Presenting descriptive statistics using graphs and tables Flashcards

1
Q

What is important when trying to work out how to draw graphs?

A
  • What information do you want to present?
  • Diagnostic or summary
  • What data type?
  • Interval or ratio
  • Ordinal or nominal
  • How many different types of data e.g. no. of variables
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What do descriptive statistics do?

A
  • Check for errors
  • Describe & summarise
  • Spread of the data
  • Ensure appropriate analysis
  • Check your anticipated result
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What are univariate graphs? and give examples

A
Graphs are only showing one variable
e.g.
• Pie charts
•Histograms
• Box plots
• Percentiles
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

When should pie charts be used?

A
  • Ordinal or nominal data types only
  • Few categories-no more than 3/4
  • Not to be used for interval/ratio data type
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

When should histograms be used?

A
  • Histograms for ratio/interval variables only
  • Too many possible values for each value to get a column of its own
  • Collapsed into classes that are of equal width e.g. classes of 5% body fat
  • Equal distances
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

When should bar charts be used?

A
  • Bar charts are for categorical variables

* Have gaps between the columns unlike histograms

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

When should a box plot be used?

A
  • Ratio/interval variables or for an ordinal level variable with lots of possible categories
  • The box plot displays statistics associated with the median
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What are the components of a box plot?

A
  • Outlier (possible)
  • Largest value that is not an outlier
  • Upper quartile
  • Median
  • Lower Quartile
  • Smallest value that is not an outlier
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What are other common percentiles?

A
  • Terciles-populaion divided into 3 equal groups
  • Quintiles-populaion divided into 5 equal groups
  • Decile- populaion divided into 10 equal groups
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What can be detected using box plots?

A

Skewness
Positive- quartiles move to the left as well as median
Negative- quartiles move to the right as well as media

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is bivariate?

A

Involves two variables

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What is bivariate presentation dependent on?

A

• Dependent on variable type

  • Ratio/interval and ratio/interval
  • Categorical and categorical (i.e. not ratio/interval)
  • Ratio/interval and categorical
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

How can bivariate relationships be shown?

A

• A bar chart can display values of one categorical variable broken down by another
- If appropriate, can be stacked
• Box plot
• Scatter graph
• Histograms and pie charts can only display one variable at a time

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What can be used to show two categorical (bivariate) data?

A

Stacked bar charts

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What can be used to show ratio/interval and categorical variable (bivariate) data?

A

Box plots e.g. different box plot for each category to compare on same graph

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What can be used to show two ratio/interval (bivariate) data?

A

Scatterplot/gram

17
Q

What do findings show about what ways people like to see data presented?

A
  • 1 Position on a common scale
  • 2 Length
  • 3 Area
  • 4 Position on non‐aligned scales
  • 5 Angle
  • 6 Slope
  • 7 Greyscale
18
Q

How can data be misleading?

A
  • Not drawn to scale
  • Condensed axis
  • Percentages don’t add up