Historical (week5) Flashcards
Origins of Statistics
Statistics derived from ‘State,’ developed during the 18th-century Enlightenment to manage modern nation-states, involving the collection of information about populations, economic activities, and growth.
Census and Statistical Records
Governments initiated population documentation, such as the first British census in 1801, alongside statistics like tax records, GDP, and economic growth estimates for effective governance.
Statistical Collection Process
Statistics aren’t pre-existing; they require meticulous collection and decisions on what to count. Kranzberg’s law parallels technology neutrality, suggesting statistics also possess inherent biases.
Significance Testing in Statistics
Guinness pioneered statistical analysis, utilizing significance testing (e.g., Gosset’s t-test) for cost-effective refinement of brewing techniques, emphasizing decision-making based on limited evidence.
Role of Randomized Data
Fisher expanded Gosset’s ideas, promoting the significance of randomness in experimental design, laying the foundation for randomized control trials vital in scientific research, like drug development.
Statistical Theory Development
Divergent views between Fisher’s pursuit of scientific truth and Neyman & Pearson’s focus on flawed scientific models led to distinct implications in academic research and statistical applications.
Statistical Application at Bletchley Park
Bletchley Park’s code-breaking necessitated swift testing of multiple hypotheses, deviating from traditional statistical methods, prioritizing practical problem-solving over academic conventions.
Statistical Development and AI
The development of AI, initially separate from academia, transitioned from rule-based programming to machine learning, relying on data rather than carefully crafted logical rules.
Evolution to Machine Learning
Machine learning’s rise in AI was driven by increased computing power and data-driven learning, shifting the focus from theoretical insights to pattern recognition and predictive capabilities.
Role of Data Scientist
In the era of machine learning, the data scientist’s role became central, emphasizing understanding and utilizing data for generating insights and making informed decisions.
Data Science Applications
Data science finds diverse applications in various sectors, demonstrating the critical role of understanding and leveraging data in modern businesses and organizations.
Data Influence in Advertising
Data science’s impact on online advertising has reshaped strategies, using algorithms to engage users with content and ads, resembling Fisher’s randomized controlled trial in ad selection.
Data Biases and Decision-Making
The historical context of statistical methodologies and biases in data raise questions about using data for evidence or information and ensuring its impartiality in decision-making.
Data-Driven Decision-Making
The reliance on data shapes decision-making, yet challenges persist regarding the objectivity and biases embedded within data, impacting the generation and testing of ideas.