frequency tables, stem and leaf plots, histograms, box and whisker plots and scatter plots
are most often used to display quantitative continuous data.
Although it is suggested in many introductory statistical texts that a pie chart can be
used to display percentages and count data, this method of data display is seldom seen in
journal articles. It is usually more difficult to compare angles and sectors of a pie chart
than heights or lengths of bars in a bar chart. The use of pie charts is therefore not
recommended.
The relationship between level of measurement and possible data display method for
the A-level data set is shown in Table 3.3.

Table 3.3: Relationship between variable level of

measurement and method of data display

Level of Measurement and Variables

Category

nominal

(Binary)

(^) Category ranked
(Ordinal)
(^) Quantitative continuous
(interval/ratio)
SUB SEX DEGP# ASCORE1 NUMV AGEY
Data display:
Bar chart + + + +
Stem and leaf
plot

• 
  
  
  
  • 
    
    
    
    • 
      
      
      
      • 
        
        
        
        • Box and
          whisker plot
          +* +
          Frequency
          table
        
        
        
        
      
      
      
      
    
    
    
    
  
  
  
  


• 
  
  
  
  • 
    
    
    
    • 
      
      
      
      • 
        
        
        
        • Histogram +
        
        
        
        
      
      
      
      
    
    
    
    
  
  
  
  


• denotes an appropriate method of data display
  +* denotes appropriate for continuous variables but can be used with discrete variables which have
  many distinct values

  in this example, because there is no variation in the value of DEGP, none of the methods of data

  
  

  display would be appropriate
  Bar Charts
  Bar charts can be constructed to summarize information about qualitative variables. A
  frequency bar chart shows the number of observations that fall into each category of the
  qualitative variable. Relative frequency bar charts show the proportion or percentage of
  the total number of observations that fall in each category of the qualitative variable and
  are useful when comparing data distributions which have different numbers of
  observations.
  Example 3.5
  The distribution of number of A-levels that 1 st class honours graduates had obtained on
  entry to university is shown inFigure 3 10as a vertical bar chart and following this is a
  Initial data analysis 51