|
Introduction to Graphs |
Introduction
to Graphs
1. Know the advantages and disadvantages of frequency distributions and graphs compared to statistics to describe distributions. Tables and graphs are good for quick, overview of distributions (frequency distributions) and serves as visual comparison of many distributions. They are especially useful to evaluate the shape of a distributions. Statistics are better than graphs in providing specific parametric characteristics of distributions such as central tendency (e.g. Mean) and variability (e.g. Standard deviations). 2. Know the advantages and disadvantages of grouping and situations when grouping may be helpful. The primary use of grouping data is for making graphs or table summaries. Grouping (graphs or tables summaries) of data allows key characteristics of the data to be more easily interpreted or presented (a picture is worth a thousand words principle) than would be true if the raw data especially if large would be presented. Grouping does not result in as much loss of detail as would describing groups with a single statistical parameter such as the mean. Grouping or statistical parameters does results in loss of precision. 3. Know the characteristics of common
charts or graphs for summarizing data.
4. Know the basic principles for proper construction of charts and graphs. The number of class intervals should be between 5 and 20 The class width can be determined by:
The ratio of the vertical axis to that of the horizontal axis should be such that the vertical axis (Y) is ¾ the length of the horizontal axis. The location of the zero value for the vertical axis must be
clear indicated (by a broken axis if necessary). Otherwise it is called
a truncated chart. Figure 2b.1a shows a truncated chart where
the zero point of the y-axis in not clearly shown. Truncated graphs often
are misleading.
5. Know how to compare the shapes of two distributions with different samples sizes on the same axis. To compare the shapes of two distributions with different sample sizes on the same axis it is necessary to use percentage as the vertical axis. Example: A survey of the opinions of 50 men and 80 women on an
issue with Yes, No or None response is tabulated and graphed below for
comparison
6. Know the shape
of the following types of distributions, circumstances when each occur,
and recognize examples of variables that would result in each shape:
Figure 2b.3 Normal Distribution
Figure 2b.4 Positively Skewed
(skewed to the right)
Figure 2b.5 Negatively Skewed (skewed
to the left)
Figure 2b.6 J-shaped curves (Exponential)
Figure 2b.7 Bimodal Distribution
Figure 2b.8 U-shaped distribution
Figure 2b.8 Rectangular
(Uniform) Distribution
|