focalinterest.net

Tag: General (Business & Management)

  • CSR shifts earnings management form under creditor oversight

    What the study found

    The study found that corporate social responsibility (CSR) does not uniformly reduce earnings management. Instead, CSR is linked to different patterns of accrual-based earnings management and real earnings management, and these patterns change with the level of creditor oversight.

    Why the authors say this matters

    The authors conclude that CSR may shift managerial opportunism across reporting channels rather than simply restraining it. They suggest this matters because monitoring intensity can shape whether managers rely more on accounting choices or on real business actions.

    What the researchers tested

    The researchers analyzed 92 Vietnamese manufacturing firms from 2017 to 2021 using panel data. They distinguished between accrual-based earnings management and real earnings management, and they used fixed effects, random effects, feasible generalized least squares, and instrumental variable GMM to address unobserved heterogeneity and endogeneity.

    What worked and what didn't

    Under weak creditor oversight, CSR was associated with higher accrual-based earnings management and lower real earnings management. Under stronger creditor oversight, leverage was associated with lower accrual-based earnings management in the validated specifications, while the real earnings management results suggested partial reallocation toward less observable real activity manipulation.

    What to keep in mind

    The study is based on manufacturing firms in Vietnam, so the results are limited to that sample and setting. The abstract also notes that the suspect-firm subsample analysis supports the proposed substitution mechanism, but it does not provide additional detail in the available summary.

    • CSR was not a uniform constraint on earnings management.
    • Weak creditor oversight was linked to higher accrual-based earnings management and lower real earnings management.
    • Stronger creditor oversight was associated with lower accrual-based earnings management in validated specifications.
    • The results suggested partial shifting toward less observable real activity manipulation under stronger monitoring.
    • A suspect-firm subsample analysis was consistent with the substitution mechanism.

  • Culture and institutions shape goodwill impairment under IFRS

    What the study found

    The study found that goodwill impairment losses under IFRS are linked to both cultural differences and institutional quality across countries. It also found that stronger institutional quality is associated with larger impairment amounts and that firms in stronger institutional settings report impairments that more closely reflect underlying economic conditions.

    Why the authors say this matters

    The authors conclude that goodwill impairment under IFRS is not only an accounting issue but is also shaped by the institutional and cultural environment in which firms operate. They say that revising accounting standards alone is not enough for comparable reporting, and that consistent enforcement across jurisdictions remains essential.

    What the researchers tested

    The researchers analyzed 2,466 companies from 17 IFRS-adopting countries, using 14,898 firm-year observations from 2007 to 2013. They estimated a pooled Tobit model with year and industry fixed effects, using Hofstede's cultural dimensions and validated institutional quality measures from Alshehabi et al. (2021).

    What worked and what didn't

    Goodwill impairment losses were associated with firm-level economic factors and managerial reporting incentives. Power distance and individualism had direct effects in the predicted direction, while uncertainty avoidance was positively associated with impairment, which the authors interpret as consistent with income smoothing. Institutional quality, especially quality of legality and equity market development, was positively associated with impairment magnitude, and moderation analyses showed that stronger institutions reduced the sensitivity of impairment to goodwill carrying values.

    What to keep in mind

    The study used firm-level data rather than cash-generating-unit-level data. The sample period was limited by the availability of the validated institutional measures, and the authors note that their results provide a pre-2014 enforcement baseline for future research on later regulatory changes.

    • Goodwill impairment losses under IFRS were associated with both culture and institutional quality.
    • Power distance and individualism showed direct effects on impairment amounts in the predicted direction.
    • Uncertainty avoidance was positively associated with impairment, which the authors link to income smoothing.
    • Higher institutional quality, especially legality and equity market development, was associated with larger impairment magnitude.
    • Stronger institutional environments were linked to impairment reports that more closely reflected economic conditions.

  • Technologies supported audit capabilities during COVID-19 turbulence

    What the study found

    The study found that technologies acted as catalysts for dynamic audit capabilities during turbulence. It also found that audit firm size and perceived relevance affected how efficient the technology was considered to be.

    Why the authors say this matters

    The authors conclude that audit firms need to develop and anchor their operations through dynamic capabilities made possible by technologies in order to navigate a crisis. They also say the results may be useful for audit regulators, auditors, and audit firms in understanding technology use during turbulent times.

    What the researchers tested

    The researchers studied auditors in Sweden during the COVID-19 pandemic, a setting where no regulatory exemptions were granted. They analyzed data from 237 auditors using descriptive statistics, regression analysis, and the Mann–Whitney U test.

    What worked and what didn't

    Technologies were reported as helping dynamic audit capabilities during turbulence. The perceived efficiency of technology was associated with audit firm size and with whether the technology was seen as relevant. The abstract does not report any specific technology that did not work.

    What to keep in mind

    The authors note that the study is based on the COVID-19 pandemic and may not apply to other events, such as a financial crisis. The abstract does not describe additional limitations.

    • Technologies were found to act as catalysts for dynamic audit capabilities during turbulence.
    • Audit firm size influenced the perceived efficiency of technology.
    • Perceived relevance also affected how efficient technology was seen to be.
    • The study used data from 237 auditors in Sweden during the COVID-19 pandemic.
    • The authors say the findings may not apply to events other than the pandemic.

  • Brexit sentiment was linked to weaker UK markets

    What the study found

    The study found that Brexit-related social media sentiment had a significant negative impact on the UK stock market and exchange rate performance. It also found only a minor gap between opponents and proponents of Brexit, which matched the 2016 referendum results.

    Why the authors say this matters

    The authors conclude that sentiment analysis, meaning the study of opinions expressed on social media, can be used to understand public opinion and market performance. They say the research adds to the literature by examining the long-term relationship between Twitter/X sentiment and market performance.

    What the researchers tested

    The researchers used time-series analysis to study the relationship between Brexit sentiments and UK market outcomes after the decision. They examined GBP exchange rates against the U.S. dollar and the euro, the FTSE-100 Index, and overall social media sentiment.

    What worked and what didn't

    The analysis showed a minor difference between Brexit opponents and proponents, but this was associated with a significant negative impact on UK stock market performance and exchange rates. The abstract does not report any results that worked positively or any null findings beyond this comparison.

    What to keep in mind

    The available summary does not describe detailed limitations, sample size, or the specific social media data used beyond Twitter/X. It also does not provide the underlying statistical results or explain the size of the negative impact in numerical terms.

    • Brexit-related social media sentiment was associated with worse UK stock market and currency exchange performance.
    • The study found only a minor gap between Brexit opponents and proponents, matching the 2016 referendum result.
    • The researchers analyzed GBP exchange rates against the U.S. dollar and euro, plus the FTSE-100 Index.
    • The authors frame sentiment analysis as a way to study public opinion and market performance.
    • The abstract does not report detailed limitations or numerical effect sizes.

  • AI and ML are reshaping international business research

    What the study found

    The authors argue that AI and ML can transform international business research by making it possible to analyze large-scale, multimodal data and detect patterns relevant to theory and evidence. They also present a structured roadmap for bringing these techniques into international business research.

    Why the authors say this matters

    The study suggests that AI and ML are not just analytical tools but may be transformative for the future of international business research. The authors say this matters because linking methodological innovation with conceptual advancement can support new work on international business topics such as foreignness, legitimacy, and deglobalization.

    What the researchers tested

    This is a research article that reviews AI- and ML-based techniques for international business research. The paper covers supervised methods, unsupervised methods, generative AI, and multimodal approaches, and it discusses how these can be applied to core international business constructs.

    What worked and what didn't

    The paper says these methods can enrich understanding of foreignness, legitimacy, internationalization strategy, corporate governance, distance, and deglobalization. It also notes that the methodological breadth and technical complexity of AI and ML create significant challenges for many international business scholars.

    What to keep in mind

    The abstract does not report empirical testing or specific quantitative results. It also does not provide detailed limitations beyond noting the technical and methodological challenges of integrating AI and ML into international business research.

    • AI and ML are presented as tools that can analyze large-scale, multimodal data in international business research.
    • The paper offers a structured roadmap for integrating AI- and ML-based techniques into the field.
    • The authors review supervised, unsupervised, generative AI, and multimodal approaches.
    • The paper says these methods can enrich constructs such as foreignness, legitimacy, and deglobalization.
    • The abstract notes that AI and ML pose methodological and technical challenges for many scholars.

  • Spreadsheets can support classroom psychometric analysis

    What the study found

    The study found that spreadsheet software can be turned into a practical environment for psychometric analysis in classroom assessment. It showed that teachers can use spreadsheets to go beyond total scores and examine item-level information.

    Why the authors say this matters

    The authors conclude that this approach can help teachers interpret test data more fully, identify weak items, refine distractors, and build small item banks aligned with competence-based curricula. They also say the approach contributes to Sustainable Development Goal 4, which concerns accessible, equitable, and high-quality education.

    What the researchers tested

    The researchers presented a methodological demonstration using a simulated dataset of 40 students answering 20 dichotomous items, meaning items with two possible responses. They used spreadsheet formulas to calculate descriptive statistics and Classical Test Theory indices, including item difficulty, discrimination, and corrected item-total correlations, and then extended the demonstration to Item Response Theory with 1PL, 2PL, and 3PL logistic models using forward-calculated item parameters.

    What worked and what didn't

    The results show that spreadsheets can support the interpretation of test data beyond total scores. The demonstration also illustrated the interpretability of the indices and the generation of Item Characteristic Curves, which are graphs showing how likely a student is to answer an item correctly at different ability levels. The abstract does not report comparative performance against other tools.

    What to keep in mind

    The authors note that Item Response Theory parameter estimation can be unstable in small samples. They also say teacher training is needed, and they call for future research using real classroom data, more automation in spreadsheet environments, and examination of artificial intelligence for adaptive assessment.

    • Spreadsheet software was shown as a practical environment for psychometric analysis.
    • The demonstration used simulated data from 40 students and 20 dichotomous items.
    • Classical Test Theory indices included item difficulty, discrimination, and corrected item-total correlations.
    • Item Response Theory was illustrated with 1PL, 2PL, and 3PL logistic models.
    • The authors say the approach can help identify weak items and refine distractors.
    • The abstract notes instability of Item Response Theory estimates in small samples.

  • Financial literacy and trust shape trusted-contact adoption

    What the study found

    The study found that fewer than 38% of eligible U.S. investors had ever adopted a trusted contact, which is a safeguard under FINRA Rule 4512 that lets an investor voluntarily name a trusted person. The authors report that financial literacy, portfolio sophistication, and social trust are important determinants of whether people adopt this safeguard.

    Why the authors say this matters

    The authors suggest the findings point to substantial frictions in precautionary financial behavior, meaning some investors do not take a simple protective step even when it is available. They conclude that social capital, which refers to trust and connections within a community, strongly shapes how financial knowledge relates to adoption.

    What the researchers tested

    The researchers used microdata from the 2021 National Financial Capability Study. They examined both naming a trusted contact and being named by others, and they addressed endogeneity, a situation where one factor may be tied to another in ways that complicate causal interpretation, by using exposure to mandatory high-school financial education as an instrument for financial literacy.

    What worked and what didn't

    Financial literacy had a large causal effect on compliance. Portfolio sophistication predicted both naming a trusted contact and being named by others, and social capital strongly moderated these relationships. In low-trust environments, literacy weakly predicted or even deterred delegation, while in high-trust regions it strongly increased adoption and peer recognition.

    What to keep in mind

    The summary does not describe other limitations beyond the study's focus on U.S. investors in the 2021 National Financial Capability Study. The results are specific to the behaviors studied: naming a trusted contact and being named by others.

    • Fewer than 38% of eligible U.S. investors had ever named a trusted contact.
    • Financial literacy had a large causal effect on compliance with the trusted-contact safeguard.
    • Portfolio sophistication predicted both naming a trusted contact and being named by others.
    • Social capital strongly changed how financial literacy related to adoption.
    • In low-trust areas, literacy weakly predicted or deterred delegation; in high-trust regions, it strongly increased adoption and peer recognition.

  • Adaptive formative assessment showed high estimation accuracy

    What the study found

    The study found that a formative adaptive assessment framework for engineering education can provide competency-oriented feedback, learning monitoring, and instructional interpretation. It also showed high estimation accuracy and satisfactory reliability for formative use across most learner profiles.

    Why the authors say this matters

    The authors conclude that adaptive assessment can be a pedagogically grounded tool for formative learning support, instructional decision-making, and quality assurance in engineering education. They frame this as important because assessment in this setting should support learning regulation and educational quality, not only measurement efficiency.

    What the researchers tested

    The researchers proposed and evaluated a formative adaptive assessment framework that combines an item response theory (IRT) computer-adaptive testing engine with a Bayesian network diagnostic component. The framework used dichotomous multiple-choice items aligned with engineering learning outcomes, with item calibration based on data from 612 university students in computer science and a simulation study involving 500 simulated learners.

    What worked and what didn't

    The results showed high estimation accuracy, with r = 0.912, and satisfactory reliability for formative use across most learner profiles. Reduced precision at the extremes of the proficiency continuum and imbalances in item exposure were also observed.

    What to keep in mind

    The abstract says the main structural limits were tied to item bank coverage and curriculum representation rather than to the adaptive algorithms themselves. No other limitations are described in the available summary.

    • The framework combined IRT-based computer-adaptive testing with Bayesian network diagnostic modelling.
    • The study reported high estimation accuracy, with r = 0.912.
    • Reliability was described as satisfactory for formative use across most learner profiles.
    • Reduced precision appeared at the extremes of the proficiency continuum.
    • Item exposure was imbalanced, mainly because of item bank coverage and curriculum representation.