Subal C. Brian Layton. Barnett Parker. Barnett R. Bill Cooper. William Cooper. PL Brockett. Patrick Brockett. Estimating elasticities with frontier and other regressions in evaluating two advertising strategies for US Army recruiting. Brocketta,1, W. Goldenc,2, Subal C. Kumbhakard,3, Michael J.
Thus, in contrast to the customary use of a single central tendency method, such as OLS, on different models, we here use multiple methods on a single model to cross check and validate results. In addition to serving as cross checks, the methods can be used to identify classic problems, including biases in the data and shortcomings in one or more of the methodologies employed.
To avoid dealing with problems on identifying motives underlying various patterns in advertising expenditure, data are drawn from a statistically designed experiment where the expenditures were controlled as part of the experiment to resolve the issue of a choice between the two advertising strategies of interest. E-mail addresses: brockett mail. Brockett , cooperw mail.
Cooper , mkllg mail. Golden , kkar binghamton.watch
Kumbhakar , michael. Layton , docbrparker aol. All rights reserved. Brockett et al.
Refine your editions:
Introduction Military recruitment, with advertising as an important component of the effort, is of great importance, especially for the US Army in the current US social and political setting. See the penultimate section of the paper for an illustrative application with a new method for coordinating such activities.
The problem of choice between Joint vs. However, neither Bozell-Eskew nor CBO offered any empirical evidence in support of their respective recommendations. This is an issue of great and abiding importance for the military services6; so, partly in response to the CBO recommendation, DoD had earlier commissioned an empirical study that was undertaken by the Wharton Center for Applied Research WCAR at the University of Pennsylvania.
See the National Research Council Report [3, p. Here, we begin by reviewing several studies that used these data to address the same issue.
From Genius to Madness
Note that each used different methodologies for reasons that will soon become apparent. See the discussion of Table 1, below. Adjustments were therefore undertaken that included a two-stage combination of DEA and OLS regressions, which we refer to as frontierized least squares FLS as one of the two frontier regressions we use in this paper. One issue addressed in the current paper involves customary uses of OLS that provide central tendency estimates of the elasticities in question.
See also the National Academy study on military recruitment edited by Sackett and Mavor . Thus, one might infer that the elasticity is positive and very large whereas, in actuality, it is numerically very small and negative, as determined from the appropriate frontier function. This possibility needs to be investigated by using other methodologies when important issues of public policy are being addressed. See also Charnes et al. These mathematical errors in the data had been discovered by use of an inequality constrained absolute deviation regression in , which took the form of a goal-programming model with only one-sided deviations permitted.
It would appear that mathematical errors were introduced into the data when Evans and Heckman  changed the base period to which the data had previously been adjusted. This introduced mathematical errors in the data, and led to the non-existence of solutions for the then new frontier type regression that Charnes et al. This non-existence made it necessary to search for the problem source that, in turn, led to discovery of this mathematical error. Charnes et al. Hence, again, a different methodology provided different insights, which is also true in the present study, as will be seen in the discussions of Tables 1 and 4, below.
Evaluating military advertising and recruiting : theory and methodology - University Of Pikeville
See Charnes et al. This phenomenon was discovered by using DEA, a methodology that optimizes on each observation, in contrast to statistical methods that optimize across all observations. For the current test, the ADIs developed by Arbitron in were used. They partition the United States into geographical regions. Note that the expenditures budgeted for the different types of advertising constituted part of the experimental design so that there need be no concern about varying underlying motives or rationales for choices of expenditure.
Finally, WCAR decided for reasons that were not made clear to annualize the data and thereby greatly reduced the number of observations.
- Theory and Methodology.
- Orgonomic Functionalism - Vol II.
- Enemy of God (The Arthur Books, Book 2);
- Evaluating Military Advertising and Recruiting: Theory and Methodology — [email protected];
- All Course Descriptions?
- The Nature/ Flower series 08.
Kwinn  and Thomas  for detailed discussions and critiques. One such critical assessment was by Charnes et al. Although this work did not use the WCAR data set as it was not released to the authors at the time , they did use an Army- only subset with the same variables employed by WCAR.
Although the Charnes et al. See, also, the Brockett et al. In a study designed to test the Navy recruitment program, Carroll et al. They also report in favor of local vs.
- Paper Love: Searching for the Girl My Grandfather Left Behind.
- Related Free eBooks.
- Cosmic And Human Metamorphoses: A Course Of Seven Lectures.
- Case Files Psychiatry (2nd Edition) (LANGE Case Files).
- Green Power: Perspectives on Sustainable Electricity Generation.
- Evaluating Military Advertising And Recruiting Theory And Methodology 2004.
- Bibliographic Information.
As we shall later see, the cross-effects of such advertising are of major concern; we thus do not further discuss the Carroll et al. As is common in regression studies of military recruiting9, he assumed a Cobb—Douglas production function that translates into a log-linear form to obtain the elasticities. Using this same model, Dertouzos  also conducted a separate regression analysis for each of the services. As we will later use the RAND formulation and results for direct comparisons with our own approaches, we do not present them here in further detail.
For the moment, we simply summarize their results. While the RAND study further exposed, and subsequently corrected, analytical inadequacies in the WCAR study, there are some points with which we take issue. Many of these issues are described in detail in Kwinn , and, so, are not repeated here. We do comment on one point that speaks to possible differences in our analyses later in this paper. This means that for any combination of inputs, the output, represented as the dependent regressand variable, apart from statistical error, is maximal for every input combination that might be used, so that the function to 9 See Sackett and Mavor .
See Cooper et al. For, as shown in Table 1, taken from Brockett et al. DEA-regression combinations for frontier regressions The uses of DEA outlined in Section 3 did not yield the desired elasticity estimates so we turn to another approach that, as described in Brockett et al. The focus in Brockett et al. This use is questionable, we think, for reasons such as those outlined in , and so it is not included in our model. Borrowing from Brockett et al. See Ray  for a 10 As noted earlier, the Cobb—Douglas function is the one commonly used in military recruitment studies.
Hence, spillover into Army recruits from Marines advertising is likely to occur. In any case, except for Marines, advertising by the other services has a negative impact on Army recruitment in this MLS regression. Moreover, this effect is reinforced by Joint Advertising, and is consistent with the negative effects from the other services by Joint with advertising that is directed to recruitment for all services.
Hence, from these results, we infer that Joint advertising, like Navy and Air Force advertising, has a relatively large negative effect on Army recruiting. This effect from Joint advertising may be due, in part, to the content of such advertising which is generally emphasizes the advantages of a military career without according preference to particular services.
An example of such competition with Army recruiting is exhibited by Navy advertising; it currently emphasizes that its forces do not engage in ground combat which, in the present context of the Iraq war, is likely to have negative effects on Army recruiting. Stochastic frontier analysis SFA In order to further evaluate these results, we now add a third method of estimation in the form of a SFA regression. See the Addendum, below, where the SFA model also includes quotas.
Conversely, when quotas are not being achieved, recruiter efforts are increased. To understand what this all means, we consider Table 4. The bottom portion contains the advertising variables on which our interest centers. As noted earlier, the present study is directed to Joint vs. Joint, and by a margin of nearly 2—1. This is true despite details of the RAND regression, discussed earlier. See, also, Coelli, et al. At the same time, we can also conclude that advertising from the other services, including Joint generally competes with Army recruiting.
This suggests the need for some type of coordination, which may be what CBO and Bozell—Eskew had in mind. We therefore now show how such coordination of advertising effort can be accomplished without requiring a wholly new organization to do so. Coordinating advertising elasticity budgets We begin our investigation into coordinating advertising efforts with the following formulation. Because it is the better known of our two frontier regressions, we will employ SFA in our illustration. The resulting fold difference in these two types of budget increases highlights the effects of negative elasticities in advertising for the other services on Army recruiting.
Budgetary reallocations A possible reason for the CBO and Bozell—Eskew recommendation lies in the behavior that we have just examined. However, other alternatives might also be explored. We suggest, for example, a model or models like the one developed below. We re-focus on the Cobb—Douglas form as it is the one used above, and, as already noted, has been widely used in military recruitment advertising studies. Typically, organizations develop pre- and post-hire objectives and incorporate these objectives into a holistic recruitment strategy.
This typically starts by advertising a vacant position. There are numerous professional associations for human resources professionals. Such associations typically offer benefits such as member directories, publications, discussion groups, awards, local chapters, vendor relations, government lobbying, and job boards. Professional associations also offer a recruitment resource for human resources professionals.
From Wikipedia, the free encyclopedia. Business portal. Applicant tracking system Candidate submittal Dismissal employment Ethnic penalty Employment agency Human resource consulting Human resource management Industrial and organizational psychology Onboarding Personality—job fit theory Personnel selection Purple squirrel Recruitment tool Recruiting metrics Sourcing personnel Trends in pre-employment screening.
Reviewed Papers Form 17th International Conference. Mekon Retrieved Employee referrals are the highest quality hires The Atlantic. Don't hunt the 'purple squirrel ' ". Crain's Detroit Business. Consulting Psychology Journal: Practice and Research. Business NH Magazine. Valdosta State University. Retrieved 7 April July 30, Retrieved 28 October Retrieved 22 July Wall Street Journal. Series 5. International Journal of Management and Decision Making. Hiring Resources. Breaugh The Oxford Handbook of Recruitment.
SHRM Foundation. Blue-collar Green-collar Grey-collar Pink-collar White-collar. Affirmative action Equal pay for women Gender pay gap Glass ceiling. Corporate abuse Accounting scandals Corporate behaviour Corporate crime Control fraud Corporate scandals Discrimination Dress code Employee handbook Employee monitoring Evaluation Labour law Sexual harassment Sleeping while on duty Wage theft Whistleblower Workplace bullying Workplace harassment Workplace incivility. See also templates Aspects of corporations Aspects of jobs Aspects of occupations Aspects of organizations Aspects of workplaces Corporate titles Organized labor.