Using secondary data

1
Using secondary data

• JN602
• Week 06
• Sources and Reading
• Veal Chapter 5
• SLT Chapter 7

2
Lecture outline

• Primary versus secondary data
• Types of secondary data
• Sources of secondary data
• Evaluation of data quality
• Reliability Validity

3
Primary vs Secondary data (Veal, 2005)

• Primary data
• new data specifically collected in current
  research project
• researcher is the primary user.
• Secondary data
• data already exist - collected for some other
  (primary) purpose
• researcher is the secondary user.
• Secondary data analysis Chapter 12

4
Uses/roles of secondary data

• Background/preparation
• Complementary comparison/validation of primary
  data collected
• Whole basis of project re-analysis of data
• Context setting (in report)

5
Use of Secondary Analysis

• Economics
• Accounting
• Political Science
• Geography
• History

6
Advantages of Secondary Data(SLT, 2003, Section
7.4)

• Fewer resource requirements
• Unobtrusive
• Often longitudinal
• Means of comparison with primary data
• Can provide contextual data
• Can result in unforeseen discoveries
• Permanence of data often stored in archives

7
Disadvantages

• Collected for an alternate purpose
• May not match your own
• Difficulty of access
• Expense
• Degree of aggregation
• Data quality

8
Types of Secondary Data

• Data already collected for some other purpose

SLT Figure 7.1
9
Documentary

• Written
• Organisation documents personnel records, safety
  audits
• Reports company, government bodies, committees
• Public documents books, journals, newspapers
• Non-written
• Television and radio
• Video and audio tapes

10
Company/Organisation data INTERNAL

• Financial accounts
• Sales data
• Prices
• Product development
• Advertising expenditure
• Purchase of supplies
• Human resources records
• Customer complaint logs

11
Company/Organisation data EXTERNAL

• Company information is available from a variety
  of sources, eg.
• Biz_at_advantage
• www.whowhere.com
• www.hoovers.com 12,000 companies, USA others
• Australian Stock Exchange (www.asx.com.au)
• AGSM Annual reports
• Kompass, Dun Bradstreet (www.dnb.com), Fortune
  500

12
Possible documentary data?

• Journals and books
• Case study materials
• Committee minutes
• AIRC documentation
• Hansard transcripts
• Mailing list discussions
• Web-site content
• Advertising banners

13
Multiple Source(SLT, 2003)

• Geographically-based
• FT and IMF country reports
• ABS Basic Community Profiles
• Time-series based
• Industry statistics and reports
• Employer associations (ACCI, VECCI, AIG)
• Government publications
• Australian Bureau of Statistics
• Bureau of Labor Statistics (U.S.A.)

14
ABS Subject/Area codes

• Subject Geographical areas
• 1 General 0 Australia
• 2 Census of population and housing 1 New South
  Wales
• 3 Demography 2 Victoria
• 4 Social statistics 3 Queensland
• 5 National accounts, international trade
  finance 4 South Australia
• 6 Labour statistics and prices 5 Western
  Australia
• 7 Agriculture 6 Tasmania
• 8 Secondary industry and distribution 7
  Northern Territory
• 9 Transport, tourism 8 Aust. Capital
  Territory
• 9 External Territories
• Examples of publications and code numbers
• 6356.0 Employer Training Practices, Australia
• 8141.0 Small Medium Enterprises Business Growth
  Performance Survey Australia
• 6201.1 Labour Force, NSW

15
Censuses

• Australia
• 2001 Census data
• http//www.abs.gov.au
• International
• New Zealand http//www.stats.govt.nz/
• U.S.A. http//www.stats-usa.gov
• IPUMSI http//www.hist.umn.edu/rmccaa/IPUMSI/

16
On-going and recurring surveys

• Australian Bureau of Statistics
• http//www.abs.gov.au
• Reserve Bank of Australia
• http//www.rba.gov.au/Statistics/
• Statistics New Zealand
• http//www.stats.govt.nz/
• The World Bank
• http//www.worldbank.org/data/

17
Ad-hoc surveys

• Social Science Data Archive (ANU)
• http//ssda.anu.edu.au
• The Data Archive (Uni. of Essex, UK)
• http//www.data-archive.ac.uk/
• Interuniversity Consortium for Political and
  Social Research
• http//www.icpsr.umich.edu
• Qualitdata
• http//qualidata.essex.ac.uk/

