Lecture 2 - Critical Thinking and Data Sleuthing Framework

The focus of the lecture is on developing critical thinking skills when interpreting data from research articles, news articles, or any reporting that presents results.
Framework introduced: Data Sleuthing Framework
- Purpose: Help students question and analyze data presented in various contexts.
- Importance of understanding components before delving deeper into data quantification.

Data Sleuthing - A process that involves critically evaluating data to better understand its implications and context.
Emphasis on the need for a framework to digest information and be skeptical about results presented in public forums.

Discussion of Leighton Smith and his radio discussions on climate change.
- Presenter’s commentary on whether Smith's audience and guests represent public opinion.
- Key point: Possible bias in sample due to the nature of the show and selection of guests.
Example of a door-to-door survey by a police officer regarding drug use.
- Question posed: How does the interviewer's identity affect honesty in responses?
- Conclusion: The persona of the interviewer can create biases in data collection.

A series of questions and critical considerations as you analyze data from any source.

Importance of knowing who funded the research.
- Different stakeholders: government, private companies, universities, etc.
- Example: Census conducted by government to make informed policy decisions.
- Relevance of funding source to potential biases in research outcomes.
Issues of transparency often arise with online articles or news articles that may omit funding sources.

Understanding who collected the data and their qualifications.
- Not all data collectors have the expertise to accurately interpret sensitive information.
- Example of health care data collection requiring specific knowledge for accurate outcomes.

Limitations based on the population from which data is collected.
- Example: Research focusing solely on patients with cardiovascular disease.
- Responses from volunteer participants may differ from non-volunteers, influencing data outcomes.

How data is measured can significantly affect results.
- Example with crime statistics comparing Germany and Ireland; definitions of crimes vary.
- Importance of validity in measuring methods and clarity in questioning techniques.
Discussion on leading questions in surveys and how they influence respondents.
- Example cited about survey design and how the phrasing of questions can bias results.

The environment where data is collected can impact accuracy (e.g., clinical settings).
- Specific examples include proper procedures for taking blood pressure measurements.
- Example: Variability in blood pressure readings based on setting and methodology.

Consideration of confounding variables when analyzing different groups.
- Significance of understanding whether differences in outcomes are due to the variable of interest or other factors.
- Example: Correlation found between ice cream sales and beach drownings explained by summer weather as a confounding variable.

Always measure claims with quantifiable outcomes.
- Example claim: Aspirin reduces heart attacks without numeric reference lacks robustness.
- Improved claim: Reporting a reduction from 17 to 9.4 attacks per thousand gives clearer insight into drug efficacy.
Necessary to critically examine data claims for accuracy based on decimal values and context.

The lecture encourages ongoing development of awareness and skills for critically engaging with data.
Future lectures will expand on each of the seven components with detailed examples and case studies.
The aim is to foster independent analysis and informed consumer of research outcomes.