SOCIOL 3FF3 Lecture Notes - Lecture 3: Bar Chart, Ordinal Data, Level Of Measurement
22 Oct 2018
School
Department
Course
Professor
Monday, September 24, 2018
1
SOCIOL 3FF3 -
Week 3: Graphs, Central Tendency, and Variability
- Nominal and Ordinal variables
- Bar chart
- Pie Chart
- Interval/ ratio variables
- Histogram
- Line Graph
Bar Chart
- For nominal and ordinal variables only
- discrete meausres
- A few rules of bar charts
- Response categories always on x - axis (column)
- Frequencies on y - axis
- Axis titles should be brief
- Axis scales efficient and discrete
- Numbers stated in equal increments
- Chart title brief, include data source and N
Pie Chart
- Can also use pie charts to show nominal and ordinal variable frequencies and
percentages
- Same rules apply that did for bar charts
Monday, September 24, 2018
2
- Except no axes, instead use legend
Presenting Inerval/ Ratio
- Cannot use bar or pie charts for continuous interval/ ratio data
- Instead:
- Histogram
- Line Chart
Histogram
- Same as bar chart, except bars touch
- Represent continuours nature of data
Measures of Central Tendency
- Measures of Central Tendency are another form of univariate descriptive statistics
that help describe the distribution of data
- Summarize information about the most typical, central, middle, or common scores of
a variable
- Often used to generalize or compare values across populations
Three Measures of Central Tendency
- Mean, Median and Mode
- Three different statistics
- Value of each depends on level of measurement (nominal, ordinal, ratio/ interval)
- They report three different kinds of information and will have the same value only in
certain situations
- Which we’ll learn about later with normal distributions
The Mode
- The most common or frequent score
Monday, September 24, 2018
3
- Not an arithmetic measure
- Can be obtained from all variables as all levels of measurement
- Most often used with nominal level variables
- Eye colour, support for political party, religion, favourite coffee brans
- least powerful measure of tendency
How to Find the Mode
- Count the number of times each value occurs
- The value with the highest count is the mode
- Graphically, the mode will be the biggest pie slice, bar, or highest point of a line
BiModality
- what if there are two frequently occurring numbers?
- In this situation we call the distribution “bimodal”
- Meaning, two modes
The Median
- The middle score of any variable
- Cannot be used with nominal data
- i.e., no middle score for “eye colour”
- Can often only be used ordinal data (most common)
- Can be used without restriction on interval/ ratio data
- Can be used without restriction on interval/ ratio data
- Benefit — not influenced by outliers (i.e., extreme scores)
Find the Median for Ordinal Data
Document Summary
Response categories always on x - axis (column) Chart title brief, include data source and n. Can also use pie charts to show nominal and ordinal variable frequencies and percentages. Same rules apply that did for bar charts. Cannot use bar or pie charts for continuous interval/ ratio data. Same as bar chart, except bars touch. Measures of central tendency are another form of univariate descriptive statistics that help describe the distribution of data. Summarize information about the most typical, central, middle, or common scores of a variable. Often used to generalize or compare values across populations. Value of each depends on level of measurement (nominal, ordinal, ratio/ interval) They report three different kinds of information and will have the same value only in certain situations. Which we"ll learn about later with normal distributions. Can be obtained from all variables as all levels of measurement. Most often used with nominal level variables.