18
Secondary Data Sets

• Longitudinal Survey of Australian Youth
• Australian Centre for Educational Research
• 1989 Present
• Progress from 15 through to post university
• Business Longitudinal Survey
• 5100 Australian small-to-medium enterprises
• 5 year (1994-98 panel study conducted by ABS
• World Values Survey
• 60 countries, every 5 years (approx.)
• Attitudes towards various social, economic and
  political issues

19
Considerations with Secondary Data

• Suitability
• Intended for another purpose
• Content versus external validity
• Reliability
• Measurement bias
• Original research methodology
• Cost
• Time

20
Example Research Problem

• What distinguishes individuals who join trade
  unions from those who do not?
• What distinguishes individuals who leave trade
  unions from those who do not?
• Four possibilities

1996 In 1996 Out
1990 In Stayer Leaver
1990 - Out Joiner Never member
21
Core research issues

• Union instrumentality and ideology
• Work context
• Life context
• Economic situation
• Family history
• Related attitudes

22
Choice of Data

• Australian Workplace Industrial Relations Survey
• 19000 employees in 2000 workplaces
• National Social Science Survey
• 1200 respondents in both 1990 and 1996
• Makes it possible to assess change over time
• Account for differences in Work, Family, Income,
  Location, Attitudes

23
Advantages of secondary data

• Time
• Economy
• Generally inexpensive in comparison to collecting
  one's own data
• High initial cost AWIRS 1995 budget of 3 mil.
• Savings from re-use and re-cycling
• Often free for academic researchers or can be
  acquired for tens or hundreds of dollars
• E.g. post-graduate students are typically
  precluded from collecting national samples
  because of cost.

24
Methodological Advantages

• Large-scale, representative samples
• Longitudinal, broad geographical
• Strong on external validity (the degree of
  confidence with which findings about a sample can
  be generalised to a population)
• Often generated by well resourced teams that have
  access to specialists
• For example, high level sample design expertise
• Does not normally require approval from ethics
  committees

25
Goodness of Measures

• Reliability
• A matter of whether a particular technique,
  applied repeatedly to the same object, yields the
  same result each time.
• How stable and consistent is the measuring
  instrument?
• Validity
• The extent to which an empirical measure
  adequately reflects the real meaning of the
  concept under consideration.
• Are we measuring the right thing?

26
Reliability

• Stability
• refers to the ability of a measure to maintain
  consistency over time, despite uncontrollable
  testing conditions or the state of the
  respondents themselves
• Internal consistency
• indicates how well the items hang together as a
  set and can independently measure the same
  concept, so respondents attach the same overall
  meaning to each of the items

27
Forms of validity

• Face validity
• That quality of an indicator that makes it seem a
  reasonable measure of a variable.
• Criterion related validity
• The degree to which a measure relates to some
  external criterion. For example, the validity of
  the VCE tests is shown in their ability to
  predict the college success of students.

28
Forms of validity (cont.)

• Construct validity
• The degree to which a measure relates to other
  variables as expected within a system of
  theoretical relationships.
• Content validity
• Refers to how much a measure covers the range of
  meanings included within a concept.

29
Evaluating potential secondary data sources

• Assess overall suitability of data to research
  question(s) and objectives
• Measurement validity, coverage
• Evaluate precise suitability of data for analyses
  needed to answer research question(s) and to meet
  objectives
• Validity, reliability, measurement bias
• Judge whether to use data based on an assessment
  of costs and benefits in comparison with
  alternative sources
• If you consider the data are definitely
  unsuitable DO NOT proceed beyond this stage

30
Unions - Content validity

• Includes various measures of
• Job satisfaction
• Atittudes towards trade unions
• Perceptions of class
• Employment situation
• Related attitudes
• No direct measure of union instrumentality
• Need to use proxy work instrumentality

31
Conclusions

• Secondary data can save time, money and effort
• However, it needs to be carefully assessed for
  suitability
• Be sure to check the research methodology used to
  collect the data

32
Sources

• Veal (2005) Chapter 5
• Saunders, Lewis and Thornhill (2003), Chapter 7