Abstract
Problem structuring results in a list of elementary issues to crack or elementary hypotheses to test. You will address them one by one through various types of analyses. Start with an analysis plan, in which you identify the analyses and sources you need to address each elementary item. You must give priority to the analyses that can radically change the overall solution. Contrary to popular belief, not all analyses are quantitative, or even entirely fact based: many require assumptions and judgments. When you perform analyses, beware the most common mistakes: (1) biased data selection; (2) assumptions that are unrealistic, inconsistent, untested, or—worst of all—unstated; (3) numerical errors; and (4) data interpretation errors, especially with correlations.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Alternatively, such “givens” could be included in the problem statement and wouldn’t need to be repeated in the hypothesis pyramid (or issue tree). In practice, however, hypothesis pyramids often include “given” components that are necessary for the logic to hold, and that have not been expressly formulated in the problem statement.
- 2.
Kahnemann , D., & Lovallo, D. (1993). Timid Choices and Bold Forecasts: A Cognitive Perspective on Risk Taking. Management Science, 39(1), 17–31.
- 3.
Ramanesh, R.V., & Browning, T.R. (2014). A Conceptual Framework for Tackling Unknown Unknowns in Project Management. Journal of Operations Management, 32, 190–204.
- 4.
Walters, D.J., Fernbach, P.M., Fox, C.R., & Sloman, S.A. (2017). Known Unknowns: A Critical Determinant of Confidence and Calibration. Management Science, 63(12), 4298–4307.
- 5.
Landsberger, H.A. (1958). Hawthorne Revisited. Ithaca: Cornell University Press.
- 6.
Harford, T. (2017). Where the Truth Really Lies with Statistics. In Tim Harford, The Undercover Economist (blog). Retrieved from http://timharford.com/2017/06/where-the-truth-really-lies-with-statistics/.
- 7.
Notwithstanding the dividends paid by GE during that period, which, as mentioned above, are part of the return to shareholders.
- 8.
Survivor bias is also frequent in quantitative analysis, for instance, when analyzing a sample of “surviving” companies and neglecting the ones that disappeared over the period of study.
- 9.
Ben-David, I., Graham, J.R., & Harvey, C.R. (2013). Managerial Miscalibration. Quarterly Journal of Economics, 128(4), 1547—1584.
- 10.
Russo, E., & Shoemaker, P. (1992). Managing Overconfidence. Sloan Management Review, 33(2), 7–17.
- 11.
Here’s why: if sales are down 10 percent from, say, $1 million the previous year, then current year sales are $900,000. Increasing this amount by 10 percent (i.e., 2 × 5 percent) results in sales of $990,000 ($10,000 short of the goal). While this may seem a trivial difference, the case interviewer will not be inviting you back for the next round.
- 12.
Gandel, S. (2013, April 17). Damn Excel! How the ‘Most Important Software Application of All Time’ is Ruining the World. Fortune. Retrieved from http://fortune.com/2013/04/17/damn-excel-how-the-most-important-software-application-of-all-time-is-ruining-the-world/.
- 13.
Reinhart, C.M., & Rogoff, K.S. (2009). This Time Is Different: Eight Centuries of Financial Folly. Princeton, NJ: Princeton University Press.
- 14.
Wejnert, B. 2002. Integrating Models of Diffusion of Innovations: A Conceptual Framework. Annual Review of Sociology, 28, 297–326.
- 15.
Calantone, R.J., Cavusgil, S.T., & Zhao, Y. (2002). Learning Orientation, Firm Innovation Capability, and Firm Performance. Industrial Marketing Management, 31(6), 515–524.
- 16.
Spurious Correlations. Retrieved from http://tylervigen.com/view_correlation?id=1703.
- 17.
A statistical association, but not a bivariate, linear correlation. Conventional correlations are linear, while many causal relationships are non-linear. Trying to fit a straight line between two non-linearly related variables will often result in zero correlation. Additionally, a causal relationship may exist between two variables, but detecting it may depend on knowledge of a third variable. This is a confounding effect like the cause of spurious correlations, but works in the opposite way: by explicitly controlling for the influence of the third variable, the partial correlation between the two other variables can be detected.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2018 The Author(s)
About this chapter
Cite this chapter
Garrette, B., Phelps, C., Sibony, O. (2018). Solve the Problem: Eight Degrees of Analysis. In: Cracked it!. Palgrave Macmillan, Cham. https://doi.org/10.1007/978-3-319-89375-4_7
Download citation
DOI: https://doi.org/10.1007/978-3-319-89375-4_7
Published:
Publisher Name: Palgrave Macmillan, Cham
Print ISBN: 978-3-319-89374-7
Online ISBN: 978-3-319-89375-4
eBook Packages: Business and ManagementBusiness and Management (R0)