Tables and graphs for frequencies and summary statistics

1
Tables and graphs for frequencies and summary
statistics

• Module 1 Sessions 5/6

2
Learning objectives

• Participants should be able to
• Produce multi-way tables
• of counts and tables with the appropriate
  percentages
• Explain the criteria
• that dictate how complex a table needs to be
• Produce tables for continuous variables
• that include the appropriate summary statistics.
• Explain what the margins of a table are
• and know when to include the margins in a table.
• Drill-down to examine the data
• that have given rise to the elements of a table.
• Produce the charts
• that correspond to a given table

3
Contents

• This presentation initially
• Review
• Uses full demonstration Describe data well
• Practicals 1, 2 and 3Counts and percentages
• Practical 1 on demonstration
• Practical 2 uses rice survey
• Practical 3 Survey on Principles of Official
  Statistics
• Presentation continued
• Practical 4 Summarising variates in tables
• Practical 5 Tables and graphs together
• Using the Tanzania case study

4
Objectives from the rice survey
Simple objectives
Not so simple objectives
5
Objectives for different types of variable
Factor, or category, or qualitative variable
Numeric or quantitative variable
Two factors
One numeric and one factor
Two numeric
6
Review and practicals 1 and 2

• Look at the demonstration
• As a class or in pairs
• Do practical 1 at the same time
• Then practical 2
• Same ideas
• Practice using Excel
• Introduces a more complicated objectives (next
  slide)
• To understand different percentages
• Then review these ideas
• Or continue with practical 3 first

7
Review of ideas

• Practical 1
• Reminder of the steps in an analysis
• Practical 2
• One-way and two-way tables
• Which percentage is appropriate?
• Practical 3
• Frequency tables in general
• Tables with and without margins (totals in
  Excel)

8
How to describe data well

• Look for oddities in the data
• and be prepared to adapt the summaries that you
  calculate
• Study the data as tables and graphs
• Use frequencies and percentages
• to summarize categorical variables
• Use averages and measures of variability
• to summarize numeric variables
• Identify any structure in the data
• and use it in producing your summaries

9
Look at the data
The 2 types of variable are summarized in
different ways
10
Simple summaries to meet simple objectives
11
Summary of the yields
Does this summary satisfy any of the objectives?
12
In summary for the simple objectives
13
Checking for oddities
What other suggestions were made for coping with
outliers?
14
Answering more complicated objectives
AND explaining some of the variability
15
One-way and two-way tables
One way table with 2 summary statistics
Two-way table
Margins of the table
The margins of a 2-way table are one-way tables
16
Percentages in tables
Consider carefully which percentage(s) are
appropriate
Here row percentages compare types of country
Column percentages would compare where implemented
17
Sort of 2-way table?
Or one-way table with 9 summary statistics
This margin is meaningless and should be hidden
18
Tables can be 3-way or higher
Excel gives all the totals But is more limited in
giving percentages
19
Practicals 4 and 5

• Practical 4
• Rice survey again
• Tables for numeric variables
• As well as frequencies for factors
• Practical 5
• Apply the ideas
• To the data from Tanzania
• Look at power used for lighting (factor)
• At size of land holdings (numeric variate)
• And at keeping indigenous chickens both!

20
Tables that summarise variates as well
Objective How do the yields relate to the
variety grown?
21
More one-way tables summary statistics
Also Lower s.d. for Variety than Village. Why
might that be important?
22
Summary statistics in 2-way tables
How do the yields relate to fertilizer and
variety?
i.e. extends to objectives from those stated
23
Tanzania agriculture survey

• A huge survey of agriculture in Tanzania
• Involving about 1800 enumerators
• And a main questionnaire of 22 pages
• Given to 3223 households
• In the district provided for analysis
• A technical report
• Describes the process of data entry
• And checking
• Then there is the analysis
• And reporting

24
Data for lighting
25
How large a table do you need?
Are these differences sufficiently large that you
need this 2-way table?
If no, then the one-way presentation is
simpler If yes, then perhaps a one-way
table/graph hides important information
26
One way table/graph
27
Objectives and many questions

• To help the extension service
• Improve their service to households keeping
  chickens
• What proportion of households
• keeps indigenous chickens?
• When they do, how many do they keep?
• Are these values roughly the same,
• or do they differ by district?
• Do they differ by other factors,
• in particular the sex of the household head
• or the type of agriculture household (Q021).

28
Recap

• Now have the tools to process
• Factors (categorical data)
• Using frequencies and percentages
• Variates (quantitative data)
• Using means and medians
• And quartiles, extremes, standard deviations
• And proportions (risks), percentiles (return
  periods)
• You also know to use other measurements
• to reduce the unexplained variation
• And the statistics can be presented
• in tables and graphs

29
Good tables for Excel users a guide

• A guide is provided
• That summarises how to use tables in Excel
• And can be used for reference
• It also introduces some additional topics
• where we need more than Excel
• Other summaries in a table (e.g. medians)
• Weights in tables
• Multiple responses
• That are considered in the next sessions

30
Learning objectives

• Are you now able to?
• Produce multi-way tables
• of counts and tables with the appropriate
  percentages
• Explain the criteria
• that dictate how complex a table needs to be
• Produce tables for continuous variables
• that include the appropriate summary statistics.
• Explain what the margins of a table are
• and know when to include the margins in a table.
• Drill-down to examine the data
• that have given rise to the elements of a table.
• Produce the charts
• that correspond to a given table