[Q] Topic 2 - Data and Graphical summaries

Vocabulary practice flashcards covering key terms and concepts from the DATA1001/1901 Semester 1 lecture notes, including data types, EDA, probability, regression, and hypothesis testing.

What is data?

Data: Information about a subject; often a sample of a population.

What is quantitative data?

Quantitative Data: Numerical info; can be continuous (decimals) or discrete (whole numbers).

What is a continuous variable?

Continuous Variable: A type of quantitative variable that can take any decimal value.

What is a discrete variable?

Discrete Variable: A quantitative type that only has whole numbers.

What is qualitative data?

Qualitative Data: Labels or categories; can't be used for math.

What is nominal data?

Nominal Data: Qualitative data with no order (like gender).

What is ordinal data?

Ordinal Data: Qualitative data with a natural order, unequal gaps.

What is exploratory data analysis (EDA)?

EDA: First look at data; includes Background, Structure, Wrangling, Summaries.

What is tidy data?

Tidy Data: Variables are columns, observations are rows.

What is self-selection bias?

Self-selection Bias: Bias from participants excluding themselves.

What is a randomised controlled trial (RCT)?

RCT: A study design that establishes causation through random assignment.

What is data linkage?

Data Linkage: Combining data from different sources.

What is a comparative boxplot?

Comparative Boxplot: Graph for visualizing qualitative vs. quantitative variables.

What is a scatterplot?

Scatterplot: Visualizes the relationship between two quantitative variables.

What is a density histogram?

Density Histogram: Histogram showing density on the y-axis (total area = 100%).

What is the interquartile range (IQR)?

IQR: The spread of the middle 50% of data; $IQR = Q3 - Q1$.

What is standard deviation (SD)?

SD: The average distance of data points from the mean; ≥ 0.

What are robust statistics?

Robust Statistics: Measures like Median and IQR; not influenced by outliers.

What is chance error?

Chance Error: Random measurement error measured by standard deviation of replicas.

What is bias?

Bias: A consistent error in one direction, often harder to detect.

What is the normal distribution rule?

Normal Distribution Rule: About 68%, 95%, 99.7% of data within ±1, 2, 3 SD.

What is correlation (r)?

Correlation (r): Measure of linear association; range from -1 to +1.

What is the central limit theorem?

Central Limit Theorem: Sums from large samples are approximately normally distributed.

What is the prosecutor's fallacy?

Prosecutor's Fallacy: Mistaking $P(\text{evidence}\text{ | innocent})$ for $P(\text{innocent}\text{ | evidence})$.

What are residuals?

Residuals: Differences between actual data and the regression line; average is 0.

What is R-squared (R²)?

R²: Proportion of variation in dependent variable explained by an independent variable.

What is a p-value?

P-value: Probability of extremes if the null hypothesis is true.

What is Cochran's Rule?

Cochran's Rule: Expected frequency in chi-squared test must be ≥ 5.

What is YAML?

YAML: Configuration format in Quarto; correct indentation is key.