Subject category:
Finance, Accounting and Control
Published in:
2013
Length: 4 pages
Data source: Field research
Abstract
The scenario presented is that the reader is a member of a student (eg MBA) consulting team working for a real client. The reader has been asked to review a section of the draft report being prepared for the client and comment on it. The extract provided is based closely on sections of a real MBA team report that contained several severe shortcomings and required a complete rework. The accompanying raw dataset is provided. The teaching philosophy is based on recognising 'What does bad look like?' and designing and executing 'What does good look like?'… The intention is to provide an interesting and realistic context to learn or exercise core statistical methodology and tools. Key messages include: the power of graphical presentation, the importance of examining substantive and statistical significance, tailoring presentation and interpretation to the intended audience, and that there aren’t always strong patterns waiting to be found! The depth of reanalysis can be tailored to the level of the class: from a set of basic confidence interval graphs through ANOVA (which can be produced with Excel, but are not really adequate) to the non-parametric equivalent (Kruskal-Wallis, using eg SPSS) and post-hoc analysis. Note: This case has been used as an assessment several times, but has since been used as an in-class exercise under a variety of names with an example ‘solution’ (enclosed here) made available to students; if it is to be used as an ASSESSMENT then the data could be tweaked to alter the outcomes.
Location:
Industry:
About
Abstract
The scenario presented is that the reader is a member of a student (eg MBA) consulting team working for a real client. The reader has been asked to review a section of the draft report being prepared for the client and comment on it. The extract provided is based closely on sections of a real MBA team report that contained several severe shortcomings and required a complete rework. The accompanying raw dataset is provided. The teaching philosophy is based on recognising 'What does bad look like?' and designing and executing 'What does good look like?'… The intention is to provide an interesting and realistic context to learn or exercise core statistical methodology and tools. Key messages include: the power of graphical presentation, the importance of examining substantive and statistical significance, tailoring presentation and interpretation to the intended audience, and that there aren’t always strong patterns waiting to be found! The depth of reanalysis can be tailored to the level of the class: from a set of basic confidence interval graphs through ANOVA (which can be produced with Excel, but are not really adequate) to the non-parametric equivalent (Kruskal-Wallis, using eg SPSS) and post-hoc analysis. Note: This case has been used as an assessment several times, but has since been used as an in-class exercise under a variety of names with an example ‘solution’ (enclosed here) made available to students; if it is to be used as an ASSESSMENT then the data could be tweaked to alter the outcomes.
Settings
Location:
Industry: