The Collaborative Library

📊 Science improves when data is shared — but not all journals are on board yet. A review of 165 medical journals found that clear data-sharing policies are still far from universal. • Only 36% allowed researchers to deposit data in repositories • Over 60% gave no clear guidance at all • Just ~9.5% of published articles included supplementary data Open science depends not only on open access, but also on sharing raw data and research materials so others can verify and build on findings. ▶ Watch the new MIRROR 8-min video summary from the Science Integrity Alliance: https://lnkd.in/eCRGp_9c 📖 Read the written version: https://lnkd.in/e5bkhdfc #MIRROR #ScienceIntegrityAlliance #OpenScience #DataSharing #ResearchTransparency #MetaScience

🐦 The Early-Bird Papers: Who Really Uses Preprints? Preprints let scientists share research before peer review, speeding up discovery and open access. But who is actually using them? A global study analysing 100+ million papers (1991–2023) found: • Only 2.2 million papers were linked to a preprint • Physics and maths dominate (~70% of preprints) • Preprinting is uneven globally — about 10% of papers in Europe vs ~5% in Africa These gaps matter: researchers who preprint may gain visibility, collaboration, and funding opportunities. This new MIRROR summary from the Science Integrity Alliance explores how preprints are spreading across disciplines and regions — and why the benefits of open science aren’t shared equally yet. ▶ Watch the 7-min video summary: https://lnkd.in/e8J3wPU3 📖 Read the full summary https://lnkd.in/ed8bqrAm #MIRROR #ScienceIntegrityAlliance #OpenScience #Preprints #MetaScience #ResearchTransparency

🤖 Is AI fuelling false findings in science? AI is transforming research — but it may also be making it easier to produce low-quality studies with misleading results. A recent analysis of papers using the large NHANES health dataset found: • A sharp rise in simple “single-factor” studies (from 4 per year to 190 in 2024) • Many papers didn’t correct for false discoveries • Some appeared to cherry-pick data from the dataset These patterns don’t prove AI caused the problem. But they raise concerns that powerful tools and large datasets could be misused to generate fast, publishable results that don’t hold up. This new MIRROR summary from the Science Integrity Alliance explores what’s happening — and how science can respond. ▶ Watch the 8-min video summary https://lnkd.in/eD9zqVrG 📖 Read the written summary https://lnkd.in/esH5HqXy #MIRROR #ScienceIntegrityAlliance #MetaScience #ResearchIntegrity #AIinScience #OpenScience

🌪️ The 2AM Scroll: Inside the Health Misinformation Storm How does health misinformation spread online — and how reliable is the research studying it? A global analysis of nearly 10,000 papers found misinformation appears widely on social media (from <1% to nearly 1 in 3 posts depending on topic). It can lead to fear, delayed care, vaccine hesitancy, and pressure on health systems. But the study uncovered another problem: When researchers assessed the quality of reviews studying misinformation, 16 out of 17 were rated critically low quality. That means many decisions about how to fight misinformation may rely on weak evidence. This new MIRROR summary from the Science Integrity Alliance explores what we know about infodemics — and why stronger research standards are urgently needed. ▶ Watch the 9-min video summary: https://lnkd.in/eyswvteE 📖 Read the written summary: https://lnkd.in/ew8zFg5m #MIRROR #ScienceIntegrityAlliance #ResearchIntegrity #MetaScience #HealthMisinformation #Infodemic

🌪️ The 2AM Scroll: Inside the Health Misinformation Storm How does health misinformation spread online — and how reliable is the research studying it? A global analysis of nearly 10,000 papers found misinformation appears widely on social media (from <1% to nearly 1 in 3 posts depending on topic). It can lead to fear, delayed care, vaccine hesitancy, and pressure on health systems. But the study uncovered another problem: When researchers assessed the quality of reviews studying misinformation, 16 out of 17 were rated critically low quality. That means many decisions about how to fight misinformation may rely on weak evidence. This new MIRROR summary from the Science Integrity Alliance explores what we know about infodemics — and why stronger research standards are urgently needed. ▶ Watch the 9-min video summary: https://lnkd.in/eyswvteE 📖 Read the written summary: https://lnkd.in/ew8zFg5m #MIRROR #ScienceIntegrityAlliance #ResearchIntegrity #MetaScience #HealthMisinformation #Infodemic

This video was especially interesting to make! 📚🤖 Recent findings show that since LLMs are getting more and more popular, an increasing number of bad quality medical studies are appearing in our scientific databases. Importantly, this doesn't mean that AI is an evil entity set to destroy us all - it rather suggests that people are misusing AI to rapidly publish many low-quality studies. Understanding what exactly is happening is an important first step if we want to fix the system promoting this behaviour.

🤖 Is AI fuelling false findings in science? AI is transforming research — but it may also be making it easier to produce low-quality studies with misleading results. A recent analysis of papers using the large NHANES health dataset found: • A sharp rise in simple “single-factor” studies (from 4 per year to 190 in 2024) • Many papers didn’t correct for false discoveries • Some appeared to cherry-pick data from the dataset These patterns don’t prove AI caused the problem. But they raise concerns that powerful tools and large datasets could be misused to generate fast, publishable results that don’t hold up. This new MIRROR summary from the Science Integrity Alliance explores what’s happening — and how science can respond. ▶ Watch the 8-min video summary https://lnkd.in/eD9zqVrG 📖 Read the written summary https://lnkd.in/esH5HqXy #MIRROR #ScienceIntegrityAlliance #MetaScience #ResearchIntegrity #AIinScience #OpenScience

🐦 The Early-Bird Papers: Who Really Uses Preprints? Preprints let scientists share research before peer review, speeding up discovery and open access. But who is actually using them? A global study analysing 100+ million papers (1991–2023) found: • Only 2.2 million papers were linked to a preprint • Physics and maths dominate (~70% of preprints) • Preprinting is uneven globally — about 10% of papers in Europe vs ~5% in Africa These gaps matter: researchers who preprint may gain visibility, collaboration, and funding opportunities. This new MIRROR summary from the Science Integrity Alliance explores how preprints are spreading across disciplines and regions — and why the benefits of open science aren’t shared equally yet. ▶ Watch the 7-min video summary: https://lnkd.in/e8J3wPU3 📖 Read the full summary https://lnkd.in/ed8bqrAm #MIRROR #ScienceIntegrityAlliance #OpenScience #Preprints #MetaScience #ResearchTransparency

📊 Science improves when data is shared — but not all journals are on board yet. A review of 165 medical journals found that clear data-sharing policies are still far from universal. • Only 36% allowed researchers to deposit data in repositories • Over 60% gave no clear guidance at all • Just ~9.5% of published articles included supplementary data Open science depends not only on open access, but also on sharing raw data and research materials so others can verify and build on findings. ▶ Watch the new MIRROR 8-min video summary from the Science Integrity Alliance: https://lnkd.in/eCRGp_9c 📖 Read the written version: https://lnkd.in/e5bkhdfc #MIRROR #ScienceIntegrityAlliance #OpenScience #DataSharing #ResearchTransparency #MetaScience

🎤 Fame, Love, and the “Glass Wall” What happens to relationships when one partner becomes famous? A study of musicians suggests fame can quietly strain romantic relationships. Artists described how a “celebrity persona” often walks into the room before the real person does, bringing expectations, assumptions, and public attention. Over time, this can affect intimacy: • Mistrust – wondering if people care about you or your status • Guardedness – protecting yourself from gossip and scrutiny • Power imbalance – the famous partner becoming the centre of attention • Emotional distance – closeness becoming harder when you feel watched Some artists also said partners helped them drop the persona and stay grounded. The takeaway? Fame may shine publicly, but in relationships it can act like a glass wall — everyone can see you, yet real closeness becomes harder to reach. ▶ Watch the video summary here: https://lnkd.in/eeF8XBVk 📚 Read the plain-language summary here: https://lnkd.in/eugmrtZb #MusicIndustry #MentalHealth #Relationships #ScienceCommunication #TheCollaborativeLibrary