ATHK1001 Lecture Notes - Lecture 2: Australian Tertiary Admission Rank, Descriptive Statistics, Central Tendency
16 Jun 2018
School
Department
Course
Professor
Lecture and
A Paradox
• People complain that statistics is confusing and irrelevant but happily
discuss the following:
- Batting averages
- Range of rents in different areas
- ATAR scores
• None are perfect, but all provide meaningful information in an accessible
way
• Can’t we rely on someone else to interpret data for us (Wikipedia,
newspaper, government)
- Statistical analysis informs rather than reveals the truth
- There are limits on the data we can gather and experiments we can
conduct
- Easy to lie with statistic but its hard to tell the truth without them
Different measures of central tendency (average)
• Mean: sum of scored divided by 2
• Medium: score of middle data point
• Mode: most common score/category
What is the best?
Depends on what is most informative within a context
Sometimes no measure of average is useful
- Mean number of testicles people have is very close to 1.0
- Medium and mode is 2, but on distribution is meaningful
Category data- you can’t take a mean or medium, but you can say there is a mode
Psych
Summarising Data:
- Use descriptive statistics
- Can be a trade-off between clarity and efficiency
- Common Ways to summarize numerical data:
• Central tendency (average)
• Graph of distribution
• Spread
find more resources at oneclass.com
find more resources at oneclass.com
Document Summary
A paradox: people complain that statistics is confusing and irrelevant but happily discuss the following: Atar scores way: none are perfect, but all provide meaningful information in an accessible, can"t we rely on someone else to interpret data for us (wikipedia, newspaper, government) Statistical analysis informs rather than reveals the truth. There are limits on the data we can gather and experiments we can conduct. Easy to lie with statistic but its hard to tell the truth without them. Different measures of central tendency (average: mean: sum of scored divided by 2, medium: score of middle data point, mode: most common score/category. Depends on what is most informative within a context. Mean number of testicles people have is very close to 1. 0. Medium and mode is 2, but on distribution is meaningful. Category data- you can"t take a mean or medium, but you can say there is a mode. Can be a trade-off between clarity and efficiency.
