What Have We Learnt About Attendance Demand in Sports During the Last Decades A MetaRegression Analy

1
What Have We Learnt About Attendance Demand in
Sports During the Last Decades? A Meta-Regression
Analysis
Jaume García (U. Pompeu Fabra) Guillermo Villa
Plácido Rodríguez (Universidad de Oviedo)

• X IASE ANNUAL CONFERENCE
• -----------------------------------------
• MAY 2008

2
Attendance

• Siegfried (1980)
• Kesenne (1987)
• Baade (1990)
• Downward (2000)
• Forrest Simmons (2002)
• Fort (2003)

3
Surveys

• Schofield (1983)
• Cairns (1990)
• Downward Dawson (2000)
• Dobson Goddard (2001)

4
Variables

• Economics variables
• Quality variables
• Loyalty
• Outcome Uncertainty
• Oportunity Cost
• Others

5
Specific Surveys

• García Rodríguez (2003)
• Borland MacDonald (2003)

6
Meta-analysis

• It is a technique of synthesizing research
  results by using various statistical methods to
  retrieve, select, and combine results from
  previous independent but related studies.

7
Applications

• Biology
• Medicine
• Social Science

8
Characteristics

• focuses on empirical research, rather than
  theoretical.
• produces quantitative results, rather than
  qualitative findings.
• examines the same constructs and relationships.
• provides findings that are comparable in
  statistical form (effect size).

9
The effect size (e.g. elasticities)

• is a standardized index that makes comparison
  possible.
• is the dependent variable of the meta-analysis.
• represents the magnitude and direction of the
  relationship of interest.
• is independent of the sample size.

10
Strengths

• It imposes a discipline on the process of summing
  up research findings.
• It represents findings in a more sophisticated
  manner than conventional reviews.
• It allows finding relationships across studies
  that are obscured in other approaches.
• It protects against over-interpreting differences
  across studies.
• It can handle a large number of studies.

11
Weaknesses

• It requires a good deal of effort.
• It assumes that studies are really comparable.
• It is highly influenced by the selection bias.

12
Data

• Papers 102
• Regressions 481
• Years 1973-2007
• Authors 148
• Countries 16
• Sports 9
• Journals 35

13
Methodology Meta-regression

• It examines the relationship between one or more
  study-level characteristics and the sizes of
  effect observed in the studies.
• In our case, price and income elasticities are
  expressed as a function of several
  estimation-level characteristics. Since each
  paper usually provides more than one estimation,
  further statistical modeling will be required
  (hierarchical models and estimation weights).

14
Preliminary data analysis

• Different mean elasticities by countries
• Different results depending on the sample unit
• Different results depending on sports