evaluates how well the newest set of AI models (Claude 3 and Gemini) can perform qualitative coding tasks. When tested, the new AI models perform about the same or better than previous models depending on the metric tested. While Gemini and Claude 3 do not agree with human output any more than previous models, they do agree with each other slightly more, and they agree with themselves substantially more, as shown by the Kappa statistic. Disagreement still exists around codewords that lack clear distinction from one another, such as human, social, and cultural codewords. However, overall model consistency has improved, so different outputs using the same AI are likely to agree. Although the use of AI has not reached human level abilities, they possess potential to expedite qualitative coding as a useful tool
The Information Content of Earnings: The Effects of a Change in Accounting Approach
No full textThis study leverages the adoption of ASC 606, “Revenue from Contracts with Customers,” to examine the effects of transitioning from a rules-based to a principles-based approach on the information content of earnings along with two of its most important attributes: predictive value and asymmetry. By introducing a common principles-based framework, ASC 606 resolves inconsistencies found in earlier rules-based standards and fosters uniformity in financial reporting. However, this change might also diminish the uniqueness of financial information, a critical factor in the decision-making process. The findings reveal a significant increase in the overall market\u27s information content of earnings post-ASC 606 implementation. Industries with complex revenue recognition processes and long-term contracts, such as information technology, real estate, and communications, experienced the most significant improvements. These industries benefited from the increased flexibility and professional judgment allowed under ASC 606, resulting in more accurate future earnings forecasts. However, the study also finds that ASC 606 has increased information asymmetry, particularly in industries with complex transactions. This suggests that while the principles-based approach enhances the informativeness and predictive value of earnings, it also introduces challenges in interpreting and applying the new standards, leading to increased information asymmetry. These findings have important implications for standard setters, regulators, and practitioners in financial reporting
Meatless Revolution: Exploring Consumers’ Perception of Meat Alternatives
No full textThis paper employs a 4 × 2 × 3 research design with four different antecedents of meat alternative consumption (red meat consumption, consumers’ health consciousness, environmental awareness, animal welfare), two types of meat alternatives (plant-based vs. vegetarian), and three consumer behavioral outcomes (purchase intention, willingness to substitute, and willingness to pay). Our study of 425 participants shows that all four antecedents affect consumers’ meat alternative perceptions. Red meat consumption has a negative effect, while health consciousness, environmental awareness, and animal welfare have a positive relationship, increasing meat alternative purchase intention, willingness to substitute, and willingness to pay. Interestingly, plant-based alternatives had higher purchase intention, willingness to substitute, and willingness to pay relative to vegetarian alternatives. Our results estimate that plant-based alternatives are substituted for up to 50% of red meat consumption, compared to 35% for vegetarian alternatives. Also noteworthy, plant-based meat alternatives have a 10% higher willingness to pay versus vegetarian alternatives
Developmental differences in reported speech and internal state language in preschoolers’ personal narratives
Get PDFThe present study explored developmental differences in preschoolers’ use of reported speech and internal state language in personal narratives. Three-, four-, and five-year-olds attending a laboratory preschool shared 204 stories about ‘a time when you were happy/sad’. Stories were audio-recorded, transcribed, and coded for reported speech (direct, indirect, narrativized) and internal state language (cognitive states, total emotion terms, unique emotion terms). Personal narratives told by five-year-olds included more cognitive states and more narrativized speech than those told by three- and four-year-olds, even when accounting for children’s vocabulary skills, and that reported speech (narrativized, direct) were positively correlated with cognitive state talk. These findings highlight distinct shifts in children’s use of cognitive state talk and reported speech in personal narratives told at age five. Associations between reported speech and internal state language are both informed by and support Vygotsky’s (1978) fundamental claim that psychological processes are socially mediated by language
ChatGPT might slow you down: Verification moderates AI uses and learning efficiency among college students
Get PDFGenerative artificial intelligence without a doubt is one of the most discussed topics in high-er education. It is still unclear whether ChatGPT is contributing to or hindering students’ learning efficiency. Built on the Technology Acceptance Model (TAM), this study is designed to examine the influence of ChatGPT usage of college stu-dents on their learning efficiency; to investigate the moderating effects of verification on the relationship between ChatGPT usage and learn-ing efficiency; and to investigate the relationship among ChatGPT usage and its antecedents, in-cluding subjective norm, job relevance, perceived usefulness, perceived ease of use, and intention to use. Using survey data from Prolific (N=623), structural equation modeling results indicated that without verification, ChatGPT usage among college students negatively impacted their learn-ing efficiency. However, the interaction of ChatG-PT usage and verification positively contributed to students’ learning efficiency
“Trying to remain calm…but I do reach my limit sometimes”: An exploration of the meaning of gentle parenting
Get PDFRaising young children has always been hard, but evidence suggests that it may be getting harder. The isolation of the pandemic, the pressures to fulfill exacting parenting standards, and the explosion of “expert” parenting advice on social media have fueled the rise of “gentle parenting,” an approach that pivots away from older, discipline-heavy parenting typologies and which promises the development of happier, healthier children. Despite the popularity of gentle parenting, it has received no empirical scrutiny. The current study represents the first systematic investigation of what gentle parenting entails. Data were gathered from a sample (N = 100) of parents of at least one child between the ages of 2 and 7 from the Midwest, Southeast, and Southwest. Approximately half (n = 49) of the sample identified as “gentle parents.” Inductive analyses identified this approach as one that emphasizes high levels of parental affection and parents’ and children’s emotion regulation. Gentle parenting appears to be distinct from other established measures of parenting approaches in its emphasis on boundaries, yet the enactment of those boundaries is not uniform. Overall, gentle parents reported high levels of parenting satisfaction and efficacy, but a subset of gentle parents who were highly critical of themselves reported significantly lower levels of efficacy than the rest of the sample. Statements of parenting uncertainty and burnout were present in over one-third of the gentle parent sample. Implications are discussed for future research and increased support for those who identify as gentle parents
How To Be a Large Child
No full textMany years ago someone was talking bad about me, and called me a large child. I thought that was awesome and turned it into a whole series of photos, videos and this book. Embrace your inner large child. — Paul Shortthttps://scholarship.rollins.edu/book_arts/1148/thumbnail.jp
An Evaluation of Rate Calculations During Extinction
Get PDFIn practice, extinction is implemented to reduce a problem behavior by terminating reinforcement. It has historically been calculated by dividing the frequency of responses in a session by the total duration of the session. When calculated this way, the data appear as a sharp increasing trend before a downward trend begins. This is referred to as an extinction burst and is often expected to decrease with consistent implementation by practitioners who say, “It must get worse before it gets better”. The motivating operation to respond would be low for a participant who is currently receiving reinforcement. Due to this reasoning, during reinforcement or consumption time the rate of responding decreases. The current study compared two response rate calculations: one which includes consumption time in the total session duration and one which excludes consumption time in the total session time for extinction procedures. The hypothesis was that either calculation would have an effect on the rate of responding. Specifically, when the calculation which excludes consumption time is used, the rate of responding will be higher than when the calculation which includes consumption time is used. The purpose of this study was to show that if consumption time was removed from the equation, extinction bursts would occur at a lower percentage than practitioners think. The two calculations were compared to determine the prevalence of extinction bursts.
Keywords: denominator, duration, extinction, extinction burst, intervene, maladaptive, reinforce, tren