Chapter 5: Exploring and Explaining Data with Graphs Flashcards
1
Q
grouped frequency distribution
A
- choose the right number of intervals and choose score limits for those intervals to make sense
- general rule: 1 interval for every 10 observations
- number of buckets its key to understanding the shape of the distribution
2
Q
histograms
A
- most useful graph for frequency distribution
- use percentage histograms unless the raw number in each interval is of interest
3
Q
frequency and percentage polygons
A
useful when comparing two distributions within the same figure than histograms
4
Q
ogive curve graphs
A
-graphs the cumulative percent
- useful for finding the percentile of a raw score
5
Q
box and whisker plot (box plot)
A
- use the interquartile range and the max and min scores to visualize the distribution.
- great for comparing two or more distributions
- outliers are shown as dots outside the whiskers
- line in the middle of the box is the median. top of the box is the 75th percentile, bottom of the box is the 25th percentile
6
Q
pie charts
A
- not useful, use a histogram or bar chart
- appropriate when you are concerned about the percentage of the whole and you have 6 or fewer categories, and with nominal data
- great when you care about the percentages from combinations of categories
7
Q
time series graphs
A
- useful for looking at changes in a variable over time
- dotted line can be used to represent extrapolated projections into the future
8
Q
misleading graphs
A
- distorted representation (i.e., linear area fallacy)
- misleading scaling and calibration (ratio scale data should usually be plotted beginning at 0 to give proper visual perspective on the data
- combination graphs: most effective/devious way to lie to give opposite impressions
9
Q
storytelling with data
A
- understand the context
- choose an effective visual
- identify and remove clutter
- direct the audience’s focus
- think like a designer
- tell a story
