Login

Publications  •  Project Statistics

Glossary  •  Schools  •  Disciplines
People Search: 
   
Title/Abstract Search: 

Dissertation Information for Gary John Russell

NAME:
- Gary John Russell

DEGREE:
- Ph.D.

DISCIPLINE:
- Marketing

SCHOOL:
- University of Chicago (USA) (1985)

ADVISORS:
- None

COMMITTEE MEMBERS:
- None

MPACT Status: Incomplete - Not_Inspected

Title: THE IMPACT OF DECISION RULES ON ADVERTISING MEASUREMENT

Abstract: In the marketing literature, there is substantial evidence that firms use knowledge of typical sales activity to determine advertising expenditure. Although the details of the budgeting process vary across firms, surveys of large U.S. firms repeatedly find that the majority use some type of constant advertising to sales ratio heuristic. Findings of this sort are reassuring and worrisome to researchers. On the one hand, normative theories of advertising spending emphasize that optimal advertising strategy depends upon the characteristics of sales response. On the other hand, the use of a highly systematic rule can make the impact of advertising difficult or impossible to detect. The key problem is that the use of a reasonable sales-based rule can distort the appearance of the advertising-sales relationship. Consequently, inferences about advertising effectiveness drawn from historical data are fundamentally dependent upon the firm's decision behavior.

The thesis exploits this relationship between inference and decision behavior to make progress in solving two distinct problems in advertising measurement: the difficulty in detecting the impact of advertising in the presence of seasonal demand patterns, and the overestimation of the carryover of advertising known as the data interval bias. The first problem refers to the instability of advertising elasticity estimates whenever advertising expenditures are allowed to follow the exogenous shifts in demand. The second problem refers to the spurious relationship between the apparent carryover of advertising and the level of aggregation of the data. Although these problems appear quite different, both may be viewed as a misinterpretation of the variation in the observed advertising-sales relationship which is due to decision behavior. Through the use of econometric analysis and simulation, it is shown that knowledge of the firm's decision behavior allows the problems to be structured in a natural way and leads to considerable improvement in the quality of measurement. These measurement gains are realized even though there is no feedback between advertising and sales present in the data.

MPACT Scores for Gary John Russell

A = 0
C = 1
A+C = 1
T = 0
G = 0
W = 0
TD = 0
TA = 0
calculated 2008-01-31 06:14:10

Advisors and Advisees Graph