Table manners

1
Table manners
GAP Toolkit 5
Training in basic drug abuse data
management and analysis
Training session 10
2
Objectives

• Define the common terminology used to evaluate
  survey data
• Establish the information that should be reported
  with the data, whether the data take the form of
  a table, a graph or numerical summaries
• Introduce the Output Viewer in SPSS
• Describe formatting charts and tables

3
Conflicting aims

• A table has two purposes
• To put across a point
• To allow the reader to verify our claims with
  reference to the data
• These aims are not always the same and getting
  the balance right is as much an art as a
  science.
• (C. Marsh, Exploring Data An Introduction to
  Data Analysis for Social Scientists (Cambridge,
  Polity Press, 1988)).

4
Survey terminology

• Reliability
• Validity
• Internal
• External

5
Reliability

• A tool of measurement displays reliability if it
  records the same result in repeated applications
  under identical conditions
• For example
• I weigh myself on my bathroom scales on Monday
  and weigh 100 kg
• I weigh myself on my bathroom scales on Tuesday
  and weigh 72 kg
• This would be nice, but is unlikely the scales
  are probably unreliable

6
Validity

• Validity takes various meanings in survey design
• Internal validity is the ability of the measure
  (the question or observation) to adequately
  represent the underlying concept of interest

7
External validity

• External validity refers to the ability to
  abstract generalizations from the results
• Are the data from a sample?
• If yes, is that sample a reasonable
  representation of the population?
• If so, can the results be generalized to the
  population?

8
Table design

• Clarity
• Structure
• Message
• Completeness

9
Titles

• Who/What?
• When?
• Where?

10
Source

• It is essential that the source of the data is
  presented regardless of what form the statistics
  take
• The source of the data helps establish the
  authority of the data

11
Variables

• Definitions
• Units

12
Percentages

• If a variable is measured in percentages, be
  certain it is clear what it is a percentage of
• Rows
• Columns
• Totals
• Ensure that the original numbers can be
  calculated by reporting the sample size (the size
  of the denominator in the calculation of the
  percentage)

13
Method

• Sample statistics are frequently used to make
  general statements about the population
• If the statistics are drawn from a sample, state
  the method of sampling
• The method determines the external validity of
  the results

14
Missing data

• Always report the number of missing observations
  in your data set
• A large amount of missing data brings into
  question the external validity of a survey
• Those who agreed to answer and those who refused
  may well have different characteristics

15
Precision

• Avoid reporting data to unjustifiable levels of
  precision
• Round results to the level of precision of the
  original data

16
Output Viewer

• The third main window in SPSS
• Used to edit and format the results of any
  statistical analysis
• Generates output files with the suffix .spo

17
Exercise Session 8 Output

• Retrieve the output file Session 8 freq.spo
  saved in session 8

18
Output
Table of contents
19
Left-hand panel

• Use the left-hand panel to manage the output
• Clicking on a heading selects that part of the
  output
• Clicking on the sign to the left of a heading
  hides or collapses the output
• Clicking on the to the left of a heading
  displays or expands the output

20
Right-hand panel

• Select a single piece of output by pointing and
  clicking once
• Select multiple pieces of output by holding down
  the Ctrl key while clicking more than once
• Format output by pointing and double-clicking

21
Copying tables to Word

• Select the output in the left-hand panel or in
  the Viewer
• Edit/Copy
• Move to Word
• Edit/Paste Special
• Formatted Text
• Unformatted Text
• Picture

22
(No Transcript)
23
(No Transcript)
24
Copying charts to Word

• Select the output in the left-hand panel or in
  the Viewer
• Edit/Copy
• Move to Word
• Edit/Paste

25
Exercise copying to Word

• Copy the frequency table for Education to a Word
  document as
• Formatted text
• Unformatted text
• Picture
• Copy the bar chart of Education to the same Word
  document as a picture

26
Formatting tables

• Double click to enter formatting mode
• A new border appears around the output with short
  diagonal lines at a 45-degree angle
• The menu bar now contains formatting options

27
Editing the table

• Edit any component of the table by
  double-clicking that particular part of the table
• A box will appear in which the table can be edited

28
(No Transcript)
29
(No Transcript)
30
Race of clients treatment centres in the Cape
region of South Africa January-June 2001
Frequency Percent Valid Percent Cumulative Percent
Valid Coloured 722 46.0 52.8 52.8
White 520 33.1 38.0 90.8
African 109 6.9 8.0 98.8
Asian 17 1.1 1.2 100.0
Total 1368 87.1 100.0
Missing System 203 12.9
Total 1571 100.0
31
Insert

• Insert/Title
• Insert/Caption
• Insert/Footnote

32
(No Transcript)
33
Caption text box
34
Type required caption here
35
Race of clients treatment centres in the Cape
region of South Africa January-June 2001
Frequency Percent Valid Percent Cumulative Percent
Valid Coloured 722 46.0 52.8 52.8
White 520 33.1 38.0 90.8
African 109 6.9 8.0 98.8
Asian 17 1.1 1.2 100.0
Total 1368 87.1 100.0
Missing System 203 12.9
Total 1571 100.0
Source The data were collected from the
treatment centres as part of the South African
Drug Information System.
36
(No Transcript)
37
(No Transcript)
38
Race of clients treatment centres in the Cape
region of South Africa January-June 2001
Frequency Percent Valid Percent Cumulative Percent
Valid Coloured 722 46.0 52.8 52.8
White 520 33.1 38.0 90.8
African 109 6.9 8.0 98.8
Asian 17 1.1 1.2 100.0
Total 1368 87.1 100.0
Missing System 203 12.9
Total 1571 100.0
a
Source The data were collected from the
treatment centres as part of the South African
Drug Information System.
a Note that the small number of Asian clients
raises issues of the generalizability of the
results pertaining to the Asian community.
39
Format

• Format/Table Properties
• Format/TableLooks

40
TableLooks
41
Race of clients treatment centres in the Cape
region of South Africa January-June 2001
Frequency Percent Valid Percent Cumulative Percent
Valid Coloured 722 46.0 52.8 52.8
White 520 33.1 38.0 90.8
African 109 6.9 8.0 98.8
Asiana 17 1.1 1.2 100.0
Total 1368 87.1 100.0
Missing System 203 12.9
Total 1571 100.0
Source The data were collected from the
treatment centres as part of the South African
Drug Information System.
a Note that the small number of Asian clients
raises issues of the generalizability of the
results pertaining to the Asian community.
42
Formatting charts

• Double-click a chart in the Output Viewer to open
  the chart in the Chart Editor
• To edit or format the chart
• Point at the part to be edited and
  double-clickOR
• Use the menu bar options

43
(No Transcript)
44
(No Transcript)
45
(No Transcript)
46
(No Transcript)
47
(No Transcript)
48
(No Transcript)
49
(No Transcript)
50
(No Transcript)
51
(No Transcript)
52
(No Transcript)
53
(No Transcript)
54
(No Transcript)
55
(No Transcript)
56
(No Transcript)
57
(No Transcript)
58
(No Transcript)
59
(No Transcript)
60
(No Transcript)
61
(No Transcript)
62
(No Transcript)
63
Exercise

• Using the file Session 8 freq.spo, edit the
  frequency table and the bar chart for the
  variable Education to reflect the principles
  discussed here
• Copy the edited frequency table and bar chart to
  a Word document

64
Summary

• Principles
• Subject of the data (who/what?)
• Time period (when?)
• Area covered (where?)
• Variables
• Definitions
• Units of measurement
• Method

• Saving Output
• To file
• To a word-processor
• Tables
• Charts