Hierarchy of Data Analysis: Impact of Social Media on Academic Performance

The main subject is data analysis hierarchy. A sentence that captures the main takeaway is: "Data analysis hierarchy determines the credibility of evidence in decision-making." (12 tokens)

### The Hierarchy of Evidence in Data Analysis
* 👥 Personal experiences and opinions are the most basic form of evidence, but are easily countered by **even a single opposing experience or expert opinion**.
* 🧐 Expert opinions carry slightly more weight than personal experiences, but are similarly weakened by **opposing expert opinions or experiences**.
* 🔍 Case studies involve in-depth examination of a single case, providing **more reliable evidence** than personal experiences or expert opinions, but with **limited generalizability**.
* 📊 Cross-sectional studies involve analyzing data from a large sample at a single point in time, providing **more generalizable results** than case studies, but still unable to **establish causal relationships**.
* ⏳ Longitudinal studies track the same variables over time, allowing for **better understanding of causal relationships**, but still subject to **potential confounding variables and biases**.
* 🔗 "The biggest unresolved issue here is that there could be a **separate real cause**. In other words, there could be a **third party involved**."
* 💡 "When we make decisions based on data, there are different hierarchies in the evidence we use. Those who utilize data need to consider this information and determine which **level it belongs to**."

### Hierarchy of Evidence in Decision-Making: A Framework for Data-Driven Decisions
* 📊 The text introduces a hierarchy of evidence for decision-making, starting with **anecdotal evidence and expert opinions**, followed by **case studies, cross-sectional studies, longitudinal studies, experimental research**, and **meta-analyses**.
* 🏥 The concept originates from evidence-based medicine, where different types of evidence are ranked based on their **reliability in guiding treatment decisions**.
* 🤔 The text discusses the challenges of applying this framework in everyday decision-making, especially in cases where **immediate action is required** or when **conflicting evidence exists**.
* 🧠 It emphasizes the importance of developing a **culture of inquiry and critical thinking**, where decision-makers evaluate claims based on their **position in the evidence hierarchy**.
* 🔍 The framework is presented as a tool for **assessing the reliability of claims**, rather than a strict requirement for **conducting new research before making decisions**.
* 💼 The text explores the practical implications of this hierarchy in organizational decision-making, highlighting the need for **alignment between evidence-based approaches and real-world constraints**.
* 👥 It touches on the cultural aspects of decision-making, including the tendency to **favor personal experience over data** and the challenges of **challenging existing beliefs**.
* 🏆 The text concludes by suggesting that individuals should develop a habit of **evaluating claims based on this evidence hierarchy**, especially in **organizational settings**.

데이터 분석에도 레벨이 있다

모두가 궁금해하는 데이터 과학의 모든 것을 알려드립니다.
데이터에 미쳐있는 사람들과 함께하는 데이터 이야기, 데이터홀릭! 지금 시작합니다. 

데이터홀릭 후원계좌 : 카카오뱅크 7979-20-70920 (예금주 : 박찬엽)
※ 여러분의 후원금은 콘텐츠 기획 및 운영에 소중히 사용하겠습니다.

| Audio
★ 팟빵 | http://bit.ly/dataholic
★ 애플 팟캐스트 | https://apple.co/2M3yFyc
★ 네이버 오디오클립 | http://bit.ly/2YmoEBj


| Video 
☆ 유튜브 | https://www.youtube.com/dataholic4


| Contact
■ 이메일 | dataholic4@gmail.com
■ 페이스북 페이지 | https://www.fb.com/dataholic4
※ 강연의뢰, 업무제휴, 후원, 기타 방송에 대한 질문이나 건의 등 모든 연락은 이메일로 연락 주세요.


| And More
● Rcoholic!
https://www.youtube.com/channel/UCcYkqQNlfvEGCRpofdmVqSg

● Do it! 쉽게 배우는 R 데이터 분석
http://www.yes24.com/Product/goods/43868089

● 데이터펭귄
https://www.facebook.com/datagentoo



데이터홀릭

Park, Team Leader Kim, and Manager Jo are passionate about data and its analysis.

The Data Leak podcast is supported by contributions from Kakao Bank and JS 1891 on YouTube.

Team Leader Kim shares updates on AI's impact on artists, while Manager Jo discusses job descriptions and the management of code infrastructure.

The Recent Release of The Legend of Zelda and Its Impact

Team members discuss the recent release of The Legend of Zelda game and its influence on other games.

They talk about the game's long tradition, the main character, and its popularity.

The conversation also touches on the difference between console and mobile games, as well as the cultural significance and educational value of playing video games.

The discussion leads to considering the possibility of including gaming topics in future broadcasts.

The Impact of Social Media on Work Productivity

Frequent use of social media during work hours can disrupt concentration and lead to decreased productivity.

Different perspectives on the impact of social media usage at work were based on personal experiences and observations, leading to varying conclusions.

When making decisions based on data, it's important to consider the hierarchy of evidence and the trustworthiness of the information sources.

The abundance of information available from different sources can make it challenging to determine reliable data for decision-making in the workplace.

The Hierarchical Order of Evidence in Data Analysis

Personal experiences are often the basis of arguments, but they provide the lowest level of evidence in data analysis, as they can be easily countered by a single contradictory experience.

Expert opinions, while more persuasive, also have a low level of evidence and can be undermined by opposing expert views.

The use of unmeasured claims and subjective standards in everyday conversations further weakens the credibility of evidence, making it challenging to determine the trustworthiness of the data.

Case studies involve deeply exploring a specific case to examine the context and relationships around it, providing a thorough understanding of a phenomenon.

These studies often use interviews, diaries, and subjective interpretation to analyze the data, but they can be influenced by researcher bias.

Case studies face challenges in proving causal relationships, controlling for other variables, and generalizing findings to a broader context.

Using a cross-sectional study with a larger sample size can improve the generalizability of the findings, but it still struggles to determine causal relationships and the order of cause and effect.

The Limitations of Regression Analysis in Determining Causal Relationships

Regression analysis does not indicate causal relationships, only linear associations between variables.

Longitudinal studies can help establish the temporal order and potential causal effects between variables.

The influence of third variables and the lack of control for other factors in causal relationships is a significant limitation in research studies.

The Impact of Instagram Usage on Grades: A Call for Controlled Experiments

Increased Instagram usage may lead to a decline in grades, but it's crucial to conduct controlled experiments to establish a causal relationship.

A randomized controlled trial with a large and diverse sample is needed to directly apply interventions and ensure similarity among the groups.

The generalizability of the results should be confirmed through replication in different settings and with different age groups to establish a more comprehensive and reliable understanding of the impact of Instagram usage on academic performance.

Meta-analysis involves combining results from multiple studies to make a comprehensive judgment, providing the most conclusive evidence.

It allows for the determination of the level of trustworthiness of a hypothesis based on the support from a large number of studies.

Meta-analysis also helps in calculating the effect size of variables and identifying the strength of their influence in different situations.

The Importance of Establishing a Hierarchy of Evidence

AI interview scores do not reliably predict job performance, highlighting the need for a culture that values and evaluates evidence.

Decision-makers in organizations should prioritize evidence-based claims and establish a hierarchy for evaluating the strength of evidence.

This approach, commonly used in medicine as evidence-based practice, is crucial for making well-informed decisions and avoiding the reversal of evidence hierarchy.

The Use of Hierarchical Treatment Methods in Medical Decision-Making

Conferences and researchers are now adopting hierarchical treatment methods for causal inference in medical decision-making, especially with the increasing online data generation.

The practicality of using traditional methods like regression analysis or machine learning models in this context is being questioned.

The variation in adopting these research methods exists among different organizations and their cultures, leading to a dilemma in prioritizing decisions based on the hierarchy of rationale.

The Importance of Data-Driven Decision Making in Medicine and Business

The concept of data-driven decision-making has been a longstanding discussion in the medical community, emphasizing the need to consider the hierarchy of evidence when making crucial decisions.

This is particularly crucial in the medical field, where the impact of decisions on human lives is significant.

In business, there is a constant struggle between following proper data analysis and making decisions based on personal advantage, highlighting the challenge of balancing the two approaches.

The Importance of Data-Driven Decision Making

Data-driven decision making is crucial for problem-solving and effective leadership in both professional and personal life.

It is essential to be open to the possibility of being wrong and to create a culture that allows for admitting mistakes without blame.

Making a hypothesis is necessary before data analysis, and it should be separate from personal claims and ego involvement.

Assessing the trustworthiness of data and developing a habit of reflection are important for those making decisions in companies.